Dissertations / Theses on the topic 'Radiomic'

Radiomics, a non-invasive and quantitative mining medical imaging information method, could extract molecular biological features and enormous feature combinations to customise individualised treatment and solve the problem of heterogeneity, satisfying the standards of precision medicine. However, it faces many challenges in the feature selection process, including redundant features, irrelevant features and the overfitting risk. More important, people know little about radiomics biological background and its connection to radiology, so it is difficult to apply radiology directly to medicine as it lacks interpretability. The core of this thesis is radiomic biology analysis that connects radiomic imaging information with molecular biology information to achieve a medical “gold standard” for cancer management. Developing methods to succeed in the feature selection process of data of varying dimensions is the main goal of this paper. Our major contributions in this thesis can be summarised as below: 1. We firstly proposed an unsupervised learning framework to guide supervised learning in the reduction of feature dimensions from large cohorts Non-Small Cell Lung Cancer data (NSCLC) on both clinical data and radiomic data for survival prediction. 2. An interpretable machine learning approach measures the contribution of features for each case and the connection of radiomics to its underlying biological features to make clinical decisions in leukemia and breast cancer cases. The weight of the feature can be estimated by measuring the distance of the approximate perturbation centre. 3. Based on the framework of feature selection that we proposed, to ensure the fairness and stability of the data split when processing classification results, cross-validation is embedded in the training process. We further propose a traversal selection method, optimising the computational complexity of the selection process to obtain the most robust feature set.

Radiomics treats images as quantitative data and promises to improve cancer prediction in radiology and therapy response assessment in radiation oncology. However, there are a number of fundamental problems that need to be solved in order to potentially apply radiomic features in clinic. The first basic step in computed tomography (CT) radiomic analysis is the acquisition of images using selectable image acquisition and reconstruction parameters. Radiomic features have shown large variability due to variation of these parameters. Therefore, it is important to develop methods to address these variability issues in radiomic features due to each CT parameter. To this end, texture phantoms provide a stable geometry and Hounsfield Units (HU) to characterize the radiomic features with respect to image acquisition and reconstruction parameters. In this project, normalization methods were developed to address the variability issues in CT Radiomics using texture phantoms. In the first part of this project, variability in radiomic features due to voxel size variation was addressed. A voxel size resampling method is presented as a preprocessing step for imaging data acquired with variable voxel sizes. After resampling, variability due to variable voxel size in 42 radiomic features was reduced significantly. Voxel size normalization is presented to address the intrinsic dependence of some key radiomic features. After normalization, 10 features became robust as a function of voxel size. Some of these features were identified as predictive biomarkers in diagnostic imaging or useful in response assessment in radiation therapy. However, these key features were found to be intrinsically dependent on voxel size (which also implies dependence on lesion volume). The normalization factors are also developed to address the intrinsic dependence of texture features on the number of gray levels. After normalization, the variability due to gray levels in 17 texture features was reduced significantly. In the second part of the project, voxel size and gray level (GL) normalizations developed based on phantom studies, were tested on the actual lung cancer tumors. Eighteen patients with non-small cell lung cancer of varying tumor volumes were studied and compared with phantom scans acquired on 8 different CT scanners. Eight out of 10 features showed high (Rs > 0.9) and low (Rs < 0.5) Spearman rank correlations with voxel size before and after normalizations, respectively. Likewise, texture features were unstable (ICC < 0.6) and highly stable (ICC > 0.9) before and after gray level normalizations, respectively. This work showed that voxel size and GL normalizations derived from texture phantom also apply to lung cancer tumors. This work highlights the importance and utility of investigating the robustness of CT radiomic features using CT texture phantoms. Another contribution of this work is to develop correction factors to address the variability issues in radiomic features due to reconstruction kernels. Reconstruction kernels and tube current contribute to noise texture in CT. Most of texture features were sensitive to correlated noise texture due to reconstruction kernels. In this work, noise power spectra (NPS) was measured on 5 CT scanners using standard ACR phantom to quantify the correlated noise texture. The variability in texture features due to different kernels was reduced by applying the NPS peak frequency and the region of interest (ROI) maximum intensity as correction factors. Most texture features were radiation dose independent but were strongly kernel dependent, which is demonstrated by a significant shift in NPS peak frequency among kernels. Percent improvements in robustness of 19 features were in the range of 30% to 78% after corrections. In conclusion, most texture features are sensitive to imaging parameters such as reconstruction kernels, reconstruction Field of View (FOV), and slice thickness. All reconstruction parameters contribute to inherent noise in CT images. The problem can be partly solved by quantifying noise texture in CT radiomics using a texture phantom and an ACR phantom. Texture phantoms should be a pre-requisite to patient studies as they provide stable geometry and HU distribution to characterize the radiomic features and provide ground truths for multi-institutional validation studies.

Radiomics refers to the analysis of quantitative features extracted from medical images including Positron Emission Tomography (PET), Computerized Tomography (CT), Magnetic Resonance Imaging (MRI), and other medical imaging techniques. Radiomic features can be used to build models providing diagnostic, prognostic, and predictive information. The aim of this work was to build a machine learning model able to predict survival probability in 85 cervical cancer patients, using the radiomics features extracted from CT and PET medical images. A thorough feature selection process was conducted employing different techniques to select the best predictors among the original features in the dataset. In particular, the Genetic Algorithm revealed to be the best of the feature selection methods employed, with promising applications in the field of radiomics. Two different survival models have been developed, a Cox proportional hazard model and a Random Survival Forest. A Decision Tree Classifier was also implemented as a further model to evaluate. All the models were trained on 80% of the available data and tested on the remaining 20%. The Concordance Index (CI) was used as the evaluation metric for the two survival models, while the area under the Receiver Operating Characteristic curve (ROC AUC) was used as the evaluation metric for the classifier. The Cox model trained using 9 selected CT features was superior to all the other models tested. It achieved a Concordance Index score of 0.71 on the test set, showing promising predictive capabilities on external data. Finally, the recurrence outcome was used as an additional feature, producing a general improvement of all the models.

Introduction There are several radiological techniques, and, in this study, we focused the Single Photon Emission Computed Tomography (SPECT) imaging. It is possible to reconstruct the unknown tracer distribution inside the body by applying tomographic reconstruction algorithms such as Filtered Back Projection (FBP) and Ordered Subset Expectation Maximisation (OSEM) to the acquired data. Nowadays, thanks to technological innovations, a new branch of research has rapidly evolved: the radiomics. In practice, radiomics tries to assess whether the “textural features” of images in regions related to specific diseases could provide added value in a diagnostic process, in the evaluation of prognosis or could guide therapeutic choices. A general concern that has to be accounted for when performing a clinical study is whether changes in acquisition and reconstruction parameters could affect the value of the features. To the best of our knowledge, in literature, there is no unique method for identifying robust features; here, a generalised method to study the effects of the variation of reconstruction parameters on radiomic features is proposed and applied to asses stability and reliability in SPECT imaging. Materials and methods Only simulation studies could asses the link between features extracted from reconstructed images and their original values. From a preliminary statistical analysis, it emerged that at least 66 phantoms (representing different original textures) were needed to achieve a statistical power higher than 90%. These synthetic phantoms derived from abdominal CT scans and “Visible Human Project” image sets. Then, using a proper model, we simulated SPECT acquisitions of each phantom and reconstructed the corresponding images changing parameters in FBP and OSEM tomographic algorithms. Features extraction was conducted with PyRadiomics, an open-source software. Six feature classes were considered, based on Intensity, Grey-Level Co-occurrence Matrix (GLCM), Grey Level Dependence Matrix (GLDM), Grey Level Run Length Matrix (GLRLM), Grey Level Size Zone Matrix (GLSZM) and Neighbourhood Grey-Tone Difference Matrix (NGTDM). Ultimately, 93 different radiomics features for each phantom were calculated. In this way, data-set has a series of repeated measurements and the method of Generalised Estimating Equations (GEE) is suitable for analysing databases with a similar structure. In this study, two different GEE models were developed: one to analyse if the radiomic features calculated in the reconstructed images (Vr) reproduce the same feature in the original VOI (Vo); another to study if they are stable or not with reconstruction parameters variations. Results 32 different reconstructions for each available phantom were obtained, for a total of 2112 images stacks. The results of the two GEE models, features could be classified according to four possible groups: a) feature with a correlation between Vo and Vr, without reconstruction parameters variation effect; b) features without a correlation between Vo and Vr and without a significant impact of the reconstruction parameters variation; c) features with a statistically significant correlation between Vo and Vr and with the effect of the reconstruction parameters variation on the Vr value; d) reconstruction parameters variation affects Vr. Moreover, there is not a correlation between the values obtained from the reconstructed images and Vo. Discussion In literature, as far as we know, there are no trustworthy works of the reproducibility or repeatability applied to SPECT imaging. Here, with software simulations, we tried to answer the following two questions: 1) Are the features extracted from the reconstructed images (Vr) correlated to those of the original images (Vo)? 2) Are the features extracted from the reconstructed images robust when the reconstruction parameters vary? To answer these questions, two GEE models were developed. Most features showed a correlation between Vo and Vr, but with a relevant impact of reconstruction parameters variation. For clinical studies, in our opinion, features like a) would be the optimal choice. However, also features like c) could be used, but researchers have to handle with care these features for which the reconstruction parameters variations affect Vr. Using the remaining features is not recommended as the lack of correlation between Vo and Vr makes random any link with clinical end-points, so it could be difficult to reproduce any result on cohorts of patients other than the one used to develop the radiomic model. Conclusions From this study, it emerges how reconstruction parameters could affect radiomic features in SPECT imaging. In our opinion, researchers should take into account this dependency in both retrospective and prospective radiomic studies. Ultimately, the method described in this work, although complicated, represents a logical approach to carry out propaedeutic evaluations about the selection of imaging parameters or radiomic features to be used for clinical studies.

Radiomics is the process of converting medical images into minable high-dimensional data to support clinical decision. A dedicated software is exploited to extract many synthetic biomarkers, called radiomic features, which can be correlated with specific clinical outcomes and may uncover disease characteristics that failed to be appreciated by visual inspection of the images. In pelvic and breast cancer, MRI-based radiomics has been investigated for its potential to describe the tumour heterogeneity, thus improving diagnosis, prognosis and early therapy assessment, and it has shown promising results. However, the process lacks standardisation and harmonisation of the methods employed, from the image processing to the building of predictive models. Parts of these challenges can be investigated with phantom studies, which offer the possibility of repeated acquisition in controlled experimental setup. In this thesis, a pelvic phantom dedicated to MRI-based radiomics has been designed, developed and validated for the first time. A research has been conducted to identify the materials and geometry of the phantom in such a way that it could mimic the MR signal and texture of a tumour and its surrounding tissues, as seen in a set of representative patients. The phantom has been used for a multicentric evaluation of the repeatability and reproducibility of the radiomic features extracted from images acquired on three MR scanners of two vendors and two magnetic field strengths. This study allowed to identify a set of robust radiomic features to support studies on clinical datasets of patients with pelvic cancer. In addition, three radiomic software products for feature extraction have been tested to assess the impact of the choice of a specific tool on the value of the features. A prototype of a radiomic breast phantom has been assembled in the last part of the thesis work, including a 3D-printed insert to mimic a real tumour.

