Diplomova praca sa zameriava na oblast zaostrovania obrazkov tvari. V teoretickej casti prace budu prezentovane moderne metody zaostrovania obrazkov pomocou jedineho obrazku a metody editacie obrazkov. Prakticka cast sa zameria na pristupy rekonstrukcie obrazkov zo sekvencie poskodenych obrazkov. Viacere modely neuronovych sieti so vstupom pre viacero obrazkov budu zhotovene a vyhodnotene. Alternativny pristup v podobe balika nastrojov na editaciu obrazkov bude taktiez predstaveny. Tieto nastroje budu vyuzivat najmodernejsie pristupy k editacii obrazkov s cielom spojit vizualne prvky tvari zo vstupnej sekvencie obrazkov do jedneho finalneho vystupu. V zavere prace budu vsetky metody navzajom porovnane.

The thesis deals with the classification of the retinal blood vessels in retinal image data. The first part of the thesis deals with the anatomy of the human eye and focuses on the description of the retina and its blood circulation. It further describes the principle of fundus camera and experimental video ophthalmoscope. The second part of the thesis is devoted to a literature search of academic publications that deal with the classification of the retinal vessels into arteries and veins. Subsequently, the principle of selected machine learning methods is presented. Based on the literature research, two methods for the classification of the blood vessels were proposed, the first one using the SVM classifier and the second one using the convolutional neural network U-Net. At the end, the analysis of vascular pulsations was performed. The practical part of the thesis was carried out in Matlab programming interface and images from the RITE, IOSTAR and AFIO database were used for classification and the retinal video sequences taken with an experimental video ophthalmoscope were processed in the analysis of pulsations.

In cancer diagnosis there is a goal of having the treatment being tailored to each patient. This in order to increase efficiency and reduce side effects. Using more data on each patient can help in achieving this. One such data source is histological images on tissues, such as lymph nodes. This report sets out to find a method in which such images on lymph nodes can be automatically segmented. This so that they can later be analysed and maybe tell in what stage a cancer is in. Such work is today done by hand, and this makes it a subjective process, that might differ between doctors and institutions. If there was a method done by a computer, the process would be replicable and objective. Also, a lot of time would be saved. The results show that such a method is reachable in this early stage of development. It is also quite efficient when segmenting the lymph node itself. The segmentation of smaller areas of the lymph nodes is not as efficient, but with further work in the area it might improve enough to be useful. Some issues are still had since the method relies in part on a person to decide a parameter in order to get a clean segmentation. The final conclusion is that one model is to prefer compared to the others and that further work on this might make it a useful tool in analysing histological images.

Image segmentation through neural networks and deep learning have, in the recent decade, become a successful tool for automated decision-making. For Luossavaara-Kiirunavaara Aktiebolag (LKAB), this means identifying the amount of slag inside a furnace through computer vision.  There are many prominent convolutional neural network architectures in the literature, and this thesis explores two: a modified U-Net and the PSPNet. The architectures were combined with three loss functions and three class weighting schemes resulting in 18 model configurations that were evaluated and compared. This thesis also explores transfer learning techniques for neural networks tasked with identifying slag in images from inside a furnace. The benefit of transfer learning is that the network can learn to find features from already labeled data of another context. Finally, the thesis explored how temporal information could be utilised by adding an LSTM layer to a model taking pairs of images as input, instead of one. The results show (1) that the PSPNet outperformed the U-Net for all tested configurations in all relevant metrics, (2) that the model is able to find more complex features while converging quicker by using transfer learning, and (3) that utilising temporal information reduced the variance of the predictions, and that the modified PSPNet using an LSTM layer showed promise in handling images with outlying characteristics.

Much research has been done on the use of machine learning to extract features such as buildings, lakes et cetera from satellite imagery, and while this dataset is valuable for many use cases, it is limited to time periods in which satellites were used. Historical maps have a much greater range of available time periods but the viability of using machine learning to extract data from these has not been investigated to any great extent. This case study uses a real-world use case to show the efficacy of using a U-Net convolutional neural network to extract features drawn on hand-drawn maps. By implementing a random forest as a pre-screener to the U-Net the goal was to filter out noise that could lead to false positives. By filtering out the noise the hope was to increase the accuracy of the U-Net. The pre-screener in this study has not performed well on the dataset and has not improved the performance of the U-Net. The U-Nets ability to extrapolate the location of features not explicitly drawn on the map was not clearly established. The results of this study show that the U-Net CNN could be an invaluable tool for quickly extracting data from this typically cumbersome data source, allowing for easier access to a wealth of data. The fields of archeology and climate science would find this especially useful.

No description available.

