To see the other types of publications on this topic, follow the link: Prediction of survival.
Dissertations / Theses on the topic 'Prediction of survival'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Prediction of survival.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Parast, Layla. "Landmark Prediction of Survival." Thesis, Harvard University, 2012. http://dissertations.umi.com/gsas.harvard:10085.
Full textAbstract:
The importance of developing personalized risk prediction estimates has become increasingly evident in recent years. In general, patient populations may be heterogenous and represent a mixture of different unknown subtypes of disease. When the source of this heterogeneity and resulting subtypes of disease are unknown, accurate prediction of survival may be difficult. However, in certain disease settings the onset time of an observable intermediate event may be highly associated with these unknown subtypes of disease and thus may be useful in predicting long term survival. Throughout this dissertation, we examine an approach to incorporate intermediate event information for the prediction of long term survival: the landmark model. In Chapter 1, we use the landmark modeling framework to develop procedures to assess how a patient’s long term survival trajectory may change over time given good intermediate outcome indications along with prognosis based on baseline markers. We propose time-varying accuracy measures to quantify the predictive performance of landmark prediction rules for residual life and provide resampling-based procedures to make inference about such accuracy measures. We illustrate our proposed procedures using a breast cancer dataset. In Chapter 2, we aim to incorporate intermediate event time information for the prediction of survival. We propose a fully non-parametric procedure to incorporate intermediate event information when only a single baseline discrete covariate is available for prediction. When a continuous covariate or multiple covariates are available, we propose to incorporate intermediate event time information using a flexible varying coefficient model. To evaluate the performance of the resulting landmark prediction rule and quantify the information gained by using the intermediate event, we use robust non-parametric procedures. We illustrate these procedures using a dataset of post-dialysis patients with end-stage renal disease. In Chapter 3, we consider improving efficiency by incorporating intermediate event information in a randomized clinical trial setting. We propose a semi-nonparametric two-stage procedure to estimate survival by incorporating intermediate event information observed before the landmark time. In addition, we present a testing procedure using these resulting estimates to test for a difference in survival between two treatment groups. We illustrate these proposed procedures using an AIDS dataset.
2
Jones, Margaret. "Point prediction in survival time models." Thesis, University of Newcastle Upon Tyne, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.340616.
Full text
3
Aparicio, Vázquez Ignacio. "Venn Prediction for Survival Analysis : Experimenting with Survival Data and Venn Predictors." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-278823.
Full textAbstract:
The goal of this work is to expand the knowledge on the field of Venn Prediction employed with Survival Data. Standard Venn Predictors have been used with Random Forests and binary classification tasks. However, they have not been utilised to predict events with Survival Data nor in combination with Random Survival Forests. With the help of a Data Transformation, the survival task is transformed into several binary classification tasks. One key aspect of Venn Prediction are the categories. The standard number of categories is two, one for each class to predict. In this work, the usage of ten categories is explored and the performance differences between two and ten categories are investigated. Seven data sets are evaluated, and their results presented with two and ten categories. For the Brier Score and Reliability Score metrics, two categories offered the best results, while Quality performed better employing ten categories. Occasionally, the models are too optimistic. Venn Predictors rectify this performance and produce well-calibrated probabilities.Målet med detta arbete är att utöka kunskapen om området för Venn Prediction som används med överlevnadsdata. Standard Venn Predictors har använts med slumpmässiga skogar och binära klassificeringsuppgifter. De har emellertid inte använts för att förutsäga händelser med överlevnadsdata eller i kombination med Random Survival Forests. Med hjälp av en datatransformation omvandlas överlevnadsprediktion till flera binära klassificeringsproblem. En viktig aspekt av Venn Prediction är kategorierna. Standardantalet kategorier är två, en för varje klass. I detta arbete undersöks användningen av tio kategorier och resultatskillnaderna mellan två och tio kategorier undersöks. Sju datamängder används i en utvärdering där resultaten presenteras för två och tio kategorier. För prestandamåtten Brier Score och Reliability Score gav två kategorier de bästa resultaten, medan för Quality presterade tio kategorier bättre. Ibland är modellerna för optimistiska. Venn Predictors korrigerar denna prestanda och producerar välkalibrerade sannolikheter.
4
Negassa, Abdissa. "Validation of tree-structured prediction for censored survival data." Thesis, McGill University, 1996. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=40407.
Full textAbstract:
Objectives. (i) to develop a computationally efficient algorithm of tree-growing for censored survival data, (ii) to assess the performance of two validation schemes, and (iii) to evaluate the performance of computationally inexpensive model selection criteria in relation to cross-validation.Background. In the tree-growing literature, a number of computationally inexpensive model selection criteria were suggested; however, none of them were systematically investigated for their performance. RECursive Partition and AMalgamation (RECPAM) is one of the existing tree-growing algorithms that provide such built-in model selection criteria. Application of RECPAM's different model selection criteria leads to a wide range of models (40). Since RECPAM is an exploratory data analysis tool, it is desirable to reduce its computational cost and establish the general properties of its model selection criteria so that clear guidelines can be suggested.
Methods. A computationally efficient tree-growing algorithm for prognostic classification and subgroup analysis is developed by employing the Cox score statistic and the Mantel-Haenszel estimator of the relative hazard. Two validation schemes, restricting validation to pruning and parameter estimation and validating the whole process of tree growing, are implemented and evaluated in simulation. Three model selection criteria--the elbow approach, minimum Akaike Information Criterion (AIC), and the one standard error (ISE) rule--were compared to cross-validation under a broad range of scenarios using simulation. Examples of medical data analyses are presented.
Conclusions. A gain in computational efficiency is achieved while obtaining the same result as the original RECPAM approach. The restricted validation scheme is computationally less expensive, however, it is biased. In the case of subgroup analysis, to adjust properly for influential prognostic factors, we suggest constructing a prognostic classification on such factors and using the resulting classification as strata in conducting the subgroup analysis. None of the model selection criteria studied exhibit a consistently superior performance over the range of scenarios considered here. Therefore, we propose a two-stage model selection strategy in which cross-validation is employed at the first step, and if according to this step there is evidence of structure in the data set, then the elbow rule is recommended in the second step.
5
Zhang, Haonan. "Machine Learning Approaches for Prediction of Kidney Transplant Survival." University of Toledo / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=toledo1555953011881185.
Full text
6
Spencer, David James. "Predicting early failure on probation using survival analysis and psychological predictor variables /." Digital version accessible at:, 2000. http://wwwlib.umi.com/cr/utexas/main.
Full text
7
Raoufi-Danner, Torrin. "Effects of Missing Values on Neural Network Survival Time Prediction." Thesis, Linköpings universitet, Statistik och maskininlärning, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-150339.
Full textAbstract:
Data sets with missing values are a pervasive problem within medical research. Building lifetime prediction models based solely upon complete-case data can bias the results, so imputation is preferred over listwise deletion. In this thesis, artificial neural networks (ANNs) are used as a prediction model on simulated data with which to compare various imputation approaches. The construction and optimization of ANNs is discussed in detail, and some guidelines are presented for activation functions, number of hidden layers and other tunable parameters. For the simulated data, binary lifetime prediction at five years was examined. The ANNs here performed best with tanh activation, binary cross-entropy loss with softmax output and three hidden layers of between 15 and 25 nodes. The imputation methods examined are random, mean, missing forest, multivariate imputation by chained equations (MICE), pooled MICE with imputed target and pooled MICE with non-imputed target. Random and mean imputation performed poorly compared to the others and were used as a baseline comparison case. The other algorithms all performed well up to 50% missingness. There were no statistical differences between these methods below 30% missingness, however missing forest had the best performance above this amount. It is therefore the recommendation of this thesis that the missing forest algorithm is used to impute missing data when constructing ANNs to predict breast cancer patient survival at the five-year mark.
8
Tian, Shaonan. "Essays on Corporate Default Prediction." University of Cincinnati / OhioLINK, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1352403546.
Full text
9
Orth, Walter [Verfasser]. "Multi-Period Credit Default Prediction : A Survival Analysis Approach / Walter Orth." Aachen : Shaker, 2012. http://d-nb.info/1066196826/34.
Full text
10
Kaponen, Martina. "Prediction of survival time of prostate cancer patients using Cox regression." Thesis, Uppsala universitet, Tillämpad matematik och statistik, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-354482.
Full text
11
Johnson, Terri Lynn. "Survival strategies of African-American women in community college /." Full text (PDF) from UMI/Dissertation Abstracts International, 2001. http://wwwlib.umi.com/cr/utexas/fullcit?p3008362.
Full text
12
Asfaw, Zeytu Gashaw. "Inference and Prediction in Non-stationary Stochastic Models: Survival Analysis and Kriging Interpolation." Doctoral thesis, Norges teknisk-naturvitenskapelige universitet, Institutt for matematiske fag, 2014. http://urn.kb.se/resolve?urn=urn:nbn:no:ntnu:diva-25982.
Full text
13
Krenzien, Felix, Ivan Matia, Georg Wiltberger, Hans-Michael Hau, Moritz Schmelzle, Sven Jonas, Udo X. Kaisers, and Peter T. Fellmer. "Early prediction of survival after open surgical repair of ruptured abdominal aortic aneurysms." Universitätsbibliothek Leipzig, 2014. http://nbn-resolving.de/urn:nbn:de:bsz:15-qucosa-156960.
