By combining computed tomography data with computational fluid dynamics, the cardiac hemodynamics of a patient can be assessed for diagnosis and treatment of cardiac disease. The advantage of computed tomography over other medical imaging modalities is its capability of producing detailed high resolution images containing geometric measurements relevant to the simulation of cardiac blood flow. To extract these geometries from computed tomography data, segmentation of 4D cardiac computed tomography (CT) data has been performed using two deep learning frameworks that combine methods which have previously shown success in other research. The aim of this thesis work was to develop and evaluate a deep learning based technique to segment the left ventricle, ascending aorta, left atrium, left atrial appendage and the proximal pulmonary vein inlets. Two frameworks have been studied where both utilise a 2D multi-axis implementation to segment a single CT volume by examining it in three perpendicular planes, while one of them has also employed a 3D binary model to extract and crop the foreground from surrounding background. Both frameworks determine a segmentation prediction by reconstructing three volumes after 2D segmentation in each plane and combining their probabilities in an ensemble for a 3D output.  The results of both frameworks show similarities in their performance and ability to properly segment 3D CT data. While the framework that examines 2D slices of full size volumes produces an overall higher Dice score, it is less successful than the cropping framework at segmenting the smaller left atrial appendage. Since the full size 2D slices also contain background information in each slice, it is believed that this is the main reason for better segmentation performance. While the cropping framework provides a higher proportion of each foreground label, making it easier for the model to identify smaller structures. Both frameworks show success for use in 3D cardiac CT segmentation, and with further research and tuning of each network, even better results can be achieved.

Analysis of overhead imagery has a great potential to produce real-time data cost-effectively. This can be an important foundation for decision-making for businesses and politics. Every day a massive amount of new satellite imagery is produced. To fully take advantage of these data volumes a computationally efficient pipeline is required for the analysis. This thesis proposes a pipeline which outperforms the Segment Before you Detect network [6] and different types of fast region based convolutional neural networks [61] with a large margin in a fraction of the time. The model obtains a prediction error for counting cars of 1.67% on the Potsdam dataset and increases the vehiclewise F1 score on the VEDAI dataset from 0.305 reported by [61] to 0.542. This thesis also shows that it is possible to outperform the Segment Before you Detect network in less than 1% of the time on car counting and vehicle detection while also using less than half of the resolution. This makes the proposed model a viable solution for large-scale satellite imagery analysis. / Analys av flyg- och satellitbilder har stor potential att kostnadseffektivt producera data i realtid för beslutsfattande för företag och politik. Varje dag produceras massiva mängder nya satellitbilder. För att fullt kunna utnyttja dessa datamängder krävs ett beräkningseffektivt nätverk för analysen. Denna avhandling föreslår ett nätverk som överträffar Segment Before you Detect-nätverket [6] och olika typer av snabbt regionsbaserade faltningsnätverk [61]  med en stor marginal på en bråkdel av tiden. Den föreslagna modellen erhåller ett prediktionsfel för att räkna bilar på 1,67% på Potsdam-datasetet och ökar F1- poängen for fordons detektion på VEDAI-datasetet från 0.305 rapporterat av [61]  till 0.542. Denna avhandling visar också att det är möjligt att överträffa Segment Before you Detect-nätverket på mindre än 1% av tiden på bilräkning och fordonsdetektering samtidigt som den föreslagna modellen använder mindre än hälften av upplösningen. Detta gör den föreslagna modellen till en attraktiv lösning för storskalig satellitbildanalys.