Radiomics is the process of extracting quantitative features from tomographic images (computed tomography [CT], magnetic resonance [MR], or positron emission tomography [PET] images). Thousands of features can be extracted via quantitative image analyses based on intensity, shape, size or volume, and texture. These radiomic features can then be used in combination with demographic, disease, and treatment indicators to increase precision in diagnosis, assessment of prognosis, and prediction of therapy response. However, for models to be effective and the analysis to be statistically sound, it is necessary to reduce the dimensionality of the data through feature selection or feature extraction. Supervised dimensionality reduction methods identify the most relevant features given a label or outcome such as overall survival (OS) or relapse-free survival (RFS) after treatment. For survival data, outcomes are represented using two variables: time-to-event and a censor flag. Patients that have not yet experienced an event are censored and their time-to-event is their follow up time. This research evaluates the effect of transforming a right-censored outcome into binary, continuous, and censored aware representations for dimensionality reduction of radiomic features to predict overall survival (OS) and relapse-free survival (RFS) of oropharyngeal cancer patients. Both feature selection and feature extraction are considered in this work. For feature selection, eight different methods were applied using a binary outcome indicating event occurrence prior to median follow-up time, a continuous outcome using the Martingale residuals from a proportional hazards model, and the raw right-censored time-to-event outcome. For feature extraction, a single covariate was extracted after clustering the patients according to radiomics data. Three different clustering techniques were applied using the same continuous outcome and raw right-censored outcome. The radiomic signatures are then combined with clinical variables for risk prediction. Three metrics for accuracy and calibration were used to evaluate the performance of five predictive models and an ensemble of the models. Analyses were performed across 529 patients and over 3800 radiomic features. The data was preprocessed to remove redundant and low variance features prior to either selection or clustering. The results show that including a radiomic signature or radiomic cluster label predicts better than using only clinical data. Randomly generating signatures or generating signatures without considering an outcome results in poor calibration scores. Random forest feature selectors with the continuous and right-censored outcomes give the best predictive scores for OS and RFS in terms of feature selection while hierarchical clustering for feature extraction gives similarly predictive scores with compact representation of the radiomic feature space.

Precision medicine holds the potential to revolutionize healthcare by providing every patient with personalized treatments and decisions tailored to his or her individual needs. This might be enabled by the large influx of potentially diagnostic information from new sources such as genetics and modern imaging techniques, provided the relevant information can be extracted. One such framework that has started to demonstrate promise in radiology, especially in the assessment of cancer, is radiomics; the practice of characterizing images by extracting a substantial amount of quantitative mathematical descriptors. This success has largely been enabled by artificial intelligence (AI) and machine learning developments that are capable of handling the big data arrays. Using radiomics, researchers have been able to build prediction models capable of assisting and informing doctors in important decisions such as risk assessment or the choice of treatment. But even though radiomics has shown promise in preliminary studies, there is still a long way to go before radiomics and related AI applications can become routine tools in clinics. The road from patient admission to release is long, and all its intricate steps need to be studied in detail to establish the AI models' benefits and safety. Deep learning is an incredibly powerful AI technique that has revolutionized many areas of science and industry such as recommender systems and protein folding. The technique has demonstrated particular capabilities in image analysis, such as the ability to drive cars autonomously and generate realistic-looking images from scratch. However, the recent advances in deep learning have largely been segregated from the radiomics domain, even though they can synergize with radiomics by performing complementary tasks such as image segmentation and denoising. There is considerable potential for DL and radiomics to cooperatively reinforce each other that so far has been majorly unexplored. This thesis investigates the application of radiomics and deep learning in the context of prostate cancer. It focuses on the clinical perspective of where machine learning implementations are most likely to have a beneficial real-world impact. A key contribution is the deployment aspect: the models are not simply proofs of concept but are conceived and applied in a practical scenario, from patient admission to treatment decision. The specific areas studied include automatic organ segmentation in medical images, automatic quality assurance of segmentations, image processing, and radiomic feature analysis. Finally, a comprehensive study is performed on predicting essential pathological variables with AI, which so far has not been studied previously. Taken together, the methods outlined in this thesis constitute a concrete pathway of how AI can be used to bolster the steps along the patient's clinical trajectory. Successful applications of these methods hold the potential to reduce the workload of clinicians and improve patient outcomes.

Arterial spin labelling (ASL) radiomics analysis to predict IDH mutation and MGMT methylation status in gliomas Fabio M. Doniselli1,2, Riccardo Pascuzzo1, Eleonora Bruno3, Domenico Aquino1, Mattia Verri, Alberto Redolfi, Valeria Cuccarini1, Marco Moscatelli1,2, Maria Grazia Bruzzone1, Luca Maria Sconfienza2,4 Abstract Objectives: To evaluate the strength and ability of radiomics features extracted from multiple tumor subregions on MR brain images to predict MGMT promoter (MGMT) methylation status and isocitrate dehydrogenase (IDH) mutation in glioma patients through a multiparametric MRI-based radiomics model, using arterial-spin labelling (ASL) perfusion imaging. Methods: Retrospective single-institution study in a cohort of 52 glioma patients. Radiomics-based models with a minimal set of relevant features and clinical parameters were built for MGMT methylation and IDH-mutation prediction from a training cohort (31 patients) and tested on an validation cohort (13 patients). Results: Feature selection methods (Boruta, RFE and LR-EL) identified age and 3 radiomics features for MGMT prediction and 3 features for IDH prediction. For IDH prediction, SVM classifier achieved average 96.8% accuracy and 0.929 AUC during the training phase, and 84.6% accuracy and 0.60 AUC on the test set. For MGMT methylation prediction, SVM classifier achieved average 67.7% accuracy and 0.765 AUC during the training phase, and 38.5% accuracy and 0.429 AUC on the test set. Conclusions: The classification model based on both demographic (age) and radiomic ASL perfusion characteristics had the best performance in predicting the IDH mutational status of gliomas. This result suggests that the proposed method has promising efficacy in predicting IDH mutational status. We have not obtained a sufficient result trying to correlate radiomics with the MGMT mutational pattern.
Assessment of quality and classification performances of MRI-based radiomics studies on MGMT methylation in gliomas: a systematic review Fabio M. Doniselli1,2*, Riccardo Pascuzzo1*, Massimiliano Agrò3, Domenico Aquino1, Federica Mazzi1, Francesco Padelli1, Marco Moscatelli1, Maria Grazia Bruzzone1, Luca M. Sconfienza2,4 Objectives To evaluate the quality of MRI-based radiomics studies predicting the O6-methylguanine-DNA methyltransferase (MGMT) methylation status in gliomas, using radiomics quality score (RQS) and Image Biomarkers Standardization Initiative (IBSI) guidelines, and examine their classification performance. Methods PubMed Medline and EMBASE were searched to identify MRI-based radiomics studies on MGMT methylation in gliomas until January 31, 2022. Included studies were scored according to RQS (16 components) and IBSI (six items) scales by two raters. Results We included 20 out of 62 identified studies. The median RQS total score was 32% of the maximum (11.5 out of 36), ranging between 8% and 44%. Eleven studies performed external validation; only three studies performed decision curve analysis to report potential clinical utility. All studies reported area under the curve (AUC) or accuracy, and 14 computed these statistics using resampling methods (e.g., cross-validation). No study performed phantom study, cost-effectiveness analysis, and prospective validation. Regarding IBSI items, 14 studies (70%) performed signal intensity normalization, while few performed N4 bias-field correction (4, 20%) and skull stripping (3, 15%). Good classification performance (AUC>0.75) was obtained by 11 (55%) studies, but only four of them performed external validation (on sets with 20-60 patients). On the contrary, seven out of the nine studies with lower classification results performed external validation (on sets with 27-126 patients). Conclusions Adherence to RQS and IBSI guidelines was generally low. MGMT methylation status appears to be correlated with radiomic features, but with great heterogeneity of results. To confirm this trend, strict implementation of RQS and IBSI criteria is needed.
Radiomics for MGMT methylation detection in GBM using conventional pre-operative MRI Fabio M. Doniselli1,2, Riccardo Pascuzzo1, Massimiliano Agrò3, Domenico Aquino1, Elena Anghileri, Bianca Pollo, Valeria Cuccarini1, Marco Moscatelli1, Francesco DiMeco, Maria Grazia Bruzzone1, Luca M. Sconfienza2,4 Abstract Objectives: To evaluate the strength and ability of radiomics features extracted from multiple tumor subregions on MR brain images to predict MGMT promoter (MGMT) methylation status in GBM patients through a multiparametric MRI-based radiomics model. Methods: Retrospective single-institution study in a cohort of 277 GBM patients. Radiomics-based models with a minimal set of relevant features and clinical parameters were built for MGMT methylation prediction from a training cohort (196 patients) and tested on an validation cohort (81 patients). Radiomic Quality Score (RQS) was equal to 15. Results: Feature selection methods (Boruta, RFE and LR-EL) identified age and 218 radiomics features. SVM classifier achieved average 73.6% (standard deviation: 6.5%) accuracy and 0.836 (0.054) AUC during the training phase, and 59.3% (95% confidence interval: 47.8%-70.0%) accuracy and 0.553 (0.412-0.686) AUC on the test set. Conclusions: We agree on the probable presence of subtle association between imaging characteristics and MGMT methylation status. However, further verification on the strength of this association is needed, as the low diagnostic performance in the validation cohort is still not sufficiently robust to allow clinically meaningful predictions.