Prior to radiation therapy planning, tumours and organs at risk need to be delineated. In recent years, deep learning models have opened the possibility of automating the contouring process, speeding up the procedures and helping clinicians. However, deep learning models, trained using ground truth labels from different clinicians, inevitably incorporate the human-based inter-observer variability as well as other machine-based uncertainties and biases. Consequently, this affects the accuracy of segmentation, representing the primary source of error in contouring tasks. Therefore, clinicians still need to check and manually correct the segmentation and still do not have a measure of reliability. To tackle these issues, researchers have shifted their focus to the topic of probabilistic neural networks and uncertainties in deep learning models. Hence, the main research question of the project is whether a 3D U-Net neural network trained on CT lung cancer images can enhance clinical contouring practice by implementing a probabilistic auto-contouring system. The Monte Carlo dropout technique was employed to generate probabilistic and uncertainty maps. The model calibration was assessed using reliability diagrams, and subsequently, a clinical experiment with a radiation oncologist was conducted. To assess the clinical validity of the uncertainty maps two novel metrics were identified, namely mean uncertainty (MU) and relative uncertainty volume (RUV). The results of this study demonstrated that probability and uncertainty mapping effectively identify cases of under or over-contouring. Although the reliability analysis indicated that the model tends to be overconfident, the outcomes from the clinical experiment showed a strong correlation between the model results and the clinician’s opinion. The two metrics exhibited promising potential as indicators for clinicians to determine whether correction of the predictions is necessary. Hence, probabilistic models revealed to be valuable in clinical practice, supporting clinicians in their contouring and potentially reducing clinical errors.

The performance of deep neural networks and estimations of their robustness has been rapidly developed. In contrast, despite the broad usage of deep convolutional neural networks (CNNs)[1] for medical image segmentation, research on their uncertainty estimations is being far less conducted. Deep learning tools in their nature do not capture the model uncertainty and in this sense, the output of deep neural networks needs to be critically analysed with quantitative measurements, especially for applications in the medical domain. In this work, epistemic uncertainty, which is one of the main types of uncertainties (epistemic and aleatoric) is analyzed and measured for volumetric medical image segmentation tasks (and possibly more diverse methods for 2D images) at pixel level and structure level. The deep neural network employed as a baseline is 3D U-Net architecture[2], which shares the essential structural concept with U-Net architecture[3], and various techniques are applied to quantify the uncertainty and obtain statistically meaningful results, including test-time data augmentation and deep ensembles. The distribution of the pixel-wise predictions is estimated by Monte Carlo simulations and the entropy is computed to quantify and visualize how uncertain (or certain) the predictions of each pixel are. During the estimation, given the increased network training time in volumetric image segmentation, training an ensemble of networks is extremely time-consuming and thus the focus is on data augmentation and test-time dropouts. The desired outcome is to reduce the computational costs of measuring the uncertainty of the model predictions while maintaining the same level of estimation performance and to increase the reliability of the uncertainty estimation map compared to the conventional methods. The proposed techniques are evaluated on publicly available volumetric image datasets, Combined Healthy Abdominal Organ Segmentation (CHAOS, a set of 3D in-vivo images) from Grand Challenge (https://chaos.grand-challenge.org/). Experiments with the liver segmentation task in 3D Computed Tomography (CT) show the relationship between the prediction accuracy and the uncertainty map obtained by the proposed techniques. / Prestandan hos djupa neurala nätverk och estimeringar av deras robusthet har utvecklats snabbt. Däremot, trots den breda användningen av djupa konvolutionella neurala nätverk (CNN) för medicinsk bildsegmentering, utförs mindre forskning om deras osäkerhetsuppskattningar. Verktyg för djupinlärning fångar inte modellosäkerheten och därför måste utdata från djupa neurala nätverk analyseras kritiskt med kvantitativa mätningar, särskilt för tillämpningar inom den medicinska domänen. I detta arbete analyseras och mäts epistemisk osäkerhet, som är en av huvudtyperna av osäkerheter (epistemisk och aleatorisk) för volymetriska medicinska bildsegmenteringsuppgifter (och möjligen fler olika metoder för 2D-bilder) på pixelnivå och strukturnivå. Det djupa neurala nätverket som används som referens är en 3D U-Net-arkitektur [2] och olika tekniker används för att kvantifiera osäkerheten och erhålla statistiskt meningsfulla resultat, inklusive testtidsdata-augmentering och djupa ensembler. Fördelningen av de pixelvisa förutsägelserna uppskattas av Monte Carlo-simuleringar och entropin beräknas för att kvantifiera och visualisera hur osäkra (eller säkra) förutsägelserna för varje pixel är. Under uppskattningen, med tanke på den ökade nätverksträningstiden i volymetrisk bildsegmentering, är träning av en ensemble av nätverk extremt tidskrävande och därför ligger fokus på dataaugmentering och test-time dropouts. Det önskade resultatet är att minska beräkningskostnaderna för att mäta osäkerheten i modellförutsägelserna samtidigt som man bibehåller samma nivå av estimeringsprestanda och ökar tillförlitligheten för kartan för osäkerhetsuppskattning jämfört med de konventionella metoderna. De föreslagna teknikerna kommer att utvärderas på allmänt tillgängliga volymetriska bilduppsättningar, Combined Healthy Abdominal Organ Segmentation (CHAOS, en uppsättning 3D in-vivo-bilder) från Grand Challenge (https://chaos.grand-challenge.org/). Experiment med segmenteringsuppgiften för lever i 3D Computed Tomography (CT) vissambandet mellan prediktionsnoggrannheten och osäkerhetskartan som erhålls med de föreslagna teknikerna.

A brain tumor is an abnormal growth of cells within the brain, which can be categorized into primary and secondary tumor types. The most common type of primary tumors in adults are gliomas, which can be further classified into high-grade gliomas (HGGs) and low-grade gliomas (LGGs). Approximately 50% of patients diagnosed with HGG pass away within 1-2 years. Therefore, the early detection and prompt treatment of brain tumors are essential for effective management and improved patient outcomes.  Brain tumor segmentation is a task in medical image analysis that entails distinguishing brain tumors from normal brain tissue in magnetic resