Dissertationen zum Thema „Ultrasound Navigation“

We have designed and demonstrated a new class of medical navigation methods that use the fingerprint-like biometrically distinct ultrasound echo patterns produced by different locations in tissue. As an example of this new biometric navigation approach, we have constructed and tested a system that uses ultrasound data to achieve prospective motion compensation in MRI, especially for respiratory motion during interventional MRI procedures in moving organs such as the liver. The ultrasound measurements are collated with geometrical information from MRI during a training stage to form a mapping table that relates ultrasound measurements to positions. During prospective correction, the system makes frequent ultrasound measurements and uses the map to determine the corresponding position. Results in motorized linear motion phantoms and freely breathing animals indicate that the system performs well. Apparent motion is reduced by up to 97.8%, and motion artifacts are reduced or eliminated in 2D Spoiled Gradient-Echo images. The motion compensation is sufficient to permit MRI thermometry of focused ultrasound heating during respiratory-like motion, with results similar to those obtained in the absence of motion. This new technique may have applications for MRI thermometry and other dynamic imaging in the abdomen during free breathing. We have also extended this technique to situations in which external position information during training is unavailable or incomplete, by extending the concept of Simultaneous Localization and Mapping to include determining the topology of a dense motion path through a gaussian random field. In the course of these investigations, we have also developed modified forms of referenceless MRI thermometry and Kalman filtering, specially adapted to optimize accuracy under our experimental conditions.

The need for spectrally efficient transmission on mobile and wireless channels is prevalent. A promising scheme for such transmission is adaptive coded modulation. In this thesis, techniques for assessing the performance of such systems are presented. One of the vulnerable points of such systems is the need for a reliable feedback channel. Channel prediction is proposed as a technique to combat the harmful effects of feedback delay.

The Nakagami distribution is often employed in a model for the fading envelope of a wireless channel; this leads to a gamma-distributed signaltonoise ratio. Nakagami (1960) provides expressions for the probability density function (PDF) of the product, sum, and ratio of two correlated gamma-distributed random variables (RVs). However, such an expression for the difference between two such RVs has not been provided by Nakagami.

A new expression for this PDF is provided in this dissertation, and it is shown that it is closely related to a distribution first described by McKay (1932). Applications of the new PDF include outage probability calculation in an environment with self-interference and assessment of the quality of certain channel estimation techniques.

Epidural anesthesia is a common but challenging procedure in obstetrics and surgery, especially for the obese patient, and can result in complications such as dural puncture and nerve injury. Ultrasound has the potential to significantly improve epidural needle guidance, by being able to depict the spinal anatomy and the epidural space. An ultrasound guidance system is therefore proposed, using a transducer-mounted camera to create 3D panorama images of the spine relative to markings on the skin. Guidance will include depiction of the spinal anatomy, identification of individual vertebrae, and selection of a suitable puncture site, trajectory and depth of needle insertion. The camera tracks the transducer movement using a specialized strip of markers attached to the skin surface. This enables 6-DOF absolute position estimation of the transducer with respect to the patient over the full range of the spine. The 3D panorama image can then be resliced in various parasagittal planes to show either the target epidural spaces or the laminae. The overall accuracy of the panorama reconstruction is validated by measuring inter-feature distances of a phantom of steel beads against measurements obtained from an optical tracking system (Optotrak), resulting in an average error of 0.64 mm between camera and Optotrak. The algorithm is then tested in vivo by creating panorama images from human subjects (n=20), obtaining measurements for depth of insertion to the epidural space, intervertebral spacings, and registration of interspinous gaps to the skin, and validating these against independent measurements by an experienced sonographer. The results showed an average error of 1.69 mm (4.23%) for the depth measurements, average error of 4.44 mm (15.2%) for the interspinous distance measurements, and an average error of 6.65 mm for registering the interspinous gaps to the skin (corresponding to 18.5% of the interspinous distances). Tracking of ultrasound images with respect to the marker is implemented in real time and visualized using the 3D Slicer software package.