Le cancer du sein est une pathologie fréquente pour lequel les examens TEP/TDM au ¹⁸F-FDG et IRM mammaire sont fréquemment réalisés en routine. Il existe cependant une sous-utilisation des informations apportées par chacune de ces techniques d'imagerie. En pratique, l’interprétation de ces examens est principalement basée sur l’analyse visuelle et l'analyse « quantitative » se résume généralement au SUVmax seul en TEP/TDM et à l’étude du rehaussement du signal après injection de produit de contraste en IRM mammaire (DCE-MRI). L’arrivée de nouvelles machines hybrides TEP/ IRM, nous a amené à évaluer l'apport d’une quantification avancée des images issues de chacune de ces modalités séparément et en combinaison. Cela rejoint un domaine en expansion « la radiomique » qui consiste à extraire un grand nombre de caractéristiques quantitatives des images médicales pour décrypter l’hétérogénéité tumorale ou améliorer la prédiction du pronostic.L’objectif de notre travail était d’étudier l’apport des données radiomiques extraites de l’imagerie TEP au ¹⁸F-FDG et de l’IRM avec injection de produit de contraste réalisées avant traitement pour caractériser l’hétérogénéité tumorale dans le cancer du sein, en prenant en compte les différents sous-types moléculaires de cancer du sein, à savoir les tumeurs luminales (Lum A, Lum B HER2- et Lum B HER2+), triple-négatives et HER2+. Une importance particulière a été portée sur la valeur prédictive des informations radiomiques extraites de ces 2 techniques d’imagerie pour prédire le pronostic dans un groupe de patientes traitées par chimiothérapie néo-adjuvante. L’influence de variations physiologiques telles que l’âge sur le calcul des données radiomiques dans le tissu mammaire normal et cancéreux séparément a également été explorée, de même que la variabilité multicentrique des index radiomiques. L’extraction de ces données radiomiques a été effectuée grace au logiciel LiFex développé au sein du laboratoire IMIV sur une base de données-patientes recueillie en rétrospective.Nous avons rapporté pour la première fois l’influence de l’âge sur le calcul des indices « radiomiques » en TEP dans le tissu mammaire sain dans 2 institutions différentes mais aussi dans les tumeurs mammaires notamment celle triple-négatives. Des associations significatives entre le « phénotype tumoral radiomique » en imagerie TEP et IRM et des données pronostiques reconnues dans le cancer du sein ont été mises en évidence. En outre, nous avons démontré l’existence d’une grande variabilité pour le « profil radiomique » en TEP parmi les tumeurs présentant le même sous-type moléculaire. Cela suggére l’existence d’informations non-redondantes au sein du « phénotype tumoral métabolique » de chaque tumeur mammaire défini par les données radiomiques. L’exploration de cette variabilité s’est révélée intéressante pour améliorer la prédiction de la réponse histologique chez les patientes avec des tumeurs triple-négatives traitées par chimiothérapie néo-adjuvante. Par ailleurs, les mesures effectuées dans la région mammaire péri-tumorale chez les patientes traitées par chimiothérapie néo-adjuvante se sont montrées prédictives pour les patientes avec des tumeurs Lum B HER2-. En IRM nous avons montré l’importance de standardiser la méthode de mesure des caractéristiques radiomiques. Nous avons observé que les caractéristiques radiomiques issues des images DCE-MRI étaient moins associées aux caractéristiques moléculaires des tumeurs et avaient une valeur prédictive moindre. Nous avons également proposé une nouvelle méthode relativement standardisée pour le calcul des données radiomiques en IRM mammaire avec des résultats intéressants mais cette méthode doit encore être optimisée. Cependant, nos résultats suggèrent que les données extraites de la totalité du volume tumorale en IRM compléteraient efficacement les caractéristiques radiomiques TEP et le sous-type moléculaire pour prédire la réponse à la chimiothérapie néo-adjuvante
Breast cancer is a common disease for which ¹⁸F-FDG PET/CT and breast MRI are frequently performed in routine practice. However, the different information provided by each of these imaging techniques are currently under-exploited. Indeed, in routine the interpretation of these scans is mainly based on visual analysis whereas the « quantitative » analysis of PET/CT data is generally limited to the sole use of the SUVmax while in breast MRI, simple parameters to characterize tumor enhancement after injection of contrast medium are used. The advent of PET/MRI machines, calls for an evaluation of the contribution of a more advanced quantification of each of the modalities separately and in combination in the setting of breast cancer. This is along with the concept of « Radiomics » a field currently expanding and which consists in extracting many quantitative characteristics from medical images used in clinical practice to decipher tumor heterogeneity or improve prediction of prognosis. The aim of our work was to study the contribution of radiomic data extracted from ¹⁸F-FDG PET and MRI imaging with contrast injection to characterize tumor heterogeneity in breast cancer taking into account the different molecular subtypes of breast cancer, namely luminal (Lum A, Lum B HER2- and Lum B HER2 +), triple-negative and HER2 + tumors. In this context, we focused on the prediction of prognosis in patients treated with neo-adjuvant chemotherapy. The influence of physiological variations such as age on the calculation of radiomic data in normal breast and breast tumors separately was also explored, as well as the multi-center variability of radioman features. Radiomic features were extracted using the LiFex software developed within IMIV laboratory. The patient database used for the studies were all retrospective data. We reported for the first time the influence of age on the values of radiomic features in healthy breast tissue in patients recruited from 2 different institutions but also in breast tumors especially those with a triple-negative subtype. Similarly, significant associations between the radiomic tumor phenotype in PET and MRI imaging and well-established prognostic factors in breast cancer have been identified. In addition, we showed a large variability in the PET « radiomic profile » of breast tumors with similar breast cancer subtype suggesting complementary information within their metabolic phenotype defined by radiomic features. Moreover, taking into account this variability has been shown to be of particular interest in improving the prediction of pathological response in patients with triple-negative tumors treated with neoadjuvant chemotherapy. A peri-tumoral breast tissue region satellite to the breast tumor was also investigated and appeared to bear some prognostic information in patients with Lum B HER2- tumors treated with neoadjuvant chemotherapy. In MR, we demonstrated the need to harmonize the methods for radiomic feature calculation. Overall, we observed that radiomic features derived from MR were less informative about the molecular features of the tumors than radiomic features extracted from PET data and were of lower prognostic value. Yet, the combination of the enhanced tumor volume in MR with a PET radiomic feature and the tumor molecular subtype yielded enhanced the accuracy with which response to neoadjuvant therapy could be predicted compared to features from one modality only or molecular subtype only

Purpose: Focal pattern in multiple myeloma (MM) seems to be related to poorer survival and differentiation from diffuse to focal pattern on computed tomography (CT) has inter-reader variability. This study aims evaluate if Radiomic approach could help radiologists in differentiating diffuse from focal patterns on CT.
Methods: We retrospectively reviewed imaging data of 70 patients with MM with CT, PET-CT or MRI available before bone marrow transplant. Two general radiologist evaluated, in consensus, CT images to define a focal (at least one lytic lesion > 5 mm in diameter) or a diffuse (lesions < 5 mm, not osteoporosis) pattern. N = 104 Radiomics features were extracted and evaluated with an open source software. Results: We found, after feature reduction, 9 features were different (p < 0.05) in the diffuse and focal patterns AUC of the Radiologists versus Reference Standard was 0.64. Conclusion: A Radiomics approach improves radiological evaluation of focal and diffuse pattern of MM on CT.

Le glioblastome multiforme (GBM) est une tumeur de grade IV représentant 49% de toutes les tumeurs cérébrales. Malgré des modalités de traitement agressives (radiothérapie, chimiothérapie et résection chirurgicale), le pronostic est mauvais avec une survie globale médiane de 12 à 14 mois. Les aractéristiques issues de la neuro imagerie des GBM peuvent fournir de nouvelles opportunités pour la classification, le pronostic et le développement de nouvelles thérapies ciblées pour faire progresser la pratique clinique. Cette thèse se concentre sur le développement de modèles pronostiques exploitant des caractéristiques de radiomique extraites des images multimodales IRM (T1 pré- et post-contraste, T2 et FLAIR). Le contexte méthodologique proposé consiste à i) recaler tous les volumes multimodaux IRM disponibles et en segmenter un volume tumoral unique, ii) extraire des caractéristiques radiomiques et iii) construire et valider les modèles pronostiques par l’utilisation d’algorithmes d’apprentissage automatique exploitant des cohortes cliniques multicentriques de patients. Le coeur des méthodes développées est fondé sur l’extraction de radiomiques (incluant des paramètres d’intensité, de forme et de textures) pour construire des modèles pronostiques à l’aide de deux algorithmes d’apprentissage, les machines à vecteurs de support (support vector machines, SVM) et les forêts aléatoires (random forest, RF), comparées dans leur capacité à sélectionner et combiner les caractéristiques optimales. Les bénéfices et l’impact de plusieurs étapes de pré-traitement des images IRM (re-échantillonnage spatial des voxels, normalisation, segmentation et discrétisation des intensités) pour une extraction de métriques fiables ont été évalués. De plus les caractéristiques radiomiques ont été standardisées en participant à l’initiative internationale de standardisation multicentrique des radiomiques. La précision obtenue sur le jeu de test indépendant avec les deux algorithmes d’apprentissage SVM et RF, en fonction des modalités utilisées et du nombre de caractéristiques combinées atteignait 77 à 83% en exploitant toutes les radiomiques disponibles sans prendre en compte leur fiabilité intrinsèque, et 77 à 87% en n’utilisant que les métriques identifiées comme fiables.Dans cette thèse, un contexte méthodologique a été proposé, développé et validé, qui permet la construction de modèles pronostiques dans le cadre des GBM et de l’imagerie multimodale IRM exploitée par des algorithmes d’apprentissage automatique. Les travaux futurs pourront s’intéresser à l’ajout à ces modèles des informations contextuelles et génétiques. D’un point de vue algorithmique, l’exploitation de nouvelles techniques d’apprentissage profond est aussi prometteuse
Glioblastoma multiforme (GBM) is a WHO grade IV tumor that represents 49% of ail brain tumours. Despite aggressive treatment modalities (radiotherapy, chemotherapy and surgical resections) the prognosis is poor, as médian overall survival (OS) is 12-14 months. GBM’s neuroimaging (non-invasive) features can provide opportunities for subclassification, prognostication, and the development of targeted therapies that could advance the clinical practice. This thesis focuses on developing a prognostic model based on multimodal MRI-derived (Tl pre- and post-contrast, T2 and FLAIR) radiomics in GBM. The proposed methodological framework consists in i) registering the available 3D multimodal MR images andsegmenting the tumor volume, ii) extracting radiomics iii) building and validating a prognostic model using machine learning algorithms applied to multicentric clinical cohorts of patients. The core component of the framework rely on extracting radiomics (including intensity, shape and textural metrics) and building prognostic models using two different machine learning algorithms (Support Vector Machine (SVM) and Random Forest (RF)) that were compared by selecting, ranking and combining optimal features. The potential benefits and respective impact of several MRI pre-processing steps (spatial resampling of the voxels, intensities quantization and normalization, segmentation) for reliable extraction of radiomics was thoroughly assessed. Moreover, the standardization of the radiomics features among methodological teams was done by contributing to “Multicentre Initiative for Standardisation of Radiomics”. The accuracy obtained on the independent test dataset using SVM and RF reached upto 83%- 77% when combining ail available features and upto 87%-77% when using only reliable features previously identified as robust, depending on number of features and modality. In this thesis, I developed a framework for developing a compréhensive prognostic model for patients with GBM from multimodal MRI-derived “radiomics and machine learning”. The future work will consists in building a unified prognostic model exploiting other contextual data such as genomics. In case of new algorithm development we look forward to develop the Ensemble models and deep learning-based techniques