Full textAbstract:
Background: Scoring models are widely established in the intensive care unit (ICU). However, the importance in patients with ruptured abdominal aortic aneurysm (RAAA) remains unclear. Our aim was to analyze scoring systems as predictors of survival in patients undergoing open surgical repair (OSR) for RAAA. Methods: This is a retrospective study in critically ill patients in a surgical ICU at a university hospital. Sixty-eight patients with RAAA were treated between February 2005 and June 2013. Serial measurements of Sequential Organ Failure Assessment score (SOFA), Simplified Acute Physiology Score II (SAPS II) and Simplified Therapeutic Intervention
Scoring System-28 (TISS-28) were evaluated with respect to in-hospital mortality. Eleven patients had to be excluded from this study because 6 underwent endovascular repair and 5 died before they could be admitted to the ICU. Results: All patients underwent OSR. The initial, highest, and mean of SOFA and SAPS II scores correlated significant
with in-hospital mortality. In contrast, TISS-28 was inferior and showed a smaller area under the receiver operating curve. The cut-off point for SOFA showed the best performance in terms of sensitivity and specificity. An initial SOFA score below 9 predicted an in-hospital mortality of 16.2% (95% CI, 4.3–28.1) and a score above 9 predicted an
in-hospital mortality of 73.7% (95% CI, 53.8–93.5, p < 0.01). Trend analysis showed the largest effect on SAPS II. When the score increased or was unchanged within the first 48 h (score >45), the in-hospital mortality rate was 85.7% (95% CI, 67.4–100, p < 0.01) versus 31.6% (95% CI, 10.7–52.5, p = 0.01) when it decreased. On multiple regression analysis, only the mean of the SOFA score showed a significant predictive capacity with regards to mortality (odds ratio
1.77; 95% CI, 1.19–2.64; p < 0.01). Conclusion: SOFA and SAPS II scores were able to predict in-hospital mortality in RAAA within 48 h after OSR. According to cut-off points, an increase or decrease in SOFA and SAPS II scores improved sensitivity and specificity.
14
van, Miltenburg Jelle. "Conformal survival predictions at a user-controlled time point : The introduction of time point specialized Conformal Random Survival Forests." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-232060.
Full textAbstract:
The goal of this research is to expand the field of conformal predictions using Random Survival Forests. The standard Conformal Random Survival Forest can predict with a fixed certainty whether something will survive up until a certain time point. This research is the first to show that there is little practical use in the standard Conformal Random Survival Forest algorithm. It turns out that the confidence guarantees of the conformal prediction framework are violated if the Standard algorithm makes predictions for a user-controlled fixed time point. To solve this challenge, this thesis proposes two algorithms that specialize in conformal predictions for a fixed point in time: a Fixed Time algorithm and a Hybrid algorithm. Both algorithms transform the survival data that is used by the split evaluation metric in the Random Survival Forest algorithm. The algorithms are evaluated and compared along six different set prediction evaluation criteria. The prediction performance of the Hybrid algorithm outperforms the prediction performance of the Fixed Time algorithm in most cases. Furthermore, the Hybrid algorithm is more stable than the Fixed Time algorithm when the predicting job extends to various time points. The hybrid Conformal Random Survival Forest should thus be considered by anyone who wants to make conformal survival predictions at usercontrolled time points.Målet med denna avhandling är att utöka området för konformitetsprediktion med hjälp av Random Survival Forests. Standardutförandet av Conformal Random Survival Forest kan förutsäga med en viss säkerhet om någonting kommer att överleva fram till en viss tidpunkt. Denna avhandling är den första som visar att det finns liten praktisk användning i standardutförandet av Conformal Random Survival Forest-algoritmen. Det visar sig att konfidensgarantierna för konformitetsprediktionsramverket bryts om standardalgoritmen gör förutsägelser för en användarstyrd fast tidpunkt. För att lösa denna utmaning, föreslår denna avhandling två algoritmer som specialiserar sig i konformitetsprediktion för en bestämd tidpunkt: en fast-tids algoritm och en hybridalgoritm. Båda algoritmerna omvandlar den överlevnadsdata som används av den delade utvärderingsmetoden i Random Survival Forest-algoritmen. Uppskattningsförmågan för hybridalgoritmen överträffar den för fast-tids algoritmen i de flesta fall. Dessutom är hybrid algoritmen stabilare än fast-tids algoritmen när det förutsägelsejobbet sträcker sig till olika tidpunkter. Hybridalgoritmen för Conformal Random Survival Forest bör därför föredras av den som vill göra konformitetsprediktion av överlevnad vid användarstyrda tidpunkter.
15
Noe, Montes Garcia, and Montes Garcia Noe. "Epidemiological aspects of Claviceps africana, causal agent of Sorghum ergot." Texas A&M University, 2004. http://hdl.handle.net/1969.1/1546.
Full textAbstract:
Sorghum ergot, caused by Claviceps africana Frederickson, Mantle & de Milliano, is a disease that affects non-fertilized ovaries in sorghum male-sterile plants and infects hybrids if there is pollen sterility at flowering time. Sphacelia containing macroconidia could play a role in the survival of the pathogen. This study developed risk assessment models and evaluated environmental conditions affecting viability of macroconidia and transition from sphacelial to sclerotial tissues. Effect of weather on ergot severity was evaluated under natural conditions (in monthly planting dates) in nine sorghum genotypes at College Station, Weslaco, Rio Bravo, and Celaya. Panicles were inoculated daily beginning at flower initiation with a suspension of 1.6 x 106 C. africana conidia ml-1. Weather triad values were used to identify weather parameters correlated with the disease. Ergot severity was statistically greater in A-lines than hybrids because of the possible interference of pollen on some dates. Celaya had the greatest amount of ergot in hybrids. A-line ATx2752 had the lowest average ergot severity throughout years, locations and planting dates, as did the hybrid NC+8R18. Maximum and minimum temperature had a negative correlation with ergot at Rio Bravo, College
Station and Weslaco, while at Celaya it was positive. The highest correlation was 7 to 9 days before initiation of flowering, suggesting that cooler temperatures during this period could cause male sterility. A-lines showed the same relationships between ergot and maximum and minimum temperatures after initiation of flowering. Minimum relative humidity had a positive correlation with ergot after initiation of flowering in both sorghum plant types. Sphacelia stored under cool temperatures (-3oC to 7oC) maintained conidial viability, and newly-formed sphacelia located on the sphacelia surface had the highest conidial viability. However, they show a greater viability reduction through time compared with conidia from older sphacelia, showing that conidial maturity can play a role in the survival of the conidia. Sphacelia on plants grown at 10oC, 20oC and 30oC with low relative humidity did not had any sclerotial development up to 4 weeks after formation of sphacelia. However, higher temperatures promoted an increase in the sphacelia dry weight during that time.
16
Ripley, Ruth Mary. "Neural network models for breast cancer prognosis." Thesis, University of Oxford, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.244721.
Full text
17
Frandsen, Abraham Jacob. "Machine Learning for Disease Prediction." BYU ScholarsArchive, 2016. https://scholarsarchive.byu.edu/etd/5975.
Full textAbstract:
Millions of people in the United States alone suffer from undiagnosed or late-diagnosed chronic diseases such as Chronic Kidney Disease and Type II Diabetes. Catching these diseases earlier facilitates preventive healthcare interventions, which in turn can lead to tremendous cost savings and improved health outcomes. We develop algorithms for predicting disease occurrence by drawing from ideas and techniques in the field of machine learning. We explore standard classification methods such as logistic regression and random forest, as well as more sophisticated sequence models, including recurrent neural networks. We focus especially on the use of medical code data for disease prediction, and explore different ways for representing such data in our prediction algorithms.
18
Walop, Wilhelmina. "The use of biomarkers in the prediction of survival in patients with bronchogenic carcinoma /." Thesis, McGill University, 1986. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=72797.
Full text
19
Robinson, David. "Prediction of survival in prostate cancer : aspects on localised, locally advanced and metastatic disease." Doctoral thesis, Linköping : Department of Clinical and Experimental Medicine, Linköping University, 2008. http://www.bibl.liu.se/liupubl/disp/disp2008/med1073s.pdf.
Full text
20
Yu, Jianxiong. "Pavement Service Life Estimation And Condition Prediction." See Full Text at OhioLINK ETD Center (Requires Adobe Acrobat Reader for viewing), 2005. http://www.ohiolink.edu/etd/view.cgi?toledo1132896646.
Full textAbstract:
Dissertation (Ph.D.)--University of Toledo, 2005.Typescript. "A dissertation [submitted] as partial fulfillment of the requirements of the Doctor of Philosophy degree in Engineering." Bibliography: leaves 69-74.
21
Begum, Mubeena. "Gene expression profiles and clinical parameters for survival prediction in stage II and III colorectal cancer." [Tampa, Fla] : University of South Florida, 2006. http://purl.fcla.edu/usf/dc/et/SFE0001554.