L'objectif des travaux présentés dans ce document est de développer de nouvelles méthodes de traitement d'images pour améliorer la planification et le guidage d'une thérapie par voie transœsophagienne de la fibrillation auriculaire à l'aide d'Ultrason Focalisé Haute Intensité. Le document est divisé en deux parties : la planification du traitement et le guidage de la thérapie. Pour la planification de la thérapie, l'idée est d'exploiter l'information acquise au stade préopératoire par un scanner X ou IRM afin de retrouver l'anatomie spécifique du patient et à y définir le futur geste thérapeutique. Plus particulièrement, nos différentes contributions ont porté sur une approche multi-atlas de segmentation de l'oreillette gauche et des veines pulmonaires ; le tracé des lignes de lésions sur le volume initial ou segmenté ; et la reconstruction d'un volume adapté à la future navigation transœsophagienne. Pour le guidage de la thérapie, nous proposons une nouvelle approche de recalage qui permet d'aligner les images échographiques peropératoires 2D et l'information 3D CT préopératoire. Dans cette approche, dans un premier temps nous avons sélectionné la mesure de similarité la plus adaptée à notre problématique à l'aide d'une évaluation systématique puis nous avons tiré profit des contraintes imposées à la sonde transœsophagienne par l'anatomie du patient pour simplifier la procédure de recalage. Toutes ces méthodes ont été évaluées sur des fantômes numériques ou physiques et sur des données cliniques
The work presented in this document aims at developing new image-processing methods to improve the planning and guidance of transesophageal HIFU atrial fibrillation therapy. This document is divided into two parts, namely therapy planning and therapy guidance. We first propose novel therapy planning methods that exploit high-resolution pre-operative CT or MRI information to extract patient-specific anatomical details and to define future therapeutic procedures. Our specific methodological contributions concern the following: an automatically-refined atlas-based segmentation approach to extract the left atrium and pulmonary veins; the delineation of the lesion lines on the original or segmented volume; and the reconstruction of a volume adapted to future intraoperative transesophageal navigation. Secondly, our proposal of a novel registration approach for use in therapy guidance aligns intraoperative 2D ultrasound with preoperative 3D CT information. This approach first carries out a systematic statistical evaluation to select the best similarity measure for our application and then takes advantage of the geometrical constraints of the transesophageal HIFU probe to simplify the registration process. Our proposed methods have been evaluated on digital and/or physical phantoms and on real clinical data

Le carcinome hépatocellulaire (CHC) est le sixième cancer en termes d'incidence dans le monde et le quatrième le plus meurtrier en 2018. La transplantation hépatique est le moyen le plus efficace de traiter le CHC, permettant de traiter le cancer et la maladie hépatique sous-jacente. Cependant, en raison de la pénurie de greffons, elle n'est pratiquée que chez 3 à 4 % des patients. Les traitements d'ablation thermique administrés par voie interstitielle constituent une alternative moins invasive pour le patient et présentent l'avantage de préserver une plus grande proportion de tissu non tumoral. Toutefois, ces techniques de traitement interstitiel ne permettent pas de traiter les CHC de plus de 2,5 cm de rayon ou nécessitent l'insertion de plusieurs aiguilles de traitement, ce qui augmente la complexité de la procédure. En outre, le traitement n'est pas conformationnel : le volume d'ablation thermique ne s'adapte pas à la forme de la tumeur. Cela peut parfois empêcher le traitement d'être effectué sans risquer d'endommager des structures anatomiques critiques. Les techniques de thérapie par ultrasons focalisés de haute intensité (HIFU) sont prometteuses dans le cas du CHC car elles permettraient de réaliser des ablations conformationnelles sur un rayon étendu jusqu’à 3 cm. L'utilisation de ce type de technologie pourrait traiter des CHC actuellement non traitables tout en préservant les structures à risque, élargissant ainsi le rôle des ablations interstitielles. Des simulations d'un nouveau cathéter ultrasonore ont été réalisées pour concevoir un cahier des charges et vérifier la faisabilité du projet en termes de focalisation et d'ablation. Un cathéter ultrasonore bimodal (assurant l’imagerie B-mode et la thérapie thermique avec les mêmes éléments) de 3 mm de diamètre et de 64 éléments piézoélectriques fonctionnant à 5,5 MHz a été sous-traité en conséquence. Tout d'abord, les capacités d'imagerie échographique ont été évaluées et confirmées. Le prototype a ensuite été caractérisé électroniquement et acoustiquement. Les performances thermiques du cathéter ont été étudiées en trois dimensions sous thermométrie IRM et ont validé les outils de simulation tout en démontrant l'aspect directionnel des échauffements induits. Ces résultats ont conduit à réaliser des essais in vitro sur foie animal. Des ablations radiales centimétriques ont confirmé, pour la première fois, les capacités du cathéter à effectuer des ablations thermiques. La robustesse du prototype sur l'ensemble des essais a été étudiée. Enfin, la remise en place d'une plateforme de navigation échographique a donné lieu à la reconstruction de volumes tumoraux en 3 dimensions. L'association du cathéter avec cette plateforme robotisée a permis de générer des ablations thermiques volumiques en 3 dimensions et de traiter des volumes compatibles avec les tumeurs primaires rencontrées en pratique clinique
Hepatocellular carcinoma (HCC) is the sixth most common cancer in terms of incidence worldwide, and the fourth most deadly in 2018. Liver transplantation is the most effective way to treat HCC, enabling both the cancer and the underlying liver disease to be treated. However, due to the shortage of grafts, it is performed in only 3% to 4% of patients. Interstitial thermal ablation treatments offer a less invasive alternative for the patient, and have the advantage of preserving a greater proportion of non-tumoral tissue. However, these interstitial treatment techniques are unable to treat HCCs larger than 2.5 cm in radius, or require the insertion of multiple treatment needles, increasing the complexity of the procedure. In addition, the treatment is not conformal: the thermal ablation volume does not adapt to the shape of the tumor. This can sometimes prevent the treatment from being carried out without risking damage to critical anatomical structures. High-intensity focused ultrasound (HIFU) therapy techniques hold promise for HCC, enabling conformal ablation over an extended radius of up to 3 cm. The use of this type of technology could treat currently untreatable HCC while preserving at-risk structures, thus expanding the role of interstitial ablations. Simulations of a new ultrasound catheter were carried out to design specifications and verify the feasibility of the project in terms of focusing and ablation. A bimodal ultrasound catheter (providing B-mode imaging and thermal therapy with the same elements) with a diameter of 3 mm and 64 piezoelectric elements operating at 5.5 MHz was subcontracted accordingly. First, ultrasound imaging capabilities were assessed and confirmed. The prototype was then characterized electronically and acoustically. The thermal performance of the catheter was studied in three dimensions under MRI thermometry, validating the simulation tools and demonstrating the directional aspect of induced heating. These results led to in vitro tests on animal liver. Centimeter radial ablations confirmed, for the first time, the catheter's ability to perform thermal ablations. The prototype's robustness over the full range of tests was also studied. Finally, the re-installation of an ultrasound navigation platform led to the reconstruction of tumor volumes in 3 dimensions. Combining the catheter with this robotized platform enabled the generation of 3-D volumetric thermal ablations, and the treatment of volumes compatible with primary tumors encountered in clinical practice