Radiomics is an emerging translational field of research aiming to extract mineable high-dimensional data from clinical images able to offer information about prognosis of cancer patients. The radiomics process relies on a multi-step path that ends in the construction of a predictive model, tailored on specific outcomes. The main steps of the radiomics process are: image acquisition and reconstruction, segmentation, features extraction, model building. Each of these steps shows its own challenges to make the final model robust and reliable. Patients with Non-small cell lung cancer (NSCLC) have baseline computed tomography (CT) and/or fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) imaging for diagnosis and staging. The aim of this study was to evaluate whether a model based on radiomic and clinical features may be associated with lymph node (LN) status and overall survival (OS) in NSCLC patients. Patients with a pathological stage up to T3N1 were retrospectively selected and divided into training and validation sets. For the prediction of positive LNs and OS, the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression model was used; univariable and multivariable logistic regression analysis assessed the association of clinical-radiomic variables and endpoints. All tests were repeated after dividing the groups according to the CT reconstruction algorithm. p-values < 0.05 were considered significant. 270 patients were included and divided into training (n = 180) and validation sets (n = 90). Transfissural extension was significantly associated with positive LNs. For OS prediction, high- and low-risk groups were different according to the radiomics score, also after dividing the two groups according to reconstruction algorithms. In conclusion, a combined clinical–radiomics model was not superior to a single clinical or single radiomics model to predict positive LNs. A radiomics model was able to separate high-risk and low-risk patients for OS

Essendo il tumore cerebrale primario più comune, il glioblastoma si presenta come un tumore maligno estremamente difficile da trattare con esiti spesso negativi nonostante i trattamenti applicati. L'epidemiologia molecolare variabile del glioblastoma presente tra i pazienti e l'eterogeneità intratumorale spiegano il fallimento delle attuali modalità di trattamento universali. La radiomica è un campo emergente dell’imaging quantitativo che mira a fare previsioni e derivare intuizioni mediche utilizzando funzionalità avanzate dell’imaging basate su caratteristiche quantitative estratte da immagini mediche per descrivere oggettivamente e quantitativamente i fenotipi tumorali oltre che ad essere utilizzata come mezzo per migliorare la diagnosi clinica o l'esito. Per quanto riguarda il glioblastoma, la radiomica ha fornito strumenti potenti e non invasivi per ottenere informazioni sui processi fisiopatologici e sulle risposte terapeutiche. Gli studi radiomici hanno inoltre prodotto una significativa comprensione biologica delle caratteristiche dell'imaging.

The aim of this thesis is to determine diagnostic performance of machine learning in differentiating between atypical cartilaginous tumor (ACT) and high-grade chondrosarcoma (CS) based on radiomic features derived from magnetic resonance imaging (MRI) and computed tomography (CT). In chapter 2, the concept of radiomics of musculoskeletal sarcomas is introduced and a systematic review on radiomic feature reproducibility and validation strategies is conducted. In chapter 3, a preliminary study is performed to investigate the performance of MRI radiomics-based machine learning in discriminating ACT from high-grade CS, using a single-center cohort, in comparison with an expert radiologist. In chapter 4, the influence of interobserver segmentation variability on the reproducibility of CT and MRI radiomic features of cartilaginous bone tumors is assessed. In chapter 5, the performance of CT radiomics-based machine learning in discriminating ACT from high-grade CS of long bones is determined and validated using independent data from a multicenter cohort, compared to an expert radiologist. In chapter 6, the performance of MRI radiomics-based machine learning in differentiating between ACT and grade II CS of long bones is determined and validated using independent data from a multicenter cohort, in comparison with an expert radiologist. Finally, in chapter 7, the main results and implications of this thesis are summarized and discussed.

Questa tesi ha lo scopo di presentare il deep learning e una delle sue applicazioni che ha avuto molto successo nell'analisi delle immagini: la rete neurale convoluzionale. In particolare, si espongono i vantaggi, gli svantaggi e i risultati ottenuti nell'applicazione delle reti convoluzionali alla radiomica, una nuova disciplina che prevede l'estrazione di un elevato numero di feature dalle immagini mediche per elaborare modelli di supporto a diagnosi e prognosi. Nel primo capitolo si introducono concetti di machine learning utili per comprendere gli algoritmi di apprendimento usati anche nel deep learning. Poi sono presentate le reti neurali, ovvero le strutture su cui si basano gli algoritmi di deep learning. Infine, viene spiegato il funzionamento e gli utilizzi delle reti neurali convoluzionali. Nel secondo capitolo si espongono le tecniche e gli utilizzi della radiomica e, infine, i vantaggi di usare le reti neurali convoluzionali in quest'ambito, presentando alcuni recenti studi portati a termine in merito.

Lung cancer is the leading cause of cancer-related death in the United States and worldwide. Early detection of lung cancer can help improve patient outcomes, and survival prediction can inform plans of treatment. By extracting quantitative features from computed tomography scans of lung cancer, predictive models can be built that can achieve both early detection and survival prediction. To build these predictive models, first a detected lung nodule is segmented, then image features are extracted, and finally a model can be built utilizing image features to make predictions. These predictions can help radiologists improve cancer care.

Building predictive models based on medical images is the basis of the budding field of radiomics. The hypothesis is that images contain phenotypic information that can be extracted to aid prediction and that automated methods can detect some things beyond human detection. With improved detection and predictive models radiomics aims to help assist radiologists and oncologists provide personalized care.

In this work a model is presented to predict long term survival versus short term survival. Forty adenocarcinoma diagnostic lung computed tomography (CT) scans from Moffitt Cancer Center were analyzed for survival prediction. These forty cases were in the top and bottom quartile for survival. A decision tree classifier was able to predict the survival group with an accuracy of 77.5% using five image features chosen from 219 using relief-f.

Another contribution of this work is a model for predicting cancer from suspicious nodules. The national lung screening trial was used to build a training set of 261 screening CTs and a test set of 237 CTs. These images were taken at the initial screening, one and two years before cancer developed. From these precursor images, which nodules developed into cancer, could be predicted at 76.79% accuracy with an area under the receiver operating characteristic curve of 0.82. A risk score was also developed to provide a measure of risk during screening. The developed risk score performed favorably in predictive accuracy compared to Lung-RADS on this data set.

The Data Science Bowl was also entered and this work examines the knowledge gained from a large-scale competition to improve imaging. In this competition participants were tasked with predicting cancer from 1397 training cases on 506 test cases. The winning entry performed with a logLoss of 0.39975 while making use of all the training data while our entry scored 1.56555 with a different set of training data. A lower logLoss shows greater accuracy. This work explains our approach and examines the winning entry.

An overview of the state of radiomicis as it applies to lung cancer is also provided. These contributions of predictive models will help to provide decision support to medical practitioners. By providing tools to the medical field the goal is to advance automated medical imaging to aid clinicians in creating diagnosis and treatment plans.

Lung cancer has been a major cause of death among types of cancers in the world. In the early stages, lung nodules can be detected by the aid of imaging modalities such as Computed Tomography (CT). In this stage, radiologists look for irregular rounded-shaped nodules in the lung which are normally less than 3 centimeters in diameter. Recent advancements in image analysis have proven that images contain more information than regular parameters such as intensity, histogram and morphological details. Therefore, in this project we have focused on extracting quantitative, hand-crafted features from nearly 1400 lung CT images to train a variety of classifiers based on them. In the first experiment, in total 424 Radiomics features per image has been used to train classifiers such as: Random Forest (RF), Support Vector Machine (SVM), Decision Tree (DT), Naive Bayes (NB), Linear Discriminant Analysis (LDA) and Multi-Layer Perceptron (MLP). In the second experiment, we evaluate each feature category separately with our classifiers. The third experiment includes wrapper feature selection methods (Forward/Backward/Recursive) and filter-based feature selection methods (Fisher score, Gini Index and Mutual information). They have been implemented to find the most relevant feature set in model construction. Performance of each learning method has been evaluated by accuracy score, wherewe achieved the highest accuracy of 78% with Random Forest classifier (74% in 5-fold average) and 0.82 Area Under the Receiver Operating Characteristics (AUROC) curve. After RF, NB and MLP showed the best average accuracy of 71.4% and 71% respectively.

With the importance of medical images for disease diagnosis and prognosis becoming widely recognized, medical image analysis has drawn much attention among researchers and clinicians. The goal of medical image analysis is to identify diagnostic and prognostic information from medical images and to establish diagnosis/prognosis models for assisting in clinical decision making and personalized treatments. Deep learning-based methods have achieved great success in computer vision research. This success is mainly attributed to its outstanding ability to learn high-level pattern representations from big data, and has motivated many investigators for applying deep learning-based algorithms in medical image analysis. The objectives of this thesis are to explore and develop deep learning methods for two medical image analysis tasks: medical image registration and radiomics. Firstly, we focused on medical image registration, a fundamental step of various medical image analysis tasks. We identified that a key challenge for accurate image registration is the variations in image appearance. Hence, we proposed an Appearance Adjustment Network (AAN) where we leverage anatomy edges, through an anatomy-constrained loss function, to generate an anatomy-preserving appearance transformation. We designed the AAN so that it can be readily embedded into a wide range of deep learning-based registration frameworks, to reduce the appearance differences between input image pairs and thereby improve registration accuracy. In this study, we experimented with Brain MRI data and observed improvements in registration accuracy. The results show that our AAN enhanced the baseline registration methods by roughly 2% in Dice score, while adding a fractional computational load. Secondly, we explored radiomics-based survival prediction of patients with advanced Nasopharyngeal Carcinoma (NPC). Radiomics refers to the extraction and analysis of high-dimensional quantitative features from non-invasive images. In this study, we incorporated deep learning into radiomics and developed an end-to-end multi-modality deep-learning model using pretreatment PET/CT images to predict 5-year progression-free survival. Furthermore, the deep-learning model was extensively compared with a large number of conventional radiomics methods for prognostic performance. The results show that the proposed deep-learning model outperformed conventional radiomics methods by roughly 5% in AUC. Our experimental results demonstrated that our models on two tasks outperform the existing methods, suggesting that deep learning is an effective tool for enhancing medical image analysis.

The significant clinical heterogeneity of Multiple Myeloma (MM) patients implies that a set of consolidated biomarkers is currently missing. Radiomics is an advanced, quantitative feature-based methodology for image analysis. We assess the feasibility of an AI-based approach for the automatic stratification of MM patients from CT data, and for the automatic identification of radiological biomarkers with a possible prognostic value. A retrospective analysis of n = 33 transplanted MM with focal lesion were performed via an open-source toolbox that extracted 109 radiomics features. The redundancy reduction was realized via correlation and principal component analysis. The highest sensitivity and critical success index (CSI) were obtained representing each patient, with 17 focal features selected via correlation with the 24 features describing the overall skeletal asset. The Mann– Whitney U-test showed that three among the 17 imaging descriptors passed the null hypothesis. This computational approach to the interpretation of radiomics features shows the potential for the stratification of relapsed and non-relapsed MM patients, and could represent a prognostic image-based procedure for determining the disease follow-up and therapy.