Full text
22
Orth, Walter Verfasser], Karl C. [Akademischer Betreuer] [Mosler, and Friedrich [Akademischer Betreuer] Schmid. "Multi-Period Credit Default Prediction - A Survival Analysis Approach / Walter Orth. Gutachter: Karl Mosler ; Friedrich Schmid." Köln : Universitäts- und Stadtbibliothek Köln, 2012. http://d-nb.info/1038360536/34.
Full text
23
Kamath, Vidya. "Use of Random Subspace Ensembles on Gene Expression Profiles in Survival Prediction for Colon Cancer Patients." Scholar Commons, 2005. https://scholarcommons.usf.edu/etd/715.
Full textAbstract:
Cancer is a disease process that emerges out of a series of genetic mutations that cause seemingly uncontrolled multiplication of cells. The molecular genetics of cells indicates that different combinations of genetic events or alternative pathways in cells may lead to cancer. A study of the gene expressions of cancer cells, in combination with the external influential factors, can greatly aid in cancer management such as understanding the initiation and etiology of cancer, as well as detection, assessment and prediction of the progression of cancer.
Gene expression analysis of cells yields a very large number of features that can be used to describe the condition of the cell. Feature selection methods are explored to choose the best of these features that are most relevant to the problem at hand. Random subspace ensembles created using these selected features perform poorly in predicting the 36-month survival for colon cancer patients. A modification to the random subspace scheme is proposed to enhance the accuracy of prediction. The method first applies random subspace ensembles with decision trees to select predictive features. Then, support vector machines are used to analyze the selected gene expression profiles in cancer tissue to predict the survival outcome for a patient.
The proposed method is shown to achieve a weighted accuracy of 58.96%, with 40.54% sensitivity and 77.38% specificity in predicting 36-month survival for new and unknown colon cancer patients. The prediction accuracy of the method is comparable to the baseline classifiers and significantly better than random subspace ensembles on gene expression profiles of colon cancer.
24
Hodozsán, Tamás. "Timed Recidivism. In search for critical periods to supplement interventions." Thesis, Malmö universitet, Fakulteten för hälsa och samhälle (HS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-25866.
Full textAbstract:
Assessing risk had always been the key focus when it comes to recidivism. Using risk assessment instruments, it is possible to predict the outcome of recidivism dichotomously. These measures, however, can only predict between 70-80 percent of validity, and they specify only levels of risk (low-medium-high), but not time. Therefore, the aim of this study is to define time of recidivism to supplement risk assessment with a possible new actuarial approach and fill out gaps in the existing literature. To do so a systematic literature review was conducted with a controlled search on exact time points. All the fourteen studies resulted in the final model were: published in the past 20 years, had some connection to time and were quantitative. The results highlighted the importance of the first year, especially the first half of the year as the most critical period regarding recidivism. Three different time periods were identified: (1) the end of the first month; (2) between the second and the third; (3) at the end of the 6th. Consequently, it might be beneficial to target these critical periods with more intense supervision/intervention in order to decrease the likelihood of recidivism.
25
Martinenko, Evgeny. "Prediction of survival of early stages lung cancer patients based on ER beta cellular expressions and epidemiological data." Master's thesis, University of Central Florida, 2011. http://digital.library.ucf.edu/cdm/ref/collection/ETD/id/4796.
Full textAbstract:
We attempted a mathematical model for expected prognosis of lung cancer patients based on a multivariate analysis of the values of ER-interacting proteins (ERbeta) and a membrane bound, glycosylated phosphoprotein MUC1), and patients clinical data recorded at the time of initial surgery. We demonstrate that, even with the limited sample size available to use, combination of clinical and biochemical data (in particular, associated with ERbeta and MUC1) allows to predict survival of lung cancer patients with about 80% accuracy while prediction on the basis of clinical data only gives about 70% accuracy. The present work can be viewed as a pilot study on the subject: since results confirm that ER-interacting proteins indeed inuence lung cancer patients' survival, more data is currently being collected.ID: 030646185; System requirements: World Wide Web browser and PDF reader.; Mode of access: World Wide Web.; Thesis (M.S.)--University of Central Florida, 2011.; Includes bibliographical references (p. 32-33).
M.S.
Masters
Mathematics
Sciences
Mathematical Science
26
Kakino, Ryo. "Quantitative image analysis for prognostic prediction in lung SBRT." Doctoral thesis, Kyoto University, 2021. http://hdl.handle.net/2433/263582.
Full text
27
Rodrigo, Hansapani Sarasepa. "Bayesian Artificial Neural Networks in Health and Cybersecurity." Scholar Commons, 2017. http://scholarcommons.usf.edu/etd/6940.
Full textAbstract:
Being in the era of Big data, the applicability and importance of data-driven models like artificial neural network (ANN) in the modern statistics have increased substantially. In this dissertation, our main goal is to contribute to the development and the expansion of these ANN models by incorporating Bayesian learning techniques. We have demonstrated the applicability of these Bayesian ANN models in interdisciplinary research including health and cybersecurity.
Breast cancer is one of the leading causes of deaths among females. Early and accurate diagnosis is a critical component which decides the survival of the patients. Including the well known ``Gail Model", numerous efforts are being made to quantify the risk of diagnosing malignant breast cancer. However, these models impose some limitations on their use of risk prediction. In this dissertation, we have developed a diagnosis model using ANN to identify the potential breast cancer patients with their demographic factors and the previous mammogram results. While developing the model, we applied the Bayesian regularization techniques (evidence procedure), along with the automatic relevance determination (ARD) prior, to minimize the network over-fitting. The optimal Bayesian network has 81\% overall accuracy in correctly classifying the actual status of breast cancer patients, 59\% sensitivity in accurately detecting the malignancy and 83\% specificity in correctly detecting non-malignancy. The area under the receiver operating characteristic curve (0.7940) shows that this is a moderate classification model.
We then present a new Bayesian ANN model for developing a nonlinear Poisson regression model which can be used for count data modeling. Here, we have summarized all the important steps involved in developing the ANN model, including the forward-propagation, backward-propagation and the error gradient calculations of the newly developed network. As a part of this, we have introduced a new activation function into the output layer of the ANN and error minimizing criterion, using count data. Moreover, we have expanded our model to incorporate the Bayesian learning techniques. The performance our model is tested using simulation data.
In addition to that, a piecewise constant hazard model is developed by extending the above nonlinear Poisson regression model under the Bayesian setting. This model can be utilized over the other conventional methods for accurate survival time prediction. With this, we were able to significantly improve the prediction accuracies. We captured the uncertainties of our predictions by incorporating the error bars which could not achieve with a linear Poisson model due to the overdispersion in the data. We also have proposed a new hybrid learning technique, and we evaluated the performance of those techniques with a varying number of hidden nodes and data size.
Finally, we demonstrate the suitability of Bayesian ANN models for time series forecasting by using an online training algorithm. We have developed a vulnerability forecast model for the Linux operating system by using this approach.
28
Kamath, Vidya P. "Enhancing Gene Expression Signatures in Cancer Prediction Models: Understanding and Managing Classification Complexity." Scholar Commons, 2010. http://scholarcommons.usf.edu/etd/3653.
Full textAbstract:
Cancer can develop through a series of genetic events in combination with
external influential factors that alter the progression of the disease. Gene expression
studies are designed to provide an enhanced understanding of the progression of cancer
and to develop clinically relevant biomarkers of disease, prognosis and response to
treatment. One of the main aims of microarray gene expression analyses is to develop
signatures that are highly predictive of specific biological states, such as the molecular
stage of cancer. This dissertation analyzes the classification complexity inherent in gene
expression studies, proposing both techniques for measuring complexity and algorithms
for reducing this complexity.
Classifier algorithms that generate predictive signatures of cancer models must
generalize to independent datasets for successful translation to clinical practice. The
predictive performance of classifier models is shown to be dependent on the inherent
complexity of the gene expression data. Three specific quantitative measures of
classification complexity are proposed and one measure ( f) is shown to correlate highly
(R 2=0.82) with classifier accuracy in experimental data.
Three quantization methods are proposed to enhance contrast in gene expression
data and reduce classification complexity. The accuracy for cancer prognosis prediction
is shown to improve using quantization in two datasets studied: from 67% to 90% in lung
cancer and from 56% to 68% in colorectal cancer. A corresponding reduction in
classification complexity is also observed.
A random subspace based multivariable feature selection approach using costsensitive
analysis is proposed to model the underlying heterogeneous cancer biology and
address complexity due to multiple molecular pathways and unbalanced distribution of
samples into classes. The technique is shown to be more accurate than the univariate ttest
method. The classifier accuracy improves from 56% to 68% for colorectal cancer
prognosis prediction.
A published gene expression signature to predict radiosensitivity of tumor cells is
augmented with clinical indicators to enhance modeling of the data and represent the
underlying biology more closely. Statistical tests and experiments indicate that the
improvement in the model fit is a result of modeling the underlying biology rather than
statistical over-fitting of the data, thereby accommodating classification complexity
through the use of additional variables.
29
Shay, Keegan P. "Evaluating the use of neighborhoods for query dependent estimation of survival prognosis for oropharyngeal cancer patients." Thesis, University of Iowa, 2019. https://ir.uiowa.edu/etd/6854.