L’échographie ultrarapide est une nouvelle méthode d'imagerie basée sur la transmission des ondes planes et qui permet d'imager les tissus biologiques à très haute cadence et avec une excellente résolution spatiale. Lorsqu'elle est appliqué à l'imagerie Doppler, elle permet d'améliorer considérablement la détection du flux sanguin dans les vaisseaux avec une sensibilité considérablement augmentée par rapport à l'imagerie conventionnelle basée sur la transmission des ondes focalisées. Ainsi le Doppler ultrarapide a ouvert d'autres champs d'applications à l'échographie faisant d’elle une nouvelle modalité précieuse de neuro-imagerie fonctionnelle capable de reconstruire la micro-vascularisation cérébrale mais aussi de mesurer indirectement l’activité neuronale en se basant sur le couplage neuro-vasculaire. Cependant l’expansion de l’imagerie fonctionnelle ultrasonore (fUS) est limitée par la difficulté à se repérer dans le réseau vasculaire complexe du cerveau, ce qui rend complexe le positionnement de la sonde échographique et l’analyse des données. Il est donc crucial de mettre en place des outils de neuro-informatique dédiés à l’imagerie fonctionnelle ultrasonore au même titre que l’imagerie fonctionnelle par IRM qui est la modalité de référence.Les travaux de cette thèse portent donc sur le développement et la validation du GPS cérébral, un outil de neuro-navigation automatique à partir des empreintes vasculaires Doppler ultrasensible de la souris et du rat. Dans un premier temps un template anatomique vasculaire a été construit puis recalé sur des atlas familiers (Atlas Allen pour la souris et l’atlas SIGMA pour le rat) permettant ainsi la création d’un atlas vasculaire pouvant servir de référence lors des sessions d’imagerie fUS pour recaler des données expérimentales et leur fournir un contexte anatomique. La précision du recalage vasculaire a été quantifié à partir des images vasculaires super-résolues obtenues par microscopie de localisation ultrasonore.Par la suite après avoir développé la chaîne neuro-informatique intégrée à une nouvelle génération de neuro-imageurs, nous avons montré la capacité du GPS cérébral à guider le positionnement d’une sonde linéaire sur des zones fonctionnelles choisies afin d’obtenir des cartes d’activation fonctionnelle même sur des plans obliques complexes. Nous avons aussi montré que le GPS cérébral peut être utilisé pour calculer automatiquement la matrice de connectivité en fournissant un contexte anatomique et une segmentation automatique des zones fonctionnelles.Pour aller plus loin nous avons exploré le potentiel du GPS vasculaire à guider automatiquement les injections intra-cérébrales dans des structures profondes. Des résultats encourageants confirmés par des images de microscopie de fluorescence ont pu être obtenus après injection d’un traceur neuronal dans le noyau thalamique. Toujours à titre exploratoire nous nous sommes intéressés à la reconstruction des images anatomiques structurelles du cerveau en plus des images vasculaires généralement générées afin d’évaluer l’échogénicité de certaines zones cérébrales identifiées à l’aide du GPS vasculaire. Enfin nous avons conduit une étude de quantification longitudinale dans le cadre du processus de consolidation du trace mnésique chez la souris. Ces travaux permettent d’apporter des nouveaux outils de neuro-imagerie pour renforcer le potentiel de l’imagerie fonctionnelle ultrasonore et permettre aux experts et aux non-experts neuroanatomistes de réaliser des protocoles standardisés, reproductibles, avec plus de précision et impliquant des études sur de grosses cohortes
Ultrafast ultrasound imaging is a recent method based on transmission of plane waves which enables the visualization of biological media with high spatio-temporal resolution. When applied to Doppler imaging, it enables detection of blood flow with very high sensitivity compared to conventional ultrasound Doppler usually limited to high blood flow imaging in cardiology. This advances have rendered functional ultrasound (fUS) imaging a valuable neuroimaging modality capable of mapping cerebral vascular networks, but also to indirectly capture neuronal activity with high sensitivity thanks to the neurovascular coupling. However, the expansion of fUS imaging is still limited by the difficulty to identify cerebral structures during experiments based solely on the Doppler images and the shape of the vessels, which complicates the positioning of the ultrasound probe and the data analysis of the data. It is therefore crucial to set up tools dedicated to functional ultrasound imaging in the same way as functional imaging by MRI, which remains the reference modality.This thesis focuses on the development and validation of cerebral GPS, an automatic neuro-navigation tool based on ultrasensitive Doppler vascular footprint of mice and rats. Initially, a vascular anatomical template was built and then registered on familiar atlases (Allen Atlas for the mouse and the SIGMA atlas for the rat) thus allowing the creation of a vascular atlas that can serve as a reference during fUS imaging sessions to align experimental data and provide an anatomical context. The accuracy of the vascular registration was quantified from the super-resolved vascular images obtained with ultrasound localization microscopy.Secondly, the neuro-informatic pipeline has been developed and integrated into a new generation of neuroimaging devices to perform online navigation. We then demonstrated the capability of the system to position itself automatically over chosen anatomical structures and to obtain corresponding functional activation maps even in complex oblique planes. Additionally, we show that the system can be used to acquire and estimate functional connectivity matrices automatically.To go further we explored the potential of vascular GPS to automatically guide intracerebral injections into deep structures. Encouraging results confirmed by fluorescence microscopy images were obtained after injection of a neuronal tracer into the thalamic nucleus. Still on an exploratory basis, we were interested in the reconstruction of structural anatomical images of the brain in addition to the vascular images generally reconstructed in order to assess the echogenicity of a few brain areas identified using vascular GPS. Finally we carried out a longitudinal study in the framework of memory consolidation in rats.This work provides new neuroimaging tools to strengthen the potential of functional ultrasound imaging and allow neuroanatomists experts and non-expert to carry out standardized, reproducible protocols with more accuracy and involving studies on large cohorts