Motivation: Brain and central nervous system tumours form the second most common group of cancers in children in the UK, accounting for 27% of all childhood cancers. Despite current advances in magnetic resonance imaging (MRI), non-invasive characterisation of paediatric brain tumours remains challenging. Radiomics, the high-throughput extraction and analysis of quantitative image features (e.g. texture), offers potential solutions for tumour characterisation and decision support. Aim and Methods: In search for diagnostic and prognostic oncological markers, the aim of this thesis was to study the application of MRI texture analysis (TA) for the characterisation of paediatric brain tumours. To this end, single and multi-centre experiments were carried out, within a supervised classification framework, on clinical MR imaging datasets of common brain tumour types. Results: TA of conventional MRI was successfully used for diagnostic classification of common paediatric brain tumours. A key contribution of this thesis was to provide evidence that diagnostic classification could be optimised by extending the analysis to include three-dimensional features obtained from multiple MR imaging slices. In addition to this, TA was shown to have a good cross-centre transferability, which is essential for long-term clinical adoption of the technique. Finally, fifteen textural features extracted from T2-weighted MRI were identified to be of significant prognostic value for paediatric medulloblastoma. Conclusion: It was shown that MRI TA provides valuable quantifiable information that can supplement qualitative assessments conducted by radiologists, for the characterisation of paediatric brain tumours. TA can potentially have a large clinical impact, since MR imaging is routinely used in the brain cancer clinical work-flow worldwide, providing an opportunity to improve personalised healthcare and decision-support at low cost.

The thesis examined the association between breast cancer (BC) characteristics and image-based phenotypes such as mammographic density (MD) and radiomic features. It also examined the prognostic utility of MD in BC patients and changes in MD over time following BC treatments. MD was measured using approaches such as BI-RADS4th edition, LIBRA and AutoDensity at different stages of the thesis. A MATLAB R2018a software package was used to extract 33 global radiomic features from the ipsilateral breast mammograms. BC was found to arise mostly from dense tissues, with cancers in dense regions demonstrating human epidermal growth receptor 2 and carcinoma in situ characteristics. The GLRLM-based radiomic feature from the ipsilateral breast predicted lymph node status among younger patients with high baseline MD, and fractal dimension-based radiomic feature predicted tumour size among women with low baseline MD. Relative to high MD at baseline, low MD conferred higher risk of death from breast cancer. Two distinct trends in MD changes were observed over time: MD decrease, which was modified by demographic factors and MD increase. In summary, BC is most likely to develop in dense regions of the breast and low baseline MD is associated with poor BC prognosis. In addition, global radiomic features from the ipsilateral breast mammograms predict lymph node status and tumour size in some categories of women. Furthermore, women undergoing BC treatment demonstrate changes in MD over time, and MD decrease over time is modified by patients’ demographic characteristics.

Cervical cancer remains the third most commonly diagnosed gynecological malignancy in the United States and throughout the world despite being potentially preventable. Patients diagnosed with cervical cancer may develop local recurrence in the cervix and surrounding structures (vaginal apex, parametrial, or paracervical), regional recurrence in pelvic lymph nodes, distant metastasis, or a combination of all. The management of such treatment outcomes has not been subject to rigorous investigation. Therefore, there is a need for studies and clinical trials that focus on decision making to support the choice of the best treatment modality that leads to the minimal number of adverse treatment outcomes. Medical imaging plays a vital role in the initial diagnosis, staging, and guiding treatment decisions for cancer patients. Positron Emission Tomography-Computed Tomography (PET/CT) hybrid scanner has proven to be a primary functional imaging modality in the oncology clinic. A typical oncological application of PET/CT aims to examine the whole body for high tracer uptake as a sign of tumorous lesions or metastasis using 18F-Fluoro-2-deoxy-D-glucose (18F-FDG). This radiopharmaceutical has been proven to be useful for the quantitative determination of regional glucose metabolism localized in the brain, heart, bladder, and, fortunately, in tumors. Currently, 18F-FDG measured on PET is the prominent radiotracer in cancer staging and follow-up imaging. In the –omics1 era, mining data to derive inherent information about a system has influenced the medical field, especially oncological imaging. The process of radiomics involves high throughput analysis of medical images to extract a large number of quantified features that are presented as a decision supporting tool for clinicians in terms of various clinical tasks such as staging, prediction, and prognosis. In recent studies, the focus of radiomics has exceeded the whole-tumor analysis to include the quantification of habitats, sub-regions within the tumor volume defined based on specific criteria, with the intent to investigate the diversity extent of the intratumor heterogeneity as robust descriptors and predictors of clinicopathological factors. The presented work is a retrospective analysis of a cohort consisting of pretreatment Positron Emission Tomography and Computed Tomography (PET/CT) hybrid scans of cervical cancer patients consecutively treated with radiochemotherapy. We extracted radiomic features from the primary cervical tumor volumes, and voxel intensity-based features from tumor habitats to analyze the tumors’ heterogeneity based on 18Flourodeoxyglocuse (18F-FDG) uptake of PET, and Hounsfield Units (HU) of CT to obtain useful tumor information, which might be associated with treatment outcomes. To our knowledge, a limited number of studies have focused on investigating the potential role of radiomic features on cervical cancer PET/CT images. Briefly, the workflow of this study consisted of investigating parameters that might affect radiomic features predictive performance by evaluating the reproducibility of radiomic features extracted from 18F-FDG PET images for segmentation methods, gray levels discretization, and PET reconstruction algorithms. Afterward, we used these features to predict cervical treatment outcomes after radiochemotherapy. Due to the use of human data, this research study acquired the approval of the institutional review board (IRB) at the University of South Florida.

[ES] En los últimos años, los investigadores han intentado explotar la información de las imágenes médicas a través de la evaluación de parámetros cuantitativos para ayudar a los clínicos con el diagnóstico de enfermedades. Esta práctica ha sido bautizada como radiomics. El análisis de texturas proporciona una gran variedad de parámetros que permiten cuantificar la heterogeneidad característica de diferentes tejidos, especialmente cuando se obtienen de imagen por resonancia magnética (IRM). Basándonos en esto, decidimos estudiar las posibilidades de los parámetros texturales extraídos de IRM para caracterizar varios trastornos cerebrales. El potencial de las texturas se analizó con enfoques de aprendizaje automático, usando diferentes clasificadores y métodos de selección de características para hallar el modelo óptimo para cada tarea específica. En esta tesis, la metodología radiomics se usó para realizar cuatro proyectos independientes. En el primer proyecto, estudiamos la diferenciación entre glioblastomas multiformes (GBMs) y metástasis cerebrales (MCs) en IRM convencional. Estos tipos de tumores cerebrales pueden confundirse al diagnosticarse, ya que presentan un perfil radiológico similar y los datos clínicos pueden no ser concluyentes. Con el fin de evitar procedimientos exhaustivos e invasivos, estudiamos el poder discriminatorio de texturas 2D extraídas de imágenes de referencia T1 filtradas y sin filtrar. Los resultados sugieren que los parámetros texturales proporcionan información sobre la heterogeneidad de los GBMs y las MCs que puede servir para distinguir con precisión ambas lesiones cuando se utiliza un enfoque de aprendizaje automático adecuado. En el segundo proyecto, analizamos la clasificación de las MCs según su origen primario en IRM de referencia. En un porcentaje de pacientes, las MCs son diagnosticadas como la primera manifestación de un tumor primario desconocido. Con el fin de detectar el tumor primario de una forma no invasiva y más rápida, examinamos la capacidad del análisis de texturas 2D y 3D para diferenciar las MCs derivadas de los tumores primarios más propensos a metastatizar (cáncer de pulmón, cáncer de mama y melanoma) en imágenes T1. Los resultados mostraron que se logra una alta precisión al usar un conjunto reducido de texturas 3D para diferenciar MCs de cáncer de pulmón de MCs de cáncer de mama y melanoma. En el tercer proyecto, evaluamos las propiedades del hipocampo en IRM para identificar las diferentes etapas de la enfermedad de Alzheimer (EA). Los criterios actuales para diagnosticar la EA requieren la presencia de déficits cognitivos severos. Con la idea de establecer nuevos biomarcadores para detectar la EA en sus primeras etapas, evaluamos un conjunto de texturas 2D y 3D extraídas de IRM del hipocampo de pacientes con EA avanzada, deterioro cognitivo leve y normalidad cognitiva. Muchos parámetros de textura 3D resultaron ser estadísticamente significativos para diferenciar entre pacientes con EA y sujetos de las otras dos poblaciones. Al combinar estos parámetros con técnicas de aprendizaje automático, se obtuvo una alta precisión. En el cuarto proyecto, intentamos caracterizar los patrones de heterogeneidad del ictus cerebral isquémico en IRM estructural. En IRM cerebral de individuos de edad avanzada, algunos procesos patológicos presentan características similares, como las lesiones por ictus y las hiperintensidades de la sustancia blanca (HSBs). Dado que los ictus afectan también al tejido adyacente, decidimos estudiar la viabilidad de texturas 3D extraídas de las HSBs, la sustancia blanca no afectada y las estructuras subcorticales para diferenciar sujetos afectados por ictus lacunares o corticales visibles en IRM convencional (imágenes T1, T2 y FLAIR) de sujetos sin ictus. Las texturas no sirvieron para diferenciar ictus corticales y lacunares, pero se lograron resultados prometedores para discernir pacientes qu
[CAT] En els últims anys, els investigadors han intentat explotar la informació de les imatges mèdiques a través de l'avaluació de nombrosos paràmetres quantitatius per ajudar els clínics amb el diagnòstic i la valoració de malalties. Aquesta pràctica ha sigut batejada com radiomics,. L'anàlisi de textures proporciona una gran varietat de paràmetres que permeten quantificar l'heterogeneïtat característica de diferents teixits, especialment quan s'obtenen a partir d'imatge per ressonància magnètica (IRM). Basant-nos en aquests fets, vam decidir estudiar les possibilitats dels paràmetres texturals extrets d'IRM per caracteritzar diversos trastorns cerebrals. El potencial de les textures es va analitzar amb mètodes d'aprenentatge automàtic, usant diferents classificadors i mètodes de selecció de característiques per trobar el model òptim per a cada tasca específica. En aquesta tesi, la metodologia radiomics es va emprar per realitzar quatre projectes independents. En el primer projecte, vam estudiar la diferenciació entre glioblastomes multiformes (GBMs) i metàstasis cerebrals (MCs) en IRM convencional. Aquests tipus de tumors cerebrals poden confondre's al diagnosticar-se ja que presenten un perfil radiològic similar i les dades clíniques poden no ser concloents. Per tal d'evitar procediments exhaustius i invasius, vam estudiar el poder discriminatori de textures 2D extretes d'imatges de referència T1 filtrades i sense filtrar. Els resultats suggereixen que els paràmetres texturals proporcionen informació sobre l'heterogeneïtat dels GBMs i les MCs que pot servir per distingir amb precisió ambdues lesions quan s'utilitza una aproximació d'aprenentatge automàtic adequada. En el segon projecte, vam analitzar la classificació de MCs segons el seu origen primari en IRM de referència. En un percentatge de pacients, les MCs són diagnosticades com la primera manifestació d'un tumor primari desconegut. Per tal de detectar el tumor primari d'una forma no invasiva i més ràpida, vam examinar la capacitat de l'anàlisi de textura 2D i 3D per diferenciar les MCs derivades dels tumors primaris més propensos a metastatitzar (càncer de pulmó, càncer de mama i melanoma) en imatges T1. Els resultats van mostrar que s'aconsegueix una alta precisió quan s'utilitza un conjunt reduït de textures 3D per diferenciar les MCs de càncer de pulmó de les MCs de càncer de mama i melanoma. En el tercer projecte, vam avaluar les propietats de l'hipocamp en la IRM per identificar les diferents etapes de la malaltia d'Alzheimer (MA). Els criteris actuals per diagnosticar la MA requereixen la presència de dèficits cognitius severs. Amb la idea d'establir nous biomarcadors per detectar la MA en les seues primeres etapes, vam avaluar un conjunt de textures 2D i 3D extretes d'IRM de l'hipocamp de pacients amb MA avançada, deteriorament cognitiu lleu i normalitat cognitiva. Molts paràmetres de textura 3D van resultar ser estadísticament significatius per diferenciar entre pacients amb MA i individus de les altres dues poblacions. En combinar aquests paràmetres amb tècniques d'aprenentatge automàtic, es va obtenir una alta precisió. En el quart projecte, vam intentar caracteritzar els patrons d'heterogeneïtat de l'ictus cerebral isquèmic en la IRM estructural. En la IRM cerebral d'individus d'edat avançada, alguns processos patològics presenten característiques similars, com les lesions per ictus i les hiperintensitats de la substància blanca (HSBs). Atès que els ictus tenen efecte també en teixit adjacent, vam decidir estudiar la viabilitat de textures 3D extretes de les HSBs, la substància blanca no afectada i les estructures subcorticals per diferenciar individus afectats per ictus llacunars o corticals visibles en IRM convencional (imatges T1, T2 i FLAIR) d'individus sense ictus. Les textures no foren útils per diferenciar ictus corticals i llacunars, però es van obtenir resultats prometedors per disce
[EN] Over the last years, researchers have attempted to exploit the information provided by medical images through the evaluation of numerous imaging quantitative parameters in order to help clinicians with the diagnosis and assessment of many lesions and diseases. This practice has been recently named as radiomics. Texture analysis supply a wide range of features that allow quantifying the distinctive heterogeneity of different tissues, especially when obtained from magnetic resonance imaging (MRI). With this in mind, we decided to study the possibilities of texture features from MRI in order to characterize several disorders that affect the human brain. The potential of texture features was analyzed with various machine learning approaches, involving different classifiers and feature selection methods so as to find the optimal model to accomplish each specific task. In this thesis, the radiomics methodology was used to perform four independent projects. In the first project, we studied the differentiation between glioblastomas (GBMs) and brain metastases (BMs) in conventional MRI. Sometimes these types of brain tumors can be misdiagnosed since they may present a similar radiological profile and the clinical data may be inconclusive. With the aim of avoiding exhaustive and invasive procedures, we studied the discriminatory power of a large amount of 2D texture features extracted from baseline original and filtered T1-weighted images. The results suggest that 2D texture features provide some heterogeneity information of GBMs and BMs that can help in their accurate discernment when using the proper machine learning approach. In the second project, we analyzed the classification of BMs by their primary site of origin in baseline MRI. A percentage of patients are diagnosed with BM as the first manifestation of an unknown primary tumor. In order to detect the primary tumor in a faster non-invasive way, we examined the capability of 2D and 3D texture analysis to differentiate BMs derived from the most common primary tumors (lung cancer, breast cancer and melanoma) in T1-weighted images. The results showed that high accuracy was achieved when using a reduced set of 3D descriptors to differentiate lung cancer BMs from breast cancer and melanoma BMs. In the third project, we evaluated the hippocampus MRI profile of Alzheimer's disease (AD) patients to identify the different stages of the disease. The current criteria for diagnosing AD require the presence of relevant cognitive deficits. With the purpose of establishing new biomarkers to detect AD in its early stages, we evaluated a set of 2D and 3D texture features extracted from MRI scans of the hippocampus of patients with advanced AD, early mild cognitive impairment and cognitive normality. Many 3D texture parameters resulted to be statistically significant to differentiate between AD patients and subjects from the other two populations. When combining these 3D parameters with machine learning techniques, high accuracy was obtained. In the fourth project, we attempted to characterize the heterogeneity patterns of ischemic stroke in structural MRI. In brain MRI of older individuals, some pathological processes present similar imaging characteristics, like in the case of stroke lesions and white matter hyperintensities (WMH) of diverse natures. Given that stroke effects are present not only in the affected region, but also in unaffected tissue, we investigated the feasibility of 3D texture features from WMH, normal-appearing white matter and subcortical structures to differentiate individuals who had a lacunar or cortical stroke visible on conventional brain MRI (T1-weighted, T2-weighted and FLAIR images) from subjects who did not. Texture features were not useful to differentiate between post-acute cortical and lacunar strokes, but promising results were achieved for discerning between patients presenting an old stroke and normal-ageing patients who never had a stroke.
Ortiz Ramón, R. (2019). Radiomics for diagnosis and assessing brain diseases: an approach based on texture analysis on magnetic resonance imaging [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/119118
TESIS