Full textAbstract:
Oropharyngeal Cancer diagnoses make up three percent of all cancer diagnoses in the United States per year. Recently, there has been an increase in the incidence of HPV-associated oropharyngeal cancer, necessitating updates to prior survival estimation techniques, in order to properly account for this shift in demographic. Clinicians depend on accurate survival prognosis estimates in order to create successful treatment plans that aim to maximize patient life while minimizing adverse treatment side effects. Additionally, recent advances in data analysis have resulted in richer and more complex data, motivating the use of more advanced data analysis techniques. Incorporation of sophisticated survival analysis techniques can leverage complex data, from a variety of sources, resulting in improved personalized prediction. Current survival prognosis prediction methods often rely on summary statistics and underlying assumptions regarding distribution or overall risk.
We propose a k-nearest neighbor influenced approach for predicting oropharyngeal survival outcomes. We evaluate our approach for overall survival (OS), recurrence-free survival (RFS), and recurrence-free overall survival (RF+OS). We define two distance functions, not subject to the curse of dimensionality, in order to reconcile heterogeneous features with patient-to-patient similarity scores to produce a meaningful overall measure of distance. Using these distance functions, we obtain the k-nearest neighbors for each patient, forming neighborhoods of similar patients. We leverage these neighborhoods for prediction in two novel ensemble methods. The first ensemble method uses the nearest neighbors for each patient to combine globally trained predictions, weighted by their accuracies within a selected neighborhood. The second ensemble method combines Kaplan-Meier predictions from a variety of neighborhoods. Both proposed methods outperform an ensemble of standard global survival predictive models, with statistically significant calibration.
30
Kamath, Vidya. "Use of random subspace ensembles on gene expression profiles to enhance the accuracy of survival prediction for colon cancer patients." [Tampa, Fla.] : University of South Florida, 2005. http://purl.fcla.edu/fcla/etd/SFE0001408.
Full text
31
Santos, Tiago Mendonça dos. "Avaliação do desempenho de modelos preditivos no contexto de análise de sobrevivência." Universidade de São Paulo, 2013. http://www.teses.usp.br/teses/disponiveis/45/45133/tde-03092013-111337/.
Full textAbstract:
Modelos estatísticos com objetivos preditivos são frequentemente aplicados como ferramentas no processo de tomadas de decisão em diversas áreas. Uma classe importante de modelos estatísticos é composta por modelos de análise de sobrevivência. Duas quantidades são de interesse nessa classe: o tempo até o instante do evento de interesse ou o status para um determinado instante de tempo fixado. Aplicações importantes desses modelos incluem a identificação de novos marcadores para certas doenças e definição de qual terapia será mais adequada de acordo com o paciente. Os marcadores utilizados podem ser dados por biomarcadores, assim como por marcadores baseados em modelos de regressão. Um exemplo de marcador baseado em modelos de regressão é dado pelo preditor linear. Ainda que a utilização de modelos de sobrevivência com objetivos preditivos seja de suma importância, a literatura nesse assunto é muito esparsa e não há consenso na forma de se avaliar o desempenho preditivo desses. Esse trabalho pretende reunir e comparar diferentes abordagens de se avaliar o desempenho preditivo de modelos de sobrevivência. Essa avaliação é feita principalmente utilizando-se funções de perda para o tempo de sobrevivência e quantidades associadas a diferentes definições de curva ROC para o status. Para a comparação dessas diferentes metodologias foi feito um estudo de simulação e no final aplicou-se essas técnicas em um conjunto de dados de um estudo do Instituto do Câncer de São Paulo.In many fields, predictive models are often applied as a helpful tool in the decision making process. An important class of predictive models is composed by survival models. Two quantities of special interest in these class are: time until the occurrence of a specified event and survival status for a fixed moment of time. Important applications of these models include new markers identification for certain diseases, as well as defining which therapy is the most appropriated for a patient. Markers can be given by biomarkers, but they can also be derived from regression models. An example of regression models based markers is the linear predictor. Despite the importance of survival models applications with predictive goals, literature is this subject is very sparse and there is no agreement on the best methodology to evaluate predictive performance of these models. In this work we intend to assemble and to compare different methodologies for assessing the predictive performance of survival models. This assessment is made mainly with loss functions for the survival time and ROC curve associated quantities for status. An simulation study was done in order to compare these different methodologies, which were also applied to a study about survival of patients at ICU of ICESP (Instituto do Câncer de São Paulo)
32
Abdul, Jalil Walid, and Torre Kvin Dalla. "The impact of missing data imputation on HCC survival prediction : Exploring the combination of missing data imputation with data-level methods such as clustering and oversampling." Thesis, KTH, Skolan för teknikvetenskap (SCI), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-230741.
Full textAbstract:
The area of data imputation, which is the process of replacing missing data with substituted values, has been covered quite extensively in recent years. The literature on the practical impact of data imputation however, remains scarce. This thesis explores the impact of some of the state of the art data imputation methods on HCC survival prediction and classification in combination with data-level methods such as oversampling. More specifically, it explores imputation methods for mixed-type datasets and their impact on a particular HCC dataset. Previous research has shown that, the newer, more sophisticated imputation methods outperform simpler ones when evaluated with normalized root mean square error (NRMSE). Contrary to intuition however, the results of this study show that when combined with other data-level methods such as clustering and oversampling, the differences in imputation performance does not always impact classification in any meaningful way. This might be explained by the noise that is introduced when generating synthetic data points in the oversampling process. The results also show that one of the more sophisticated imputation methods, namely MICE, is highly dependent on prior assumptions about the underlying distributions of the dataset. When those assumptions are incorrect, the imputation method performs poorly and has a considerable negative impact on classification.
33
Olešová, Kristína. "Klasifikace stupně gliomů v MR datech mozku." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2020. http://www.nusl.cz/ntk/nusl-413113.
Full textAbstract:
This thesis deals with a classification of glioma grade in high and low aggressive tumours and overall survival prediction based on magnetic resonance imaging. Data used in this work is from BRATS challenge 2019 and each set contains information from 4 weighting sequences of MRI. Thesis is implemented in PYTHON programming language and Jupyter Notebooks environment. Software PyRadiomics is used for calculation of image features. Goal of this work is to determine best tumour region and weighting sequence for calculation of image features and consequently select set of features that are the best ones for classification of tumour grade and survival prediction. Part of thesis is dedicated to survival prediction using set of statistical tests, specifically Cox regression
34
Spagnoli, Lorenzo. "COVID-19 prognosis estimation from CAT scan radiomics: comparison of different machine learning approaches for predicting patients survival and ICU Admission." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/23926/.
Full textAbstract:
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
35
Dalla, Torre Kevin, and Jalil Walid Abdul. "The impact of missing data imputation on HCC survival prediction : Exploring the combination of missing data imputation with data-level methods such as clustering and oversampling." Thesis, KTH, Skolan för teknikvetenskap (SCI), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-231023.
Full textAbstract:
The area of data imputation, which is the process of replacing missing data with substituted values, has been covered quite extensively in recent years. The literature on the practical impact of data imputation however, remains scarce. This thesis explores the impact of some of the state of the art data imputation methods on HCC survival prediction and classification in combination with data-level methods such as oversampling. More specifically, it explores imputation methods for mixed-type datasets and their impact on a particular HCC dataset. Previous research has shown that, the newer, more sophisticated imputation methods outperform simpler ones when evaluated with normalized root mean square error (NRMSE). Contrary to intuition however, the results of this study show that when combined with other data-level methods such as clustering and oversampling, the differences in imputation performance does not always impact classification in any meaningful way. This might be explained by the noise that is introduced when generating synthetic data points in the oversampling process. The results also show that one of the more sophisticated imputation methods, namely MICE, is highly dependent on prior assumptions about the underlying distributions of the dataset. When those assumptions are incorrect, the imputation method performs poorly and has a considerable negative impact on classification.Forskningen kring data imputation, processen där man ersätter saknade data med substituerade värden, har varit omfattande de senaste åren. Litteraturen om den praktiska inverkan som data imputation metoder har på klassificering är dock otillräcklig. Det här kandidatexamensarbetet utforskar den inverkan som de nyare imputation metoderna har på HCC överlevnads klassificering i kombination med andra data-nivå metoder så som översampling. Mer specifikt, så utforskar denna studie imputations metoder för heterogena dataset och deras inverkan på ett specifikt HCC dataset. Tidigare forskning har visat att de nyare, mer sofistikerade imputations metoderna presterar bättre än de mer enkla metoderna när de utvärderas med normalized root mean square error (NRMSE). I motsats till intuition, så visar resultaten i denna studie att när imputation kombineras med andra data-nivå metoder så som översampling och klustring, så påverkas inte klassificeringen alltid på ett meningsfullt sätt. Detta kan förklaras med att brus introduceras i datasetet när syntetiska punkter genereras i översampling processen. Resultaten visar också att en av de mer sofistikerade imputation metoderna, nämligen MICE, är starkt beroende på tidigare antaganden som görs om de underliggande fördelningarna i datasetet. När dessa antaganden är inkorrekta så presterar imputations metoden dåligt och har en negativ inverkan på klassificering.
36
Mohammadisohrabi, Ali. "Design and implementation of a Recurrent Neural Network for Remaining Useful Life prediction." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2020.