L’objectif principal d’une Arthroplastie Totale de Hanche (ATH) est de réduire la douleur en reproduisant les propriétés mécaniques de l’articulation de la hanche à l'aide d'implants artificiels, qui assurent également une fonctionnalité optimale. Ce travail traite de la navigation de l’ATH, à base d’échographie, en mettant l'accent sur la navigation de l'implant fémoral. La conception d'un système de navigation pour l’ATH, basé sur l’échographie peropératoire et CT préopératoire et focalisant sur l’implant fémoral, est proposée. Il fournit un feed back visuel et quantitatif sur l'insertion de l'implant. Le pipeline de navigation proposé inclut : la calibration de la sonde, la segmentation tenant compte de la géométrie d’acquisition, et le recalage profitant de la forme quasi symétrique du fémur. La précision est évaluée en termes cliniques.Les temps d'exécution et la précision obtenus sont encourageants et rendent le pipeline proposé valable pour être évalué en utilisant des données réelles des patients. Une fois notre étude réussie le défi peropératoire, d’autres applications telles que l'arthroplastie totale de l'épaule et l'ablation du cancer osseux peuvent en profiter
The principal aim of a Total Hip Arthroplasty (THA) is to reduce pain by reproducing the hip joint mechanical properties using artificial implants. This work addresses the ultrasound based navigation of THA, with a focus on the femoral implant. Using a calibrated 3D ultrasound probe, the aim is to image and recognize the femur after implant insertion, then to provide quantitative and visual navigation instructions for optimal implant placement. US bone interfaces are estimated per line of sight, then mapped to CT model profiting from the femur symmetry. The inaccuracy of the proposed navigation system is quantified in clinical terms. The obtained runtimes and accuracies make the proposed pipeline worthy to be tested in the OR. Once the proposed pipeline succeeds the intraoperative challenge, navigated bone tumor ablation and total shoulder arthroplasty can profit from our work