РЕЗИМЕИстраживање које је примењено у оквиру докторске дисертације усмерено је наевалуацију примене едукативних радионица као методичког модела у настави средње стручне школе. Реализацијом емпиријског истраживања настојали смо доћи до сазнања да ли постоје разлике у квалитету, ефикасности и ефектима образовно-васпитног процеса када се настава изводи путем модела едукативне радионице и када се изводи путем традиционалне наставе. Разлике смо посматрали у односу на квалитет, нивое и ретенцију усвојеног знања, мисаону активност ученика, педагошку комуникацију и процес индивидуализације и диференцијације у настави стручних предмета медицинске школе.Теоријски оквир истраживања представља дефинисање основних појмоваистраживања са филозофско-теоријским утемељењем и освртом на теоријске тенденције савремених истраживања у настави. Он обухвата теоријска објашњења конструктивистичке димензије интерактивне наставе, концепцију развојног приступа у раду са младима, теорије интерактивног учења и наставе, моделовање радионичарског облика наставе као и начине праћења и вредновања квалитета васпитно-образовног система.Ради провере хипотеза истраживања примењено је истраживање са паралелним групама, а од метода коришћена је дескриптивна метода. Као технике коришћене су анализа документације, анкетирање, тестирање и планско и организовано посматрање часова у експерименталној и контролној групи. Подаци су прикупљени уз помоћ већег броја инструмената: тестови знања, упитници и скале ставова за ученике и наставнике и оригинални протоколи за праћење и евалуацију наставног часа. Од савременихстатистичких поступака приликом обраде и анализе података коришћене су дескриптивна статистика, т-тест, хи квадрат и анализа коваријансе за поновљена мерења.Резултати статистичке анализе о утицају едукативних радионица на квалитет,нивое и ретенцију усвојеног знања, мисаону активност ученика, педагошку комуникацију и процес индивидуализације и диференцијације у настави показали су се детерминишућим по три од пет дефинисаних варијабли истраживања. То нас упућује на закључак да постоји узрочно-последична веза између наведених варијабли. Наше истраживање показало је дасу едукативне радионице као методички модел омогућиле да се у току реализације часа .Евалуација примене едукативних радионица као методичког модела у настави средње стручне школедокторска дисертација Бојана Перић Пркосовачки стручне наставе развије интерактивна педагошка комуникација постављајући ученика каомисаоно активног субјекта, уз поштовање индивидуализације и диференцијацијенаставног процеса. Такође, добијени подаци путем тестова знања сигнализирају да се квалитет, нивои стечених знања и умења као и њихова ретенција код ученика експерименталне и контролне групе статистички занемарљиво разликују.Посматрањем, праћењем и евалуацијом едукативних радионица током реализације наставе установили смо да се едукативним радионицама постиже интерактивна комуникација у одељењу, мисаона активност ученика добија шире размере, а ученик се препознаје као ангажовани појединац који диференцијацијом наставног процеса може да процени своје индивидуалне могућности, знање и способности.
REZIMEIstraživanje koje je primenjeno u okviru doktorske disertacije usmereno je naevaluaciju primene edukativnih radionica kao metodičkog modela u nastavi srednje stručne škole. Realizacijom empirijskog istraživanja nastojali smo doći do saznanja da li postoje razlike u kvalitetu, efikasnosti i efektima obrazovno-vaspitnog procesa kada se nastava izvodi putem modela edukativne radionice i kada se izvodi putem tradicionalne nastave. Razlike smo posmatrali u odnosu na kvalitet, nivoe i retenciju usvojenog znanja, misaonu aktivnost učenika, pedagošku komunikaciju i proces individualizacije i diferencijacije u nastavi stručnih predmeta medicinske škole.Teorijski okvir istraživanja predstavlja definisanje osnovnih pojmovaistraživanja sa filozofsko-teorijskim utemeljenjem i osvrtom na teorijske tendencije savremenih istraživanja u nastavi. On obuhvata teorijska objašnjenja konstruktivističke dimenzije interaktivne nastave, koncepciju razvojnog pristupa u radu sa mladima, teorije interaktivnog učenja i nastave, modelovanje radioničarskog oblika nastave kao i načine praćenja i vrednovanja kvaliteta vaspitno-obrazovnog sistema.Radi provere hipoteza istraživanja primenjeno je istraživanje sa paralelnim grupama, a od metoda korišćena je deskriptivna metoda. Kao tehnike korišćene su analiza dokumentacije, anketiranje, testiranje i plansko i organizovano posmatranje časova u eksperimentalnoj i kontrolnoj grupi. Podaci su prikupljeni uz pomoć većeg broja instrumenata: testovi znanja, upitnici i skale stavova za učenike i nastavnike i originalni protokoli za praćenje i evaluaciju nastavnog časa. Od savremenihstatističkih postupaka prilikom obrade i analize podataka korišćene su deskriptivna statistika, t-test, hi kvadrat i analiza kovarijanse za ponovljena merenja.Rezultati statističke analize o uticaju edukativnih radionica na kvalitet,nivoe i retenciju usvojenog znanja, misaonu aktivnost učenika, pedagošku komunikaciju i proces individualizacije i diferencijacije u nastavi pokazali su se determinišućim po tri od pet definisanih varijabli istraživanja. To nas upućuje na zaključak da postoji uzročno-posledična veza između navedenih varijabli. Naše istraživanje pokazalo je dasu edukativne radionice kao metodički model omogućile da se u toku realizacije časa .Evaluacija primene edukativnih radionica kao metodičkog modela u nastavi srednje stručne školedoktorska disertacija Bojana Perić Prkosovački stručne nastave razvije interaktivna pedagoška komunikacija postavljajući učenika kaomisaono aktivnog subjekta, uz poštovanje individualizacije i diferencijacijenastavnog procesa. Takođe, dobijeni podaci putem testova znanja signaliziraju da se kvalitet, nivoi stečenih znanja i umenja kao i njihova retencija kod učenika eksperimentalne i kontrolne grupe statistički zanemarljivo razlikuju.Posmatranjem, praćenjem i evaluacijom edukativnih radionica tokom realizacije nastave ustanovili smo da se edukativnim radionicama postiže interaktivna komunikacija u odeljenju, misaona aktivnost učenika dobija šire razmere, a učenik se prepoznaje kao angažovani pojedinac koji diferencijacijom nastavnog procesa može da proceni svoje individualne mogućnosti, znanje i sposobnosti.