Find full textAbstract:
A key idea underlying many Predictive Maintenance solutions is Remaining Useful Life (RUL) of machine parts, and it simply involves a prediction on the time remaining before a machine part is likely to require repair or replacement. Nowadays, with respect to fact that the systems are getting more complex, the innovative Machine Learning and Deep Learning algorithms can be deployed to study the more sophisticated correlations in complex systems. The exponential increase in both data accumulation and processing power make the Deep Learning algorithms more desirable that before. In this paper a Long Short-Term Memory (LSTM) which is a Recurrent Neural Network is designed to predict the Remaining Useful Life (RUL) of Turbofan Engines. The dataset is taken from NASA data repository. Finally, the performance obtained by RNN is compared to the best Machine Learning algorithm for the dataset.
37
Lu, Pascal. "Statistical Learning from Multimodal Genetic and Neuroimaging data for prediction of Alzheimer's Disease." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS636.
Full textAbstract:
De nos jours, la maladie d'Alzheimer est la principale cause de démence. Elle provoque des troubles de mémoires et de comportements chez les personnes âgées. La diagnostic précoce de la maladie d'Alzheimer est un sujet actif de recherche. Trois différents types de données jouent un role particulier dans le diagnostic de la maladie d'Alzheimer: les tests cliniques, les données de neuroimagerie et les données génétiques. Les deux premières modalités apportent de l'information concernant l'état actuel du patient. En revanche, les données génétiques permettent d'identifier si un patient est à risque et pourrait développer la maladie d'Alzheimer dans le futur. Par ailleurs, durant la dernière décennie, les chercheurs ont crée des bases de données longitudinales sur la maladie d'Alzheimer et d'importantes recherches ont été réalisées pour le traitement et l'analyse de données complexes en grande dimension. La première contribution de cette thèse sera d'étudier comment combiner différentes modalités dans le but d'améliorer leur pouvoir prédictif dans le contexte de la classification. Nous explorons les modèles multiniveaux permettant de capturer les potentielles interactions entre modalités. Par ailleurs, nous modéliserons la structure de chaque modalité (structure génétique, structure spatiale du cerveau) à travers l'utilisation de pénalités adaptées comme la pénalité ridge pour les images, ou la pénalité group lasso pour les données génétiques. La deuxième contribution de thèse sera d'explorer les modèles permettant de prédire la date de conversion à la maladie d'Alzheimer pour les patients atteints de troubles cognitifs légers. De telles problématiques ont été mises en valeurs à travers de challenge, comme TADPOLE. Nous utiliserons principalement le cadre défini par les modèles de survie. Partant de modèles classiques, comme le modèle d'hasard proportionnel de Cox, du modèle additif d'Aalen, et du modèle log-logistique, nous allons développer d'autres modèles de survie pour la combinaisons de modalités, à travers un modèle log-logistique multiniveau ou un modèle de Cox multiniveauAlzheimer's Disease (AD) is nowadays the main cause of dementia in the world. It provokes memory and behavioural troubles in elderly people. The early diagnosis of Alzheimer's Disease is an active topic of research. Three different types of data play a major role when it comes to its diagnosis: clinical tests, neuroimaging and genetics. The two first data bring informations concerning the patient's current state. On the contrary, genetic data help to identify whether a patient could develop AD in the future. Furthermore, during the past decade, researchers have created longitudinal dataset on A and important advances for processing and analyse of complex and high-dimensional data have been made. The first contribution of this thesis will be to study how to combine different modalities in order to increase their predictive power in the context of classification. We will focus on hierarchical models that capture potential interactions between modalities. Moreover, we will adequately modelled the structure of each modality (genomic structure, spatial structure for brain images), through the use of adapted penalties such as the ridge penalty for images and the group lasso penalty for genetic data. The second contribution of this thesis will be to explore models for predict the conversion date to Alzheimer's Disease for mild cognitive impairment subjects. Such problematic has been enhanced by the TADPOLE challenge. We will use the framework provided by survival analysis. Starting from basic models such as the Cox proportional hasard model, the additive Aalen model, and the log-logistic model, we will develop other survival models for combining different modalities, such as a multilevel log-logistic model or a multilevel Cox model
38
Minya, Kristoffer. "Överlevnadsanalys i tjänsteverksamhet : Tidspåverkan i överklagandeprocessen på Migrationsverket." Thesis, Linköpings universitet, Statistik, 2014. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-110428.
Full textAbstract:
Migrationsverket är en myndighet som prövar ansökningar från personer som vill söka skydd, ha medborgarskap, studera eller vill jobba i Sverige. Då det på senare tid varit en stor ökning i dessa ansökningar har tiden för vilket ett beslut tar ökat. Varje typ av ansökning (exempelvis medborgarskap) är en process som består av flera steg. Hur beslutet går igenom dessa steg kallas för flöde. Migrationsverket vill därför öka sin flödeseffektivitet. När beslutet är klart och personen tagit del av det men inte är nöjd kan denne överklaga. Detta är en av de mest komplexa processerna på Migrationsverket. Syftet är analysera hur lång tid denna process tar och vilka steg i processen som påverkar tiden. Ett steg (som senare visar sig ha en stor effekt på tiden) är yttranden. Det är när domstolen begär information om vad personen som överklagar har att säga om varför denne överklagar. För att analysera detta var två metoder relevanta, accelerated failure time (AFT) och \multi-state models (MSM). Den ena kan predicera tid till händelse (AFT) medan den andra kan analysera effekten av tidspåverkan (MSM) i stegen. Yttranden tidigt i processen har stor betydelse för hur snabbt en överklagan får en dom samtidigt som att antal yttranden ökar tiden enormt. Det finns andra faktorer som påverkar tiden men inte i så stor grad som yttranden. Då yttranden tidigt i processen samtidigt som antal yttranden har betydelse kan flödeseffektiviteten ökas med att ta tid på sig att skriva ett informativt yttrande som gör att domstolen inte behöver begära flera yttranden.The Swedish Migration Board is an agency that review applications from individuals who wish to seek shelter, have citizenship, study or want to work in Sweden. In recent time there has been a large increase in applications and the time for which a decision is made has increased. Each type of application (such as citizenship) is a process consisting of several stages. How the decision is going through these steps is called flow. The Swedish Migration Board would therefore like to increase their flow efficiency. When the decision is made and the person has take part of it but is not satisfied, he can appeal. This is one of the most complex processes at the Board. The aim is to analyze how long this process will take and what steps in the process affects the time. One step (which was later found to have a significant effect on time) is opinions. This is when the court requests information on what the person is appealing has to say about why he is appealing. To analyze this, two methods were relevant, accelerated failure time (AFT) and the multi-state models (MSM). One can predict time to event (AFT), the other to analyze the effect of time-manipulation (MSM) in the flow. Opinions early in the process is crucial to how quickly an appeal get judgment while the number of opinions increases the time enormously. There are other factors that affect the time but not so much as opinions. The flow efficiency can be increased by taking time to write an informative opinion which allows the court need not to ask for more opinions.
39
Nicolò, Chiara. "Mathematical modelling of neoadjuvant antiangiogenic therapy and prediction of post-surgical metastatic relapse in breast cancer patients." Thesis, Bordeaux, 2019. http://www.theses.fr/2019BORD0183.
Full textAbstract:
Pour les patients diagnostiqués avec un cancer au stade précoce, les décisions de traitement dépendent de l’évaluation du risque de rechute métastatique. Les outils de pronostic actuels sont fondés sur des approches purement statistiques, sans intégrer les connaissances disponibles sur les processus biologiques à l’oeuvre. L’objectif de cette thèse est de développer des modèles prédictifs du processus métastatique en utilisant une approche de modélisation mécaniste et la modélisation à effets mixtes. Dans la première partie, nous étendons un modèle mathématique du processus métastatique pour décrire la croissance de la tumeur primaire et de la masse métastatique totale chez des souris traitées avec le sunitinib (un inhibiteur de tyrosine kinase ayant une action anti-angiogénique) administré comme traitement néoadjuvant (i.e. avant exérèse de la tumeur primaire). Le modèle est utilisé pour tester des hypothèses expliquant les effets différentiels du sunitinib sur la tumeur primaire et les métastases. Des algorithmes d’apprentissage statistique sont utilisés pour évaluer la valeur prédictive des biomarqueurs sur les paramètres du modèle.Dans la deuxième partie de cette thèse, nous développons un modèle mécaniste pour la prédiction du temps de rechute métastatique et le validons sur des données cliniques des patientes atteintes d’un cancer du sein localisé. Ce modèle offre des prédictions personnalisées des métastases invisibles au moment du diagnostic, ainsi que des simulations de la croissance métastatique future, et il pourrait être utilisé comme un outil de prédiction individuelle pour aider à la gestion des patientes atteintes de cancer du seinFor patients diagnosed with early-stage cancer, treatment decisions depend on the evaluation of the risk of metastatic relapse. Current prognostic tools are based on purely statistical approaches that relate predictor variables to the outcome, without integrating any available knowledge of the underlying biological processes. The purpose of this thesis is to develop predictive models of the metastatic process using an established mechanistic modelling approach and the statistical mixed-effects modelling framework.In the first part, we extend the mathematical metastatic model to describe primary tumour and metastatic dynamics in response to neoadjuvant sunitinib in clinically relevant mouse models of spontaneous metastatic breast and kidney cancers. The calibrated model is then used to test possible hypothesis for the differential effects of sunitinib on primary tumour and metastases, and machine learning algorithms are applied to assess the predictive power of biomarkers on the model parameters.In the second part of this thesis, we develop a mechanistic model for the prediction of the time to metastatic relapse and validate it on a clinical dataset of breast cancer patients. This model offers personalised predictions of the invisible metastatic burden at the time of diagnosis, as well as forward simulations of metastatic growth, and it could be used as a personalised prediction tool to assist in the routine management of breast cancer patients
40
Krol, Agnieszka. "Consideration of multiple events for the analysis and prediction of a cancer evolution." Thesis, Bordeaux, 2016. http://www.theses.fr/2016BORD0329.