Le carcinome hépatocellulaire est le principal cancer primaire du foie. Les procédures d’ablation thermique par voie interstitielle constituent un type de traitement curatif de ce cancer. Ces méthodes ne permettent pas toujours, de par leur nature physique (radio fréquence, micro-onde, laser, cryothérapie), de générer une ablation conformationnelle pour un volume tumoral donné. Dans certains cas, cela peut entraîner l’ablation d’un volume important de tissus non tumoraux. L'utilisation d'une sonde interstitielle ultrasonore disposant d’un transducteur avec de nombreux éléments indépendants, capable de générer des ultrasons focalisés de haute intensité (HIFU), permettrait théoriquement de lever cette limitation. D’autre part, le nombre élevé d’éléments permettrait également de disposer de capacités d’imagerie. Les travaux présentés dans cette thèse ont donc d’abord porté sur la conception du transducteur associé à ce type de sonde. Un design particulier a été proposé spécifiquement pour le traitement d’une tumeur de 4 cm de diamètre. S’est ensuite posé la question de la stratégie de planification du traitement à adopter pour obtenir une lésion la plus conformationnelle possible. Différentes stratégies ont été évaluées par simulations numériques. Toutes ont présentés des caractéristiques semblables tant en termes de conformation que de temps de traitement total. La focalisation ultrasonore s’est avérée en elle-même suffisante pour générer un traitement conformationnel. Finalement, une plateforme robotique a été développée pour le pilotage de prototypes de sondes ultrasonores interstitielles bimodales, aussi bien en mode imagerie qu’en mode thérapie. Cette plateforme a permis de réaliser in vitro, les planifications de traitement automatiques de plusieurs fantômes de tumeurs, en se basant sur la reconstruction ultrasonore 3D issue de l’imagerie échographique obtenue in situ par les sondes ultrasonores. En revanche, en mode thérapie les prototypes de sondes n’ont pas atteint leurs spécifications et n’ont pas réussi à générer des lésions thermiques dans des tissus hépatiques in vitro. La modularité de la plateforme robotique a rendu possible son utilisation avec un système de thérapie ultrasonore différent, à la fiabilité éprouvé. Avec ce système, la planification automatique du traitement, ainsi que l’exécution du traitement HIFU associé, ont pu être effectué in vitro avec succès par la plateforme
Hepatocellular carcinoma is the most common primary cancer of the liver. Interstitial thermal ablation procedures constitute a type of curative treatments for this cancer. Given the physical nature of the phenomenon used to modify temperature (radio frequency, micro wave, laser, cryotherapy), those methods may not be able to generate a conformal treatment for a given tumor shape. In some cases, this limitation may induce the thermal ablation of a large volume of non-tumor tissues. The use of an ultrasound interstitial probe mounted with a multi-element transducer capable of generating high intensity focused ultrasound (HIFU) may theoretically help to overcome this limitation. Also a transducer with an important number of elements may also provide in situ imaging. As a first step, the design of a transducer for interstitial ultrasound probe was studied. A specific configuration has been proposed for the treatment of tumors with a diameter of 4 cm. The question of the treatment planning method to adopt to reach an optimal conformal treatment has been then addressed by comparing numerical simulations of different strategies. All strategies were sufficiently conformal and none presented real assets compared to the others. Ultrasound focusing in itself provided the desired conformal thermal ablation. Finally, a robotic platform was developed for driving interstitial dual mode ultrasound probes, both in imaging and in therapy mode. This platform allowed the automatic treatment planning of in vitro tumor mimic phantoms, based on 3D ultrasound reconstruction from the B mode images obtained in situ by the interstitial probe. However, in therapy mode, the probes did not reach their specifications and did not manage to create thermal lesions in in vitro liver tissue sample. The modularity of the robotic platform allowed driving a different HIFU system, which was more robust. With this system, the platform managed to perform with success an automatic treatment planning and then the associated HIFU treatment in in vitro tissue sample

This diploma thesis deals with processing of measured data from inertial navigation system in order these could be used for stabilization. There is general information about aerial vehicles called copters with emphasis on four-rotor construction called quadrocopter at first. Then mathematical model of quadrocopter in state space form is derived, the particular implementation of university developed quadrocopter is described and the design of data processing algorithm is presented with measured results. Finally achieved results are discussed.

Dans ce mémoire, nous présentons une méthode de suivi de fond pour petits véhicules autonomes sous-marins sous-actionnés de type « torpille civile ». Nous établissons les deux principales limitations qui contraignent leurs déplacements : le rayon de courbure minimal et la pente maximale de leur trajectoire. Nous proposons ensuite une méthode globale de suivi de fond prenant en compte ces contraintes. La génération des trajectoires suppose la connaissance de la position du véhicule dans un environnement bathymétrique connu a priori. Cette dernière condition n'étant pas toujours vérifiée, nous adaptons la méthode globale au cas, local, où le véhicule navigue dans un environnement inconnu dont il découvre progressivement les caractéristiques physiques. La méthode employée alors exploite les données bathymétriques acquises par deux sondeurs mono-faisceaux disposés dans le nez du véhicule. Le choix de ce type de capteurs est justifié et les modèles des faisceaux acoustiques sont étudiés. L'adjonction d'un troisième sondeur à l'avant du véhicule permet de tenir compte également de l'éventualité de se trouver face à un aplomb, c'est-à-dire une falaise sous-marine. Un algorithme basé sur un découpage du faisceau en trois zones de perception permet de classer les échos reçus et d'adapter le comportement du véhicule en conséquence. On détermine ainsi la taille de l'obstacle et la manière de l'éviter afin de garantir la sécurité du véhicule tout en tâchant de suivre le fond au plus près. Les résultats des simulations du comportement hydrodynamique du véhicule Taipan ainsi que des essais en milieu naturel sont présentés.