Thesis (M.S. in Operations Research)--Naval Postgraduate School, September 2008.
Thesis Advisor(s): Enns, John. "September 2008." Description based on title screen as viewed on November 6, 2008. Includes bibliographical references (p. 55). Also available in print.

L’imagerie multi-modale Tomographie par émission de positons (TEP)/ Tomodensitométrie(TDM) est la modalité d’imagerie la plus utilisée pour le diagnostic et le suivi des patients en oncologie. Les images obtenues par cette méthode offrent une cartographie à la fois de la densité des tissus (modalité TDM) mais également une information sur l’activité métabolique des lésions tumorales (modalité TEP). L’analyse plus approfondie de ces images acquises en routine clinique a permis d’extraire des informations supplémentaires quant à la survie du patient ou à la réponse au(x) traitement(s). Toutes ces nouvelles données permettent de décrire le phénotype d’une lésion de façon non invasive et sont regroupées sous le terme de Radiomics. Cependant, le nombre de paramètres caractérisant la forme ou la texture des lésions n’a cessé d’augmenter ces dernières années et ces données peuvent être sensibles à la méthode d’extraction ou encore à la modalité d’imagerie employée. Pour ces travaux de thèse, la variabilité de ces caractéristiques a donc été évaluée sur les images TDM et TEP à l’aide d’une cohorte test-retest : pour chaque patient, deux examens effectués dans les mêmes conditions, espacés d’un intervalle de l’ordre de quelques jours sont disponibles. Les métriques reconnues comme fiables à la suite de cette analyse sont exploitées pour l’étude de la survie des patients dans le cadre du cancer du poumon. La construction d’un modèle pronostique à l’aide de ces métriques a permis, dans un premier temps, d’étudier la complémentarité des informations fournies par les deux modalités. Ce nomogramme a cependant été généré par simple addition des facteurs de risque. Dans un second temps, les mêmes données ont été exploitées afin de construire un modèle pronostique à l’aide d’une méthode d’apprentissage reconnue comme robuste : les machines à vecteurs de support ou SVM (support vector machine). Les modèles ainsi générés ont ensuite été testés sur une cohorte prospective en cours de recrutement afin d’obtenir des résultats préliminaires sur la robustesse de ces nomogrammes
Positron emission tomography (PET) / Computed tomography (CT) multi-modality imaging is the most commonly used imaging technique to diagnose and monitor patients in oncology. PET/CT images provide a global tissue density description (CT images) and a characterization of tumor metabolic activity (PET images). Further analysis of those images acquired in clinical routine supplied additional data as regards patient survival or treatment response. All those new data allow to describe the tumor phenotype and are generally grouped under the generic name of Radiomics. Nevertheless, the number of shape descriptors and texture features characterising tumors have significantly increased in recent years and those parameters can be sensitive to exctraction method or whether to imaging modality. During this thesis, parameters variability, computed on PET and CT images, was assessed thanks to a test-retest cohort : for each patient, two groups of PET/CT images, acquired under the same conditions but generated with an interval of few minutes, were available. Parameters classified as reliable after this analysis were exploited for survival analysis of patients in the context of non-small cell lug cancer (NSCLC).The construction of a prognostic model with those metrics permitted first to study the complementarity of PET and CT texture features. However, this nomogram has been generated by simply adding risk factors and not with a robust multi-parametric analysis method. In the second part, the same data were exploited to build a prognostic model using support vector machine (SVM) algorithm. The models thus generated were then tested on a prospective cohort currently being recruited to obtain preliminary results as regards the robustness of those nomograms

Les immunothérapies ciblant les voies du récepteur de la mort cellulaire programmée 1 et de son ligand (anti-PD(L)1) se sont révélées être un traitement efficace pour de nombreux cancers. Le traitement par anti-PD(L)1 constitue un changement de paradigme en oncologie puisque son activité repose sur la restauration d'une réponse efficace des cellules T antitumorales. Deux raisons principales expliquent la nécessité d’identifier des biomarqueurs permettant de prédire la survie et l'efficacité anticancéreuse des anti-PD(L)1. Premièrement, un excès de décès a été observé dans le groupe expérimental d’essais de phase III randomisés comparant les immunothérapies anti-PD(L)1 à la chimiothérapie. Parmi les hypothèses controversées pouvant expliquer cette observation, le manque d'efficacité de l'anti-PD(L)1 chez les patients atteints d'une maladie à croissance rapide (appelés "progresseurs rapides") par rapport à un effet paradoxal de l'accélération de la maladie sous immunothérapie (dits "hyperprogresseurs") sont souvent mentionnés. Deuxièmement, les critères de réponse en imagerie jouent un rôle essentiel dans la prise en charge des patients cancéreux et définissent une "stratégie attentiste" pour les patients avec une maladie évolutive en imagerie. Le mécanisme d’action distinct des anti-PD(L)1, qui restaurent la capacité anti-tumorale du système immunitaire, conduisent à la survenue de profils de réponse non conventionnels tels que la pseudoprogression, l’hyperprogression, l’effet abscopal et les toxicités liées au système immunitaire. Nous avons tiré parti de l’apprentissage automatique pour confronter différents facteurs pronostiques / prédictifs et identifier les biomarqueurs d'imagerie associés à la mort prématurée sous immunothérapie anti-PD(L)1. Nous avons exploité des données transcriptomiques pour déterminer les voies biologiques liées à ces facteurs pronostiques / prédictifs. Nos résultats démontrent qu'un sous-ensemble limité de biomarqueurs d'imagerie peut prévoir la survie globale des patients. La classification de ces biomarqueurs d'imagerie en caractéristiques distinctives fournit une structure conceptuelle et une cohérence logique délimitant les interconnexions entre eux. Ces caractéristiques distinctives peuvent être comprises comme des circuits physiologiques distincts perturbées par le cancer et liés à une survie plus courte : organotropisme hépatique, charge tumorale élevée, hétérogénéité importante dans la vascularisation ou le métabolisme de la tumeur, infiltration le long des bordures de la tumeur, irrégularité de la forme tumorale, forte consommation de glucose, sarcopénie, et métabolisme élevé de la moelle osseuse. En utilisant l’apprentissage automatique, nous avons démontré que l’augmentation de la lactate déshydrogénase sérique et la présence de métastases hépatiques au scanner étaient deux facteurs indépendants de décès prématuré après l’initiation du traitement anti-PD(L)1. L'analyse transcriptomique a identifié des voies de signalisations susceptibles de donner lieu à de nouveaux traitements, et d'améliorer l'efficacité des anti-PD(L)1. Dans une perspective plus large, cela démontre la nécessité de continuer à développer une technologie d'imagerie de pointe pour améliorer la surveillance des patients atteints de cancer traités avec des immunothérapeutiques. Cela implique l'analyse et la liaison des données en pathologie, en oncologie, en radiologie et en médecine nucléaire, ainsi que la capacité de travailler avec de larges ensembles de données. Par conséquent, il est nécessaire de développer des programmes de radiomique pour développer des outils prédictifs utiles au diagnostic, à l'évaluation et à la gestion de tous les types de patients cancéreux. En conclusion, les approches de médecine de précision axées sur la radiomique pourraient améliorer la vie des patients cancéreux traités par immunothérapie anticancéreuse
Immunotherapies targeting the programmed cell death receptor-1 and ligand-1 pathways (anti-PD(L)1) have emerged as an effective treatment for a variety of cancers. Anti-PD(L)1 is a paradigm shift in the treatment of cancers since its activity relies on restoring an efficient anti-tumor T-cell response. Two main reasons explain the need to investigate biomarkers forecasting survival and predicting the anti-cancer efficacy of anti-PD(L)1. First, an excess of death has been observed in the experimental arm of randomized phase III trials comparing anti-PD(L)1 immunotherapies to chemotherapy for multiple cancers. Among the controversial hypotheses that would explain this observation are frequently mentioned the lack of effectiveness of anti-PD(L)1 in patients with a fast-growing disease (so-called "fast progressors") vs. a paradoxical effect of disease acceleration under immunotherapy (so-called "hyperprogressors"). Second, imaging response criteria play a pivotal role in guiding cancer patient management and define a "wait and see strategy" for patients treated with anti-PD(L)1 in monotherapy with progressive disease. The distinct mechanisms of anti-PD(L)1, which restore the immune system's anti-tumor capacity, leads to unconventional immune-related phenomena. From a medical imaging standpoint, it translates into pseudoprogression, hyperprogression, abscopal effect, and immune-related adverse events. We leveraged machine learning approaches to challenge the prognostic/predictive factors and identify which imaging biomarkers are associated with early death upon anti-PD(L)1 immunotherapy. We mined transcriptomic data to determine the biological pathways related to these prognostic/predictive factors. Our results demonstrate that a limited subset of imaging biomarkers can forecast overall survival. The classification of these imaging biomarkers into distinct hallmarks provides a conceptual structure and logical coherence delineating the interconnections between them. These hallmarks can be understood as distinct physiological circuits disrupted by cancer that are linked to shorter survival: liver organotropism, high tumor burden, high heterogeneity in tumor vascularity or metabolism, infiltration along tumor boundaries, irregularity in tumor shape, high glucose consumption, sarcopenia, and high bone marrow metabolism. Using machine-learning, we demonstrated that increased baseline serum lactate dehydrogenase and the presence of liver metastasis on CT-scan are two independent drivers of premature death after anti-PD(L)1 initiation. Transcriptomic analysis identified actionable pathways amenable to novel treatments, which could improve anti-PD(L)1 efficacy. From a broader perspective, this demonstrates the need to continue to develop advanced imaging technology to enhance the monitoring of cancer patients treated with immunotherapeutics. This involves analyzing and linking data in pathology, oncology, and radiology, and the ability to work with extensive datasets. Therefore, there is a need to develop comprehensive programs of radiomics for predictive tools that benefit diagnosis, assessment, and management of all types of cancer patients. In conclusion, radiomics driven precision medicine approaches could improve the lives of cancer patients treated with cancer immunotherapy