Full textAbstract:
Le nombre croissant d’essais cliniques pour le traitement du cancer a conduit à la standardisation de l’évaluation de la réponse tumorale. Dans les essais cliniques de phase III des cancers avancés, la survie sans progression est souvent appliquée comme un critère de substitution pour la survie globale. Pour les tumeurs solides, la progression est généralement définie par les critères RECIST qui utilisent l’information sur le changement de taille des lésions cibles et les progressions de la maladie non-cible. Malgré leurs limites, les critères RECIST restent l’outil standard pour l’évaluation des traitements. En particulier, la taille tumorale mesurée au cours de temps est utilisée comme variable ponctuelle catégorisée pour identifier l’état d’un patient. L’approche statistique de la modélisation conjointe permet une analyse plus précise des marqueurs de réponse tumorale et de la survie. En outre, les modèles conjoints sont utiles pour les prédictions dynamiques individuelles. Dans ce travail, nous avons proposé d’appliquer un modèle conjoint trivarié pour des données longitudinales (taille tumorale), des évènements récurrents (les progressions de la maladie non-cible) et la survie. En utilisant des mesures de capacité prédictive, nous avons comparé le modèle proposé avec un modèle pour les progressions tumorales, définies selon les critères standards et la survie. Pour un essai clinique randomisé porté sur le cancer colorectal, nous avons trouvé une meilleure capacité prédictive du modèle proposé. Dans la deuxième partie, nous avons développé un logiciel en libre accès pour l’application de l’approche de modélisation conjointe proposée et les prédictions. Enfin, nous avons étendu le modèle à une analyse plus sophistiquée de l’évolution tumorale à l’aide d’un modèle mécaniste. Une équation différentielle ordinaire a été mise en œuvre pour décrire la trajectoire du marqueur biologique en tenant compte les caractéristiques biologiques de la croissance tumorale. Cette nouvelle approche contribue à la recherche clinique sur l’évaluation d’un traitement dans les essais cliniques grâce à une meilleure compréhension de la relation entre la réponse tumorale et la survieThe increasing number of clinical trials for cancer treatments has led to standardization of guidelines for evaluation of tumor response. In phase III clinical trials of advanced cancer, progression-free survival is often applied as a surrogate endpoint for overall survival (OS). For solid tumors, progression is usually defined using the RECIST criteria that use information on the change of size of target lesions and progressions of non-target disease. The criteria remain the standard tool for treatment evaluation despite their limitations. In particular, repeatedly measured tumor size is used as a pointwise categorized variable to identify a patient’s status. Statistical approach of joint modeling allows for more accurate analysis of the tumor response markers and survival. Moreover, joint models are useful for individual dynamic predictions of death using patient’s history. In this work, we proposed to apply a trivariate joint model for a longitudinal outcome (tumor size), recurrent events (progressions of non-target disease) and survival. Using adapted measures of predictive accuracy we compared the proposed joint model with a model that considered tumor progressions defined within standard criteria and OS. For a randomized clinical trial for colorectal cancer patients, we found better predictive accuracy of the proposed joint model. In the second part, we developed freely available software for application of the proposed joint modeling and dynamic predictions approach. Finally, we extended the model to a more sophisticated analysis of tumor size evolution using a mechanistic model. An ordinary differential equation was implemented to describe the trajectory of the biomarker regarding the biological characteristics of tumor size under a treatment. This new approach contributes to clinical research on treatment evaluation in clinical trials by better understanding of the relationship between the markers of tumor response with OS
41
Alves, Karina Lumena de Freitas. "Análise de sobrevivência de bancos privados no Brasil." Universidade de São Paulo, 2009. http://www.teses.usp.br/teses/disponiveis/18/18140/tde-28102009-103529/.
Full textAbstract:
Diante da importância do sistema financeiro para a economia de um país, faz-se necessária sua constante fiscalização. Nesse sentido, a identificação de problemas existentes no cenário bancário apresenta-se fundamental, visto que as crises bancárias ocorridas mundialmente ao longo da história mostraram que a falta de credibilidade bancária e a instabilidade do sistema financeiro geram enormes custos financeiros e sociais. Os modelos de previsão de insolvência bancária são capazes de identificar a condição financeira de um banco devido ao valor correspondente da sua probabilidade de insolvência. Dessa forma, o presente trabalho teve como objetivo identificar os principais indicadores característicos da insolvência de bancos privados no Brasil. Para isso, foi utilizada a técnica de análise de sobrevivência em uma amostra de 70 bancos privados no Brasil, sendo 33 bancos insolventes e 37 bancos solventes. Foi possível identificar os principais indicadores financeiros que apresentaram-se significativos para explicar a insolvência de bancos privados no Brasil e analisar a relação existente entre estes indicador e esta probabilidade. O resultado deste trabalho permitiu a realização de importantes constatações para explicar o fenômeno da insolvência de bancos privados no Brasil, bem como, permitiu constatar alguns aspectos característicos de bancos em momentos anteriores à sua insolvência.The financial system is very important to the economy of a country, than its supervision is necessary. Accordingly, the identification of problems in the banking scenario is fundamental, since the banking crisis occurring worldwide throughout history have shown that and instability of the financial system generates huge financial and social costs. The banking failure prediction models are able to identify the financial condition of a bank based on the value of its probability of insolvency. Thus, this study aimed to identify the main financial ratios that can explain the insolvency of private banks in Brazil. For this, it was used the survival analysis to analize a sample of 70 private banks in Brazil, with 33 solvent banks and 37 insolvent banks. It was possible to identify the key financial indicators that were significantly to explain the bankruptcy of private banks in Brazil and it was possible to examine the relationship between these financial ratios and the probability of bank failure. The result of this work has enabled the achievement of important findings to explain the phenomenon of the bankruptcy of private banks in Brazil, and has seen some characteristic of banks in times prior to its insolvency.
42
Winter, Christof Alexander. "Protein interactions in disease: Using structural protein interactions and regulatory networks to predict disease-relevant mechanisms." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2012. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-62260.
Full textAbstract:
Proteins and their interactions are fundamental to cellular life. Disruption of protein-protein, protein-RNA, or protein-DNA interactions can lead to disease, by affecting the function of protein complexes or by affecting gene regulation. A better understanding of these interactions on the molecular level gives rise to new methods to predict protein interaction, and is critical for the rational design of new therapeutic agents that disrupt disease-causing interactions. This thesis consists of three parts that focus on various aspects of protein interactions and their prediction in the context of disease.
In the first part of this thesis, we classify interfaces of protein-protein interactions. We do so by systematically computing all binding sites between protein domains in protein complex structures solved by X-ray crystallography. The result is SCOPPI, the Structural Classification of Protein Protein Interfaces. Clustering and classification of geometrically similar interfaces reveals interesting examples comprising viral mimicry of human interface binding sites, gene fusion events, conservation of interface residues, and diversity of interface localisations. We then develop a novel method to predict protein interactions which is based on these structural interface templates from SCOPPI. The method is applied in three use cases covering osteoclast differentiation, which is relevant for osteoporosis, the microtubule-associated network in meiosis, and proteins found deregulated in pancreatic cancer. As a result, we are able to reconstruct many interactions known to the expert molecular biologist, and predict novel high confidence interactions backed up by structural or experimental evidence. These predictions can facilitate the generation of hypotheses, and provide knowledge on binding sites of promising disease-relevant candidates for targeted drug development.
In the second part, we present a novel algorithm to search for protein binding sites in RNA sequences. The algorithm combines RNA structure prediction with sequence motif scanning and evolutionary conservation to identify binding sites on candidate messenger RNAs. It is used to search for binding sites of the PTBP1 protein, an important regulator of glucose secretion in the pancreatic beta cell. First, applied to a benchmark set of mRNAs known to be regulated by PTBP1, the algorithm successfully finds significant binding sites in all benchmark mRNAs. Second, collaborators carried out a screen to identify changes in the proteome of beta cells upon glucose stimulation while inhibiting gene expression. Analysing this set of post-transcriptionally controlled candidate mRNAs for PTBP1 binding, the algorithm produced a ranked list of 11 high confident potential PTBP1 binding sites. Experimental validation of predicted targets is ongoing. Overall, identifying targets of PTBP1 and hence regulators of insulin secretion may contribute to the treatment of diabetes by providing novel protein drug targets or by aiding in the design of novel RNA-binding therapeutics.