碩士
國立中央大學
機械工程研究所
100
Ultrasonic guided biopsy is the most popular approach to diagnose whether tumor is benign or malignant. However, biopsy by free-hand or using probe-attached guidance device cannot avoid needle deflection and misalignment due to external forces. In addition, the biopsy direction must follow and along the plane of ultrasound image which restrict the flexibility to adjust the needle direction to avoid critic tissues. Therefore, it is hard for physicians to do deep and small tumor biopsy clinically. In this research, an ultrasound-assisted biopsy navigation system has been developed. By using electromagnetic tracking system and ultrasound positioning technology, the positions of ultrasound images, tumors, and biopsy needle are transferred to the same coordinate system. Real-time image display of the relative position of tumor and biopsy needle tip will assist the physician to plan and guide biopsy direction. Moreover, a biopsy assistive device, using a stepping motor to enable the rotation of the needle, is designed to reduce needle deflection due to external forces. To reduce the influence of liver displacement and deformation to biopsy accuracy due to respiration, a Laser Range Finder is applied to measure the ups and downs of the patient’s chest to control the breath status during insertion is identical to that during ultrasound scanning. In the experiments, pork liver and pork belly are applied as biopsy phantoms. The results of several experiments show that the average and maximum distance errors of using 18 Gauge needle are 3.17mm and 4.38mm respectively, while the average and maximum distance errors of using 21 Gauge needle are 3.36mm and 4.62mm respectively. It indicates that the developed system should be able to `assist physicians to biopsy deep tumor with a diameter greater than 1cm.

碩士
中華大學
資訊工程學系
106
Agriculture provides humans with food to maintain the fundamental needs of life. Agricultural economic development is directly related to the survival of humanity, national security, and social stability. At present, the decreasing birth rate in Taiwan and the reduction and ageing of its farming population have become urgent problems in agricultural development that need to be solved. There are more agriculture-related problems to be addressed, such as climate change and extreme climate caused by global warming, the resultant lower crop yield, the decreased area of farmland due to energy security issues, and market economy price competition after joining the WTO. The most important agricultural problem in Taiwan to solve is how to promote Intelligent Agriculture 4.0. Only by improving the traditional agricultural management methods and production techniques, by using advanced information technology to enhance the circulation of produce sales and marketing information, and introducing sensing technology, intelligent robots (IR), Internet of Things (IoT), Big Data analysis, artificial intelligence and deep learning (recognition of crop growth status), and other technologies to transform and develop intelligent agriculture, can we break through the current difficulties in agricultural development. The objective of this thesis is to apply the unmanned aerial vehicle (UAV) to the indoor automatic cultivation of mushrooms. In the future, drone imaging can be used in deep learning to automatically monitor and analyze the growth status of mushrooms, and automatically deal with poor growth status. The advantage of combining UAV indoor inspection and IoT environment control in the mushroom farm is that automatic control will not affect crop growth nor pollute the farm environment, and will reduce manpower demand and the impact of the ageing farming population; in addition, environmentally controlled cultivation by indoor sensors will not be affected by climate change. The main contribution of this thesis is to design and implement a UAV indoor ultrasonic flight control and obstacle avoidance system in the Arduino environment to solve the core problem of indoors UAV flight, so as to improve the technical depth of the automation and informatization of indoor mushroom cultivation.