Historical topographic maps are valuable resources for studying past landscapes, but they are unsuitable for geospatial analysis. Cartographic map elements must be extracted and digitized for use in GIS. This can be accomplished by sophisticated image processing and pattern recognition techniques, and more recently, artificial intelligence. While these methods are generally effective, they require high levels of technical expertise. This study presents a straightforward method to digitally extract historical topographic map elements from within popular GIS software, using new and rapidly evolving toolsets. A convolutional neural network deep learning model was used to extract elevation contour lines from a 1940 United States Geological Survey (USGS) quadrangle in Sevier County, TN, ultimately producing a Digital Elevation Model (DEM). The topographically derived DEM (TOPO-DEM) is compared to a modern LiDAR-derived DEM to analyze its quality and utility. GIS-capable historians, archaeologists, geographers, and others can use this method in research and land management.

The main result of this thesis is a deep learning model named BearNet, which can be trained to detect an arbitrary amount of objects as a set of points. The model is trained using the Weighted Hausdorff distance as loss function. BearNet has been applied and tested on two problems from the industry. These are: From an intensity image, detect two pocket points of an EU-pallet which an autonomous forklift could utilize when determining where to insert its forks. From a depth image, detect the start, bend and end points of a straw attached to a juice package, in order to help determine if the straw has been attached correctly. In the development process of BearNet I took inspiration from the designs of U-Net, UNet++ and a high resolution network named HRNet. Further, I used a dataset containing RGB-images from a surveillance camera located inside a mall, on which the aim was to detect head positions of all pedestrians. In an attempt to reproduce a result from another study, I found that the mall dataset suffers from training set contamination when a model is trained, validated, and tested on it with random sampling. Hence, I propose that the mall dataset is evaluated with a sequential data split strategy, to limit the problem. I found that the BearNet architecture is well suited for both the EU-pallet and straw datasets, and that it can be successfully used on either RGB,  intensity or depth images. On the EU-pallet and straw datasets, BearNet consistently produces point estimates within five and six pixels of ground truth, respectively. I also show that the straw dataset only constitutes a small subset of all the challenges that exist in the problem domain related to the attachment of a straw to a juice package, and that one therefore cannot train a robust deep learning model on it. As an example of this, models trained on the straw dataset cannot correctly handle samples in which there is no straw visible.

The aim of the following thesis was to study the issue of optical disc and retinal vessels segmentation in ophthalmologic sequences. The theoretical part of the thesis summarizes the principles of different approaches in the field of deep learning, which are used in connection with the given issue. Based on the theoretical part, methods for optical disk segmentation and retinal vessel segmentation based on the convolutional neural networks Linknet, PSPNet, Unet and MaskRCNN are proposed. The practical part of the thesis deals with the description of their implementation and subsequent evaluation.

The goal of this thesis was to propose an image segmentation method, which is capable of effective segmentation process with small datasets. Recently published ODE neural network was used for this method, because its features should provide better generalization in case of tasks with only small datasets available. The proposed ODE-UNet network was created by combining UNet architecture with ODE neural network, while using benefits of both networks. ODE-UNet reached following results on ISBI dataset: Rand: 0,950272 and Info: 0,978061. These results are better than the ones received from UNet model, which was also tested in this thesis, but it has been proven that state of the art can not be outperformed using ODE neural networks. However, the advantages of ODE neural network over tested UNet architecture and other methods were confirmed, and there is still a room for improvement by extending this method.

Deep Learning techniques are widely used across various medical imaging applications. However, they are often fine-tuned for a specific modality and are not generalizable when it comes to new modalities or datasets. One of the main reasons for this is large data variations for e.g., the dynamic range of intensity values is large across multi-modal images. The goal of the project is to develop a method to address multi-modal learning that aims at segmenting liver from Computed Tomography (CT) images and abdominal organs from Magnetic Resonance (MR) images using deep learning techniques. In this project, a self-supervised approach is adapted to attain domain adaptation across images while retaining important 3D information from medical images using a simple 3D-UNet with a few auxiliary tasks. The method comprises of two main steps: representation learning via self-supervised learning (pre-training) and fully supervised learning (fine-tuning). Pre-training is done using a 3D-UNet as a base model along with some auxiliary data augmentation tasks to learn representation through texture, geometry and appearances. The second step is fine-tuning the same network, without the auxiliary tasks, to perform the segmentation tasks on CT and MR images. The annotations of all organs are not available in both modalities. Thus the first step is used to learn general representation from both image modalities; while the second step helps to fine-tune the representations to the available annotations of each modality. Results obtained for each modality were submitted online, and one of the evaluations obtained was in the form of DICE score. The results acquired showed that the highest DICE score of 0.966 was obtained for CT liver prediction and highest DICE score of 0.7 for MRI abdominal segmentation. This project shows the potential to achieve desired results by combining both self and fully-supervised approaches.