The third part of this thesis deals with gene regulation in disease. One of the great challenges in medicine is to correlate genotypic data, such as gene expression measurements, and other covariates, such as age or gender, to a variety of phenotypic data from the patient. Here, we address the problem of survival prediction based on microarray data in cancer patients. To this end, a computational approach was devised to find genes in human cancer tissue samples whose expression is predictive for the survival outcome of the patient. The central idea of the approach is the incorporation of background knowledge information in form of a network, and the use of an algorithm similar to Google s PageRank. Applied to pancreas cancer, it identifies a set of eight genes that allows to predict whether a patient has a poor or good prognosis. The approach shows an accuracy comparable to studies that were performed in breast cancer or lymphatic malignancies. Yet, no such study was done for pancreatic cancer. Regulatory networks contain information of transcription factors that bind to DNA in order to regulate genes. We find that including background knowledge in form of such regulatory networks gives highest improvement on prediction accuracy compared to including protein interaction or co-expression networks. Currently, our collaborators test the eight identified genes for their predictive power for survival in an independent group of 150 patients. Under a therapeutic perspective, reliable survival prediction greatly improves the correct choice of therapy. Whereas the live expectancy of some patients might benefit from extensive therapy such as surgery and chemotherapy, for other patients this may only be a burden. Instead, for this group, a less aggressive or different treatment could result in better quality of the remaining lifetime.
Conclusively, this thesis contributes novel analytical tools that provide insight into disease-relevant interactions of proteins. Furthermore, this thesis work contributes a novel algorithm to deal with noisy microarray measurements, which allows to considerably improve prediction of survival of cancer patients from gene expression data.
43
Marques, Maria João Pereira Vicente Dias. "Análise retrospetiva de 92 casos de cólica em equinos admitidos em segunda opinião para tratamento hospitalar." Master's thesis, Universidade de Lisboa, Faculdade de Medicina Veterinária, 2018. http://hdl.handle.net/10400.5/15833.
Full textAbstract:
Dissertação de Mestrado Integrado em Medicina VeterináriaA cólica é uma patologia de importância preeminente em equinos, cuja identificação da causa nem sempre é fácil, fazendo com que a determinação precoce de um prognóstico seja essencial. Assim, foi realizado um estudo retrospetivo em 92 casos de cólica recebidos pelo Serviço de Cirurgia e Urgência em Equinos da FMV-ULisboa. Os objetivos do presente estudo foram: 1) caracterizar os casos de cólica referenciados para o SCUE FMV-ULisboa, avaliando o tipo de intervenção clínica, a causa de cólica e a taxa de alta hospitalar; 2) avaliar o valor prognóstico de cada um dos indicadores recolhidos na admissão; 3) comparar o valor destes indicadores entre os dois tipos de intervenção clínica, médica e cirúrgica; e 4) elaborar um modelo multivariado de predição de prognóstico. Estimou-se que 82% dos animais submetidos a intervenção cirúrgica e 75% dos animais tratados medicamente tiveram alta hospitalar, e que 25% dos animais submetidos a laparotomia sofreram íleo pós-cirúrgico. Foram recolhidos na admissão os seguintes dados: idade, tempo entre sinalização e admissão hospitalar, refluxo gastrointestinal, frequência cardíaca, hematócrito, proteínas totais séricas, proteínas totais do líquido peritoneal, lactato peritoneal e lactato sanguíneo. Nas cólicas médicas, os indicadores hematócrito, frequência cardíaca e lactato peritoneal foram considerados estatisticamente significativos (p<0,05), o lactato sanguíneo marginalmente significativo (p=0,053) e as proteínas do líquido peritoneal tendencialmente significativas (p<0,10). Foram elaborados dois modelos de predição multivariável. O modelo de 3 preditores (lactato sanguíneo, frequência cardíaca e hematócrito) com especificidade de 42,9% e sensibilidade de 96,0%. O modelo de 5 preditores (lactato sanguíneo, frequência cardíaca, hematócrito, idade e proteínas totais séricas) com especificidade de 71,4% e sensibilidade de 95,7%. Nas cólicas cirúrgicas, não foi possível determinar preditores significativos nem elaborar modelos de predição. Foi, ainda, criada uma aplicação informática de cálculo de probabilidade de alta hospitalar com base nos modelos descritos. Finalmente, conclui-se que a recolha de líquido peritoneal deverá ser feita com mais frequência pois os seus indicadores parecem transmitir informação valiosa. O modelo de 3 preditores, apesar de ter uma especificidade menor para a amostra em estudo, será provavelmente mais fiável do ponto de vista clínico, para utilização futura. Para além disso, é espectado que com o aumento da dimensão da amostra, estes modelos se tornem mais robustos.
ABSTRACT - A RETROSPECTIVE REVIEW OF 92 EQUINE COLIC CASES REFERRED FOR HOSPITAL TREATMENT - Colic is a really important syndrome in the equine species. To identify a diagnosis can be a true challenge, so the early determination of a prognosis is essential. Therefore, a retrospective study was performed in 92 colic cases admitted at the “Equine Surgery and Emergency Services” (Lisbon University). The objectives of this study were: 1) describe the colic cases and evaluate the clinical approach (medical or surgical), the origin of the problem and rate of survival; 2) estimate the prognostic value of each one of the collected predictors at the admission process; 3) compare the predictors according to the clinical approach; and 4) elaborate a multivariable prognostic prediction model. In this study, the survival rate was 82% for the horses submitted to surgical intervention and 75% for the horses treated medically; and, 25% of the horses in which laparotomy was performed developed post-operative ileus. The following data were collected during admission at the hospital: age, time between the onset of clinical signs and referral, gastrointestinal reflux, cardiac frequency, haematocrit, blood total protein, peritoneal fluid total protein, peritoneal fluid lactate, blood lactate. In medical colics, haematocrit, cardiac frequency and peritoneal fluid lactate were statistically significant (p<0,05), blood lactate was marginally significant (p=0,053) and peritoneal fluid total protein was tendentially significant (p<0,10). Two multivariable prognostic prediction models were elaborated. The three predictors model (blood lactate, cardiac frequency and haematocrit) had a specificity of 42,9% and a sensibility of 96,0%. The five predictors model (blood lactate, cardiac frequency, haematocrit, blood total protein and age) had a specificity of 71,4% and a sensibility of 95,7%. In surgical colics, it wasn’t possible to determine statistically significant predictors neither to elaborate prediction models. Based on the previously descript models, a computerized application to calculate the survival probability was created. It was concluded that peritoneal fluid should be collected more often, since peritoneal lactate and peritoneal fluid total protein seem to be providers of valuable information. Even though, the three predictors model has a reduced specificity for the study sample, it will be probably more reliable from the clinical point of view for further applications. Furthermore, it’s expected that with the increasing of the sample size, these models will get more robust.
N/A
44
Kilic, Hasan, and Eloi Munezero. "Kreditbedömning vid konkurser och varningssignaler : Att förutspå konkurser." Thesis, Luleå tekniska universitet, Institutionen för ekonomi, teknik och samhälle, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-61347.
Full textAbstract:
Varje år går tusentals företag i konkurs, vilket innebär förluster för samhället i stort och för de intressenter som på något sätt kan förknippas till företaget. För banken som lånar ut krediter till företag som går i konkurs innebär det kreditförluster om det inte finns säkerheter som täcker lånet. Därför är behovet av tidiga varningssignaler av stor betydelse för intressenter. Syftet med detta arbete är att teoretiskt analysera och empiriskt pröva varningssignaler för konkurser samt förklara signaler om förestående konkurs i kreditbedömning. Studiens resultat visar att risker för konkurser kan upptäckas med hjälp av finansiella och icke-finansiella nyckeltal. Resultatet i denna studie visar att återbetalningsförmåga, vilket består av soliditet och likviditet är den viktigaste varningssignaler bland de finansiella nyckeltalen. Revisionsanmärkningar, bankens egna mätinstrument, erfarenhet och VD:ns ålder är de viktigaste varningssignalerna bland de icke-finansiella nyckeltalen. Resultatet visar även att dessa varningssignaler blir starkare desto närmare konkurs företaget är.Every year thousands of firms file for bankruptcy, creating considerable loses for the society and stakeholders associated with the firm. A bankruptcy by a firm, that a bank have loaned money to, can also affect the bank considerably if there are not assets enough to cover the outstanding debt. Considering the negative consequences of a bankruptcy it is of paramount interest to be able to spot early warning signals. The study shows that bankruptcy risks can be detected with the help of the firm´s financial and non-financial key assessment indicators. The purpose of this paper is to theoretically and empirically study warning signals of bankruptcies, in order to identify and explain the signals in the credit assessments before occurrence of the bankruptcy. The result of this study shows that the refund assessment consisting of solidity and liquidity are the most important warning signals among the financial key assessment indicators. Remarks, the bank´s own measuring instrument, experience, and the CEO´s age are the most important warning signals among non-financial key assessment indicators. Additionally, results show that the warning signals become stronger the closer the company proceeds towards bankruptcy.