La pratica chirurgica sta costantemente sostituendo i tradizionali approcci operatori con tecniche minimamente invasive, queste ultime forniscono molti benefici per il paziente come per esempio la riduzione delle complicanze post-operative, un tempo di recupero ridotto che garantisce un’ospedalizzazione più breve. Le procedure percutanee sono probabilmente le tecniche minimamente invasive più diffuse, quest’approccio è utilizzato sia per la diagnosi sia per il trattamento di aree patologiche localizzate in molti distretti del corpo umano: testa e collo, arti, polmoni, cuore e in generale molte delle strutture in ambito addominale e pelvico. Nell’ablazione percutanea uno strumento chirurgico che assomiglia a grosso ago è accuratamente inserito attraverso la pelle del paziente in corrispondenza di un’area patologica per distruggere il tessuto malato tramite fonti d’intensa energia o bassa temperatura. Il secondo approccio è chiamato crio-ablazione percutanea e fornisce risultati nel medio termine molto incoraggiante se confrontati con altre tecniche minimamente invasive. L’ablazione percutanea richiede l’accurato posizionamento degli strumenti per garantire un completo trattamento della zona patologica, ottenendo quindi la possibilità di ricorrenza della medesima condizione patologica o di altre complicazioni pre o post operative. Anche se queste procedure possono essere eseguite alla “cieca”, l’introduzione della guida tramite immagini mediche può migliorare l’esito di queste procedure anche nel caso di utenti inesperti o situazioni patologiche critiche. L’ablazione percutanea può beneficiare in modo molto efficace dall’introduzione di un sistema di navigazione computerizzato, capace di fornire una guida integrando le immagini cliniche con la localizzazione degli strumenti chirurgici. Le caratteristiche richieste da questi sistemi sono la capacità di eseguire la registrazione automatica tra immagini intra e pre operative, pianificare la traiettoria nell’inserimento di aghi e guida nel corso dell’inserimento con interfaccia semplice e intuitiva da utilizzare. Inoltre il monitoraggio accurato del posizionamento degli aghi e della zona ablata è fondamentale per ottenere un trattamento ottimale. I sistemi di guida basati su ecografica forniscono le caratteristiche richieste in un sistema compatto e dai costi contenuti. Algoritmi di elaborazione avanzata sono necessari per superare alcuni dei limiti di questi sistemi, soprattutto legati alla ridotta qualità e risoluzione delle immagini ecografiche. L’introduzione di un sistema di rilevamento della posizione (tracker) capace di misurare la posizione e l’orientamento di una sonda ecografica e di altri strumenti nel corso dell’intervento è una parte fondamentale di un sistema di navigazione e può permettere la ricostruzione tridimensionale automatica dell’anatomia del paziente basata sull’estrazione di punti salienti dalle immagini bidimensionali. Per permettere una ricostruzione tridimensionale corretta, una procedura di calibrazione è necessaria per stimare la trasformazione sconosciuta tra il sistema di riferimento del tracker e quello dell’immagine ecografica. Nelle procedure di ablazione percutanee, monitorare l’area trattata è fondamentale per garantire la completa distruzione del tessuto malato ed evitare la ricorrenza a breve termine. Nel caso specifico della crio-ablazione è molto complesso eseguire questo monitoraggio poiché l’area congelata produce forti riflessioni e ombre che rendono estremamente complesso garantire la stima corretta dell’area ablata. In questa tesi descriveremo diversi metodi capaci di migliorare gli attuali sistemi di navigazione ecografici. Il primo contributo di questa tesi riguarda l’introduzione di una procedura di calibrazione basata su dati acquisiti con diversi settaggi di profondità per il calcolo di una singola trasformazione rigida. La procedura è basata sulla stima automatica della dimensione del fascio ultrasonoro in modo da migliorare l’integrazione di feature estratte a diverse profondità di acquisizione. Il metodo proposto permette una calibrazione della sonda ecografica con un controllo meno stringente dei parametri di acquisizione e senza effetti significativi sulla precisione e l’accuratezza della calibrazione. Questo potrebbe facilitare l’adozione di un sistema ecografico integrato con un tracker nell’ambiente clinico reale, dove i parametri di acquisizione sono cambiati molto spesso per garantire la corretta visualizzazione dell’anatomia del paziente. Il secondo contributo di questa tesi consiste nello sviluppo di un rilevatore e descrittore di punti salienti specificatamente pensato per le immagini ecografiche, dove le informazioni d’intensità non sono stabili e affidabili. Il metodo è basato sull’operatore “Local Binary Pattern” calcolato lungo diverse direzioni e scale sulle mappe di congruenza di fase. Queste scelte permettono una localizzazione robusta di punti salienti nelle immagini ecografiche anche alla presenza di trasformazioni geometriche e d’intensità. Questi punti salienti possono essere usati per la localizzazione basata su immagini degli strumenti o delle aree patologiche o per permettere la registrazione con altre tipologie d’immagini acquisite prima, durante o dopo la procedura chirurgica. Il terzo contributo consiste nell’introduzione di un sistema di guida compatto che integra un piccolo schermo direttamente sullo strumento di ablazione per fornire indicazioni su come muovere lo strumento per garantire il corretto inserimento lungo la traiettoria pianificata. Fornire indicazione all’utente in modo intuitivo e semplice è fondamentale per garantire la corretta esecuzione della traiettoria pianificata, di conseguenza migliorando i risultati finali e riducendo i tempi e i rischi delle procedure. Il quarto contributo supera l’impossibilità di monitorare l’area di ablazione con immagini ecografiche nel corso di procedure di crio-ablazione. Il metodo è basato sull’elastografia a ultrasuoni, una tecnica utilizzata per misurare l’elasticità del tessuto. Siccome il congelamento produce danni cellulari, si è deciso di misurare l’elasticità del tessuto prima e dopo il completo scongelamento del tessuto. Questo metodo permette di monitorare le aree trattate con la crio-ablazione subito dopo la fine dell’intervento, permettendo in questo modo l’adozione di azioni correttive nel caso di trattamento non completo o dell’insorgenza di complicazioni. I contributi presentati in questa tesi, se integrati in un sistema ecografico di navigazione chirurgica, possono migliorare le caratteristiche di questi sistemi avanzati con specifica attenzione per i requisiti dettati dalla pratica clinica reale.