Glioblastoma is a highly malignant tumor which mostly recurs locally around the resected contrast enhancement. However, it is difficult to identify tumor invasiveness pre-surgically, especially in non-enhancing areas. Thus, the aim of this thesis was to utilize multimodal MR technique to identify and characterize the peritumoral progression zone that eventually leads to tumor progression. Patients with newly diagnosed cerebral glioblastoma were included consecutively from our cohort between 2010 and2014. The presurgical MRI sequences included volumetric T1-weighted with contrast, FLAIR, T2-weighted, diffusion-weighted imaging, diffusion tensor and perfusion MR imaging. Postsurgical and follow-up MRI included structural and ADC images. Image deformation, caused by disease nature and surgical procedure, renders routine coregistration methods inadequate for MRIs comparison between different time points. Therefore, a two-staged non-linear semi-automatic coregistration method was developed from the modification of the linear FLIRT and non-linear FNIRT functions in FMRIB’s Software Library (FSL). Utilising the above mentioned coregistration method, a volumetric study was conducted to analyse the extent of resection based on different MR techniques, including T1 weighted with contrast, FLAIR and DTI measures of isotropy (DTI-p) and anisotropy (DTI-q). The results showed that patients can have a better clinical outcome with a larger resection of the abnormal DTI q areas. Further study of the imaging characteristics of abnormal peritumoural DTI-q areas, using MRS and DCS-MRI, showed a higher Choline/NAA ratio (p = 0.035), especially higher Choline (p = 0.022), in these areas when compared to normal DTI-q areas. This was indicative of tumour activity in the peritumoural abnormal DTI-q areas. The peritumoural progression areas were found to have distinct imaging characteristics. In these progression areas, compared to non-progression areas within a 10 mm border around the contrast enhancing lesion, there was higher signal intensity in FLAIR (p = 0.02), and T1C (p < 0.001), and there were lower intensity in ADC (p = 0.029) and DTI-p (p < 0.001). Further applying radiomics features showed that 35 first order features and 77 second order features were significantly different between progression and non-progression areas. By using supervised convolutional neural network, there was an overall accuracy of 92.4% in the training set (n = 37) and 78.5% in the validation set (n=14). In summary, multimodal MR imaging, particularly diffusion tensor imaging, can demonstrate distinct characteristics in areas of potential progression on preoperative MRI, which can be considered potential targets for treatment. Further application of radiomics and machine learning can be potentially useful when identifying the tumor invasive margin before the surgery.

Early assessment of tumour response has lately acquired big interest in the medical field, given the possibility to modify treatments during their delivery. Radiomics aims to quantitatively describe images in radiology by automatically extracting a large number of image features. In this context, PET/CT (Positron Emission Tomography/Computed Tomography) images are of great interest since they encode functional and anatomical information, respectively. In order to assess the patients' responses from many image features appropriate methods should be applied. Machine learning offers different procedures that can deal with this, possibly high dimensional, problem. The main objective of this work was to develop a method to classify lung cancer patients as responding or not to chemoradiation treatment, relying on repeated PET/CT images. Patients were divided in two groups, based on the type of chemoradiation treatment they underwent (sequential or concurrent radiation therapy with respect to chemotherapy), but image features were extracted using the same procedure. Support vector machines performed classification using features from the Radiomics field, mostly describing tumour texture, or from handcrafted features, which described image intensity changes as a function of tumour depth. Classification performance was described by the area under the curve (AUC) of ROC (Receiving Operator Characteristic) curves after leave-one-out cross-validation. For sequential patients, 0.98 was the best AUC obtained, while for concurrent patients 0.93 was the best one. Handcrafted features were comparable to those from Radiomics and from previous studies, as for classification results. Also, features from PET alone and CT alone were found to be suitable for the task, entailing a performance better than random.

Gliomas are considered to be the most common primary adult malignant brain tumor. With the dramatic increases in computational power and improvements in image analysis algorithms, computer-aided medical image analysis has been introduced into clinical applications. Precision tumor grading and genotyping play an indispensable role in clinical diagnosis, treatment and prognosis. Gliomas diagnostic procedures include histopathological imaging tests, molecular imaging scans and tumor grading. Pathologic review of tumor morphology in histologic sections is the traditional method for cancer classification and grading, yet human study has limitations that can result in low reproducibility and inter-observer agreement. Compared with histopathological images, Magnetic resonance (MR) imaging present the different structure and functional features, which might serve as noninvasive surrogates for tumor genotypes. Therefore, computer-aided image analysis has been adopted in clinical application, which might partially overcome these shortcomings due to its capacity to quantitatively and reproducibly measure multilevel features on multi-parametric medical information. Imaging features obtained from a single modal image do not fully represent the disease, so quantitative imaging features, including morphological, structural, cellular and molecular level features, derived from multi-modality medical images should be integrated into computer-aided medical image analysis. The image quality differentiation between multi-modality images is a challenge in the field of computer-aided medical image analysis. In this thesis, we aim to integrate the quantitative imaging data obtained from multiple modalities into mathematical models of tumor prediction response to achieve additional insights into practical predictive value. Our major contributions in this thesis are: 1. Firstly, to resolve the imaging quality difference and observer-dependent in histological image diagnosis, we proposed an automated machine-learning brain tumor-grading platform to investigate contributions of multi-parameters from multimodal data including imaging parameters or features from Whole Slide Images (WSI) and the proliferation marker KI-67. For each WSI, we extract both visual parameters such as morphology parameters and sub-visual parameters including first-order and second-order features. A quantitative interpretable machine learning approach (Local Interpretable Model-Agnostic Explanations) was followed to measure the contribution of features for single case. Most grading systems based on machine learning models are considered “black boxes,” whereas with this system the clinically trusted reasoning could be revealed. The quantitative analysis and explanation may assist clinicians to better understand the disease and accordingly to choose optimal treatments for improving clinical outcomes. 2. Based on the automated brain tumor-grading platform we propose, multimodal Magnetic Resonance Images (MRIs) have been introduced in our research. A new imaging–tissue correlation based approach called RA-PA-Thomics was proposed to predict the IDH genotype. Inspired by the concept of image fusion, we integrate multimodal MRIs and the scans of histopathological images for indirect, fast, and cost saving IDH genotyping. The proposed model has been verified by multiple evaluation criteria for the integrated data set and compared to the results in the prior art. The experimental data set includes public data sets and image information from two hospitals. Experimental results indicate that the model provided improves the accuracy of glioma grading and genotyping.

Computed tomography (CT) imagery is an important weapon in the fight against lung cancer; various forms of lung cancer are routinely diagnosed from CT imagery. The growth of the suspect nodule is known to be a prognostic factor in the diagnosis of pulmonary cancer, but the change in other aspects of the nodule, such as its aspect ratio, density, spiculation, or other features usable for machine learning, may also provide prognostic information. We hypothesized that adding combined feature information from multiple CT image sets separated in time could provide a more accurate determination of nodule malignancy. To this end, we combined data from multiple CT images for individual patients taken from the National Lung Screening Trial. The resulting dataset was compared to equivalent datasets featuring single CT images for each patient. Feature reduction and normalization was performed as is standard. The highest accuracy achieved was 83.71% on a subset of features chosen by a combination of manual feature stability testing and the Correlation-based Feature Selection algorithm and classified by the Random Forests algorithm. The highest accuracy achieved with individual CT images was 81.00%, on a feature set consisting solely of the volume of the nodule in cubic centimeters.

Since the start of 2020 Sars-COVID19 has given rise to a world-wide pandemic. In an attempt to slow down the spreading of this disease various prevention and diagnostic methods have been developed. In this thesis the attention has been put on Machine Learning to predict prognosis based on data originating from radiological images. Radiomics has been used to extract information from images segmented using a software from the hospital which provided both the clinical data and images. The usefulness of different families of variables has then been evaluated through their performance in the methods used, i.e. Lasso regularized regression and Random Forest. The first chapter is introductory in nature, the second will contain a theoretical overview of the necessary concepts that will be needed throughout this whole work. The focus will be then shifted on methods and instruments used in the development of this thesis. The third chapter will report the results and finally some conclusions will be derived from the previously presented results. It will be concluded that the segmentation and feature extraction step is of pivotal importance in driving the performance of the predictions. In fact, in this thesis, it seems that the information from the images achieves the same predictive power that can be derived from the clinical data. This can be interpreted in three ways: first it can be taken as a symptom of the fact that even the more complex Sars-COVID19 cases can be segmented automatically, or semi-automatically by untrained personnel, leading to results competing with other methodologies. Secondly it can be taken to show that the performance of clinical variables can be reached by radiomic features alone in a semi-automatic pipeline, which could aid in reducing the workload imposed on medical professionals in case of pandemic. Finally it can be taken as proof that the method implemented has room to improve by more carefully investing in the segmentation phase

Positron Emission Tomography (PET) is an imaging modality that has become increasingly beneficial in Radiotherapy by improving treatment planning (1). PET reveals tumor volumes that are not well visualized on computed tomography CT or MRI, recognizes metastatic disease, and assesses radiotherapy treatment (1). It also reveals areas of the tumor that are more radiosensitive allowing for dose painting - a non-homogenous dose treatment across the tumor (1). However, PET is not without limitations. The quantitative unit of PET images, the Standardized Uptake Value (SUV), is affected by many factors such as reconstruction algorithm, patient weight, and tracer uptake time (2). In fact, PET is so sensitive that a patient imaged twice in a single day on the same machine and same protocol will produce different SUV values. The objective of this research was to increase the capabilities of PET by exploring other quantitative PET/CT measures for Radiotherapy treatment applications. The technique of quantitative image feature analysis, nowadays known as radiomics, was applied to PET and CT images. Image features were then extracted from PET/CT images and how the features differed between conventional and respiratory-gated PET/CT images in lung cancer was analyzed. The influence of noise on image features was analyzed by applying uncorrelated, Gaussian noise to PET/CT images and measuring how significantly noise affected features. Quantitative PET/CT measures outside of image feature analysis were also investigated. The correlation of esophageal metabolic tumor volumes (tumor volume demonstrating high metabolic uptake) and endoscopically implanted fiducial markers was studied. It was found that certain image features differed greatly between conventional and respiratory-gated PET/CT. The differences were mainly due to the effect of respiratory motion including affine motion, rotational motion and tumor deformation. Also, certain feature groups were more affected by noise than others. For instance, contour-dependent shape features exhibited the least change with noise. Comparatively, GLSZM features exhibited the greatest change with added noise. Discordance was discovered between the inferior and superior tumor fiducial markers and metabolic tumor volume (MTV). This demonstrated a need for both fiducial markers and MTV to provide a comprehensive view of a tumor. These studies called attention to the differences in features caused by factors such as motion, acquisition parameters, and noise, etc. Investigators should be aware of these effects. PET/CT radiomic features are indeed highly affected by noise and motion. For accurate clinical use, these effects must be account by investigators and future clinical users. Further investigation is warranted towards the standardization of PET/CT radiomic feature acquisition and clinical application.