Liver segmentation is a cumbersome task when done manually, often consuming quality time of radiologists. Use of automation in such clinical task is fundamental and the subject of most modern research. Various computer aided methods have been incorporated for this task, but it has not given optimal results due to the various challenges faced as low-contrast in the images, abnormalities in the tissues, etc. As of present, there has been significant progress in machine learning and artificial intelligence (AI) in the field of medical image processing. Though challenges exist, like image sensitivity due to different scanners used to acquire images, difference in imaging methods used, just to name a few. The following research embodies a convolutional neural network (CNN) was incorporated for this process, specifically a U-net algorithm. Predicted masks are generated on the corresponding test data and the Dice similarity coefficient (DSC) is used as a statistical validation metric for performance evaluation. Three datasets, from different scanners (two1.5 T scanners and one 3.0 T scanner), have been evaluated. The U-net performs well on the given three different datasets, even though there was limited data for training, reaching upto DSC of 0.93 for one of the datasets.

This study presents the development and evaluation process of an improved CG-ResUnet neural network model for retinal blood vessel segmentation. The methodology includes preprocessing techniques such as CLAHE, Kirsch and grey filtering to improve image quality. The developed model showed the highest precision (0.961), but it also showed the lowest area under the curve (AUC) (0.919). The lowest recall (0.872) indicates that the model still has potential for improvement in minimising false results and accurately identifying vessel pixels. The precision (accuracy) of the model (0.631) is higher than other models, indicating that this model is highly sensitive. However, additional tuning is required to achieve higher accuracy and overall segmentation quality. F1-Score (0.729) and Dice score (0.729) were also higher than other models, indicating high potential for growth with further tuning. A hybrid post-processing approach combining automatic segmentation with manual adjustments is proposed to improve segmentation accuracy, especially for complex images with thin vessels. Future research should focus on improving accuracy and solving segmentation problems in areas of high complexity to further improve diagnostic efficiency and reduce manual labor in clinical settings. / В данном исследовании представлен процесс разработки и оценки усовершенствованной нейросетевой модели CG-ResUnet для сегментации кровеносных сосудов сетчатки. Методология включает в себя такие методы предварительной обработки, как CLAHE, Кирша и серая фильтрация для улучшения качества изображения. Разработанная модель показала самый высокий показатель точности (0,961), однако она также продемонстрировала самый низкий показатель площади под кривой (AUC) (0,919). Самый низкий показатель recall (0,872) указывает на то, что модель все еще имеет потенциал для улучшения в минимизации ложных результатов и точном определении пикселей сосудов. Точность (precision) модели (0,631) превышает показатели других моделей, что указывает на высокую чувствительность данной модели. Однако для достижения более высокой точности и общего качества сегментации требуется дополнительная настройка. Показатели F1-Score (0,729) и Dice score (0,729) также оказались выше, чем у других моделей, что свидетельствует о высоком потенциале для роста при последующей настройке. Для повышения точности сегментации, особенно для сложных изображений с тонкими сосудами, предлагается гибридный подход к постобработке, сочетающий автоматическую сегментацию с ручными корректировками. Будущие исследования должны быть направлены на повышение точности и решение проблем сегментации в областях с высокой сложностью для дальнейшего повышения диагностической эффективности и сокращения ручного труда в клинических условиях.