Surgical practice is constantly replacing traditional invasive approaches with minimally invasive techniques, which provide many benefits for the patient such as reduced postoperative complications, faster recovery and shorten hospitalization. Percutaneous procedures are probably the most widely used minimally invasive technique, this approach is used both for diagnosis (i.e. biopsy) and for the treatment of localized pathological areas in many regions of the human body: head and neck, limbs, lungs, hearth, abdominal and pelvic structures. In percutaneous ablation a surgical tool that resembles a thick needle is accurately inserted through the skin of the patient in correspondence of a pathological area to destroy diseased tissue thanks to high energy or low temperature. The latter approach is called percutaneous cryoablation and provides very encouraging mid-term outcome compared with other minimally invasive techniques. Percutaneous ablation requires accurate positioning of the tools to guarantee the complete treatment of diseased area, thus reducing the possible re-insurgence of same pathological condition or other intra or post operative complications. Although these procedures could be performed blindly the introduction of image guidance could improve the outcome of the procedure even in the case of inexpert users or critical cases. Therefore percutaneous ablation could greatly benefit from the introduction of a computer navigation system, able to provide integrated guidance with imaging sensor and tools localization. Required characteristics are the ability of performing automatic registration with other intra-operative and pre-operative dataset, planning needle trajectory and guidance during the insertion with effective and user-friendly interface. Furthermore the accurate monitoring of the needle position and ablation area is fundamental to obtain a full treatment of the pathological area. Ultrasound guided systems provide the needed characteristics in a compact and coste ffective device. Advance processing techniques are necessary to overcome some limitations of ultrasound based system, mainly connected to limited quality and resolution of these images. The introduction of a tracking system able to measure the position and the orientation of ultrasound probe and tools is a key component of a navigation system, and could enable the automatic 3D reconstruction of patient anatomy based on the extraction of feature points from 2D images. To enable the correct 3D reconstruction a calibration procedure is necessary to estimate an unknown rigid transformation between tracking and image reference systems. In percutaneous ablation procedures, monitoring the ablation area is fundamental to guarantee to complete suppression of the diseased tissues and avoid short term recurrence. The specific case of cryoablation is very challenging since the frozen zone produces strong re ections and shadows that make it unfeasible to guarantee the proper monitoring of the ablated area. In this thesis we describe different methods able to improve the current ultrasound based navigation systems. The first contribution of this thesis is the introduction of an ultrasound calibration procedure based on data acquired with different acquisition depth settings for the computation of a single rigid transformation. The procedure is based on the automatic estimation of the beam width to improve the integration of feature extracted from different depth settings. The proposed method enables the calibration of ultrasound probe with less strict control of the acquisition parameters and without significant effect on the calibration accuracy and precision. This would ease the adoption of tracked ultrasound system in the real clinical condition, where acquisition parameters are changed very often to guarantee the correct visualization of patient anatomy. The second contribution of this thesis is the development of a feature detector and descriptor able to localize and match salient points from ultrasound image. These methods have been designed and tested with specific attention to ultrasound images, where the intensity information are not stable. The detector is based a local energy model in place of the widely adopted gradient methods, where feature points are localized based on the phase congruency of Fourier components. The detector is based on Local Binary Pattern operator computed over a local angle and direction of the phase congruency. These choices enable the robust localization of feature point in ultrasound image in presence of intensity and geometrical transformation. These feature points could be used for the image based localization of tools or pathological areas, or to enable the registration with other imaging dataset acquired before, during or after the surgical procedure. The third contribution is the introduction of a compact navigation device that integrates a small display directly into the ablation tool to provide indication on how to move the needle in the proper way and to guarantee the correct insertion along the planned trajectory. Providing indications to the user in a confortable and effective manner is fundamental to guarantee the correct execution of the planned trajectory, consequently improving the final result of the procedure with reduced risks and reduced procedure duration. The fourth contribution overcomes the unfeasible monitoring of the ablation area with ultrasound image during cryoablation procedures, and it is based on ultrasound elastography. Ultrasound elastography is used for the measurement of tissue elasticity; since the freezing produces cellular structural damage we measure the tissue elastic properties before and after the complete thawing of the tissue. This method would enable the monitoring of cryo ablated area immediately after the intervention, thus enabling the effective adoption of corrective actions in case of not complete treatment or complications. We believe that the contributions described in this thesis, if integrated in an ultrasound guided navigation system, will improve the characteristics of these advanced systems with specific attention to real clinical requirements.