45
Dickie, Ben. "Predicting cancer patient survival using dynamic contrast enhanced MRI." Thesis, University of Manchester, 2017. https://www.research.manchester.ac.uk/portal/en/theses/predicting-cancer-patient-survival-using-dynamic-contrast-enhanced-mri(146dfe97-f892-4cdf-b916-633e9247093e).html.
Full textAbstract:
This thesis describes the use of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to study the prognostic role of microvascular physiology and heterogeneity in locally advanced cancers of the cervix, bladder, and head and neck. To increase the utility of DCE-MRI parameters for prognostication and use in heterogeneity analyses, a novel model fitting approach was developed to reduce the error in two-compartment exchange model (2CXM) parameter estimates. Using this method, precision of 2CXM parameters was increased in 35 of 42 experimental conditions (improvements between 4.7% and 50%) and bias reduced in 30 of 42 conditions (reductions between 1.8% and 49%). The prognostic value of plasma flow, permeability surface area product, and contrast agent volume transfer constant were assessed in a cervix cancer dataset. Plasma flow was the most prognostic parameter (HR = 0.25, P = 0.0086), followed by the volume transfer constant (HR = 0.33, P = 0.031), then the permeability surface area product (HR = 0.43, P = 0.090). Inclusion of plasma flow in survival modelling significantly increased the ability to discriminate between patients with short and long disease-free survival, compared to clinicopathologic factors alone (P = 0.043). The universal prognostic value of microvascular heterogeneity was assessed in cervix, bladder, and head and neck datasets. Following estimation of 2CXM parameters for each patient, a selection of previously published heterogeneity biomarkers were computed and entered into a random survival forest variable selection algorithm. Two variables (vvas, Atrans) were identified as universally prognostic and significantly improved discriminative ability of survival models compared to clinicopathologic factors alone (P < 0.001). Gaussian process models were used to decompose statistical and spatial aspects of intratumoural microvascular heterogeneity. When applied to the three cancer datasets described above, statistical variance in plasma flow (P = 0.00025) was universally prognostic and showed greater discriminative ability compared with spatial scale and average microvascular function parameters. The results of this thesis demonstrate that joint fitting reduces error in DCE-MRI parameters. DCE-MRI estimates of plasma flow appear to hold greater prognostic value than the volume transfer constant and permeability surface area product, and microvascular heterogeneity has potential to provide universal prognostic value. The biomarkers vvas, Atrans, and variance in plasma flow, were identified as universally prognostic. Future work should test the reproducibility of these biomarkers for prognostication in independent datasets.
46
Mauguen, Audrey. "Prognosis of cancer patients : input of standard and joint frailty models." Thesis, Bordeaux, 2014. http://www.theses.fr/2014BORD0240/document.
Full textAbstract:
La recherche sur le traitement des cancers a évolué durant les dernières années principalement dans une direction: la médecine personnalisée. Idéalement, le choix du traitement doit être basé sur les caractéristiques dupatient et de sa tumeur. Cet objectif nécessite des développements biostatistiques, pour pouvoir évaluer lesmodèles pronostiques, et in fine proposer le meilleur. Dans une première partie, nous considérons le problèmede l’évaluation d’un score pronostique dans le cadre de données multicentriques. Nous étendons deux mesuresde concordance aux données groupées analysées par un modèle à fragilité partagée. Les deux niveaux inter etintra-groupe sont étudiés, et l’impact du nombre et de la taille des groupes sur les performances des mesuresest analysé. Dans une deuxième partie, nous proposons d’améliorer la prédiction du risque de décès en tenantcompte des rechutes précédemment observées. Pour cela nous développons une prédiction issue d’un modèleconjoint pour un événement récurrent et un événement terminal. Les prédictions individuelles proposées sontdynamiques, dans le sens où le temps et la fenêtre de prédiction peuvent varier, afin de pouvoir mettre à jourla prédiction lors de la survenue de nouveaux événements. Les prédictions sont développées sur une série hospitalièrefrançaise, et une validation externe est faite sur des données de population générale issues de registres decancer anglais et néerlandais. Leurs performances sont comparées à celles d’une prédiction issue d’une approchelandmark. Dans une troisième partie, nous explorons l’utilisation de la prédiction proposée pour diminuer ladurée des essais cliniques. Les temps de décès non observés des derniers patients inclus sont imputés en utilisantl’information des patients ayant un suivi plus long. Nous comparons trois méthodes d’imputation : un tempsde survie moyen, un temps échantillonné dans une distribution paramétrique et un temps échantillonné dansune distribution non-paramétrique des temps de survie. Les méthodes sont comparées en termes d’estimationdes paramètres (coefficient et écart-type), de risque de première espèce et de puissanceResearch on cancer treatment has been evolving for last years in one main direction: personalised medicine. Thetreatment choice must be done according to the patients’ and tumours’ characteristics. This goal requires somebiostatistical developments, in order to assess prognostic models and eventually propose the best one. In a firstpart, we consider the problem of assessing a prognostic score when multicentre data are used. We extended twoconcordance measures to clustered data in the context of shared frailty model. Both the between-cluster andthe within-cluster levels are studied, and the impact of the cluster number and size on the performance of themeasures is investigated. In a second part, we propose to improve the prediction of the risk of death accountingfor the previous observed relapses. For that, we develop predictions from a joint model for a recurrent event anda terminal event. The proposed individual prediction is dynamic, both the time and the horizon of predictioncan evolve, so that the prediction can be updated at each new event time. The prediction is developed ona French hospital series, and externally validated on population-based data from English and Dutch cancerregistries. Its performances are compared to those of a landmarking approach. In a third part, we explore theuse of the proposed prediction to reduce the clinical trial duration. The non-observed death times of the lastincluded patients are imputed using the information of the patients with longer follow-up. We compared threemethods to impute the data: a survival mean time, a time sampled from the parametric distribution and atime sampled from a non-parametric distribution of the survival times. The comparison is made in terms ofparameters estimation (coefficient and standard-error), type-I error and power
47
Henein, Kringen M. (Kringen Margaret) Carleton University Dissertation Biology. "Predicting the survival of woodland species in human-altered landscapes." Ottawa, 1995.
Find full text
48
Webster, Elizabeth Natalie. "Health care Facilities as a Predictor of Breast Cancer Survival Rates." ScholarWorks, 2018. https://scholarworks.waldenu.edu/dissertations/6145.
Full textAbstract:
The disparity between survival rates for Black and White women with breast cancer is well documented and has been examined in terms socioeconomics, environment, tumor type, and genetics. However, there is little examination of the role of health care facilities in cancer disparities. Health care facilities are representative of societal norms and beliefs that include location, quality of care, finance, policies, and staffing; therefore, they are a proxy for social justice and social change. The purpose of this study was to examine correlations between health care facility type; social determinants of cancer such as poverty, culture, and social justice; and breast cancer survival rates. Using the social determinants of cancer theoretical framework, the breast cancer survival rate of 4,087 Black and White women in Georgia between the ages of 45 and 69 was studied. The relationship between breast cancer survival and predictors including race, income, health care facility type, grade, and tumor type (4 sub-variables) were examined using the Kaplan-Meier Method, log-rank test, and Cox proportional hazard model. The log-rank test suggested no statistically significant difference in the survival functions among patients in different health care facilities (Ï?2(2) = 0.0150, p = 0.9926). The Cox proportional hazard model suggested no statistically significant relationship between breast cancer survival and health care facility type, after controlling for other predictors (Ï?2(2) = 0.3647, p = 0.8333). This result indicates that healthcare facilities do not influence breast cancer survival rates, however, given the persistent health outcome disparities further research in the area is warranted.
49
Rudser, Kyle D. "Variable importance in predictive models : separating borrowing information and forming contrasts /." Thesis, Connect to this title online; UW restricted, 2007. http://hdl.handle.net/1773/9609.
Full text
50
Payne, Kieran. "Predicting patient length of stay and outcome using discrete conditional survival methods." Thesis, Queen's University Belfast, 2012. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.579783.
Full textAbstract:
The health and social care service is increasingly being placed under pressure to facilitate the demands of an ageing population in a difficult economic climate .. The need for better planning and resource management through statistical modelling has never been greater. This thesis adds to the current research on modelling patient length of stay and outcome within hospital by further developing the family of discrete conditional survival models. A statistical technique, consisting of a conditional and process component, is used to model length of stay within hospital based on information known on first day of admission. The approach in this thesis categorises patients into cohorts with similar characteristics and based upon this classification, accurately predicts their length of stay in hospital. Three new techniques, classification trees, ADA boosting and random forests are introduced into the family of discrete conditional survival models. The use of Coxian phase-type distributions for representing length of stay is examined and optimised with the development of more efficient expressions of the probability density function. This is validated in application by modelling length of stay of geriatric patients in Northern Ireland hospitals. The structure of the resulting distributions are discussed and compared to previous research. The advances in the discrete conditional survival model are illustrated in a model developed as a tool for predicting infant length of stay within neonatal care. With the development of late onset sepsis, the model classifies infants as both high or low risk and depending upon the classification accurately models their corresponding length of stay. Performance measures are calculated for each model and the advantages of using the techniques considered and compared against standard methods. The approach not only accurately predicts outcome and length of stay but contributes to knowledge. Development and potential integration within a hospital environment are discussed as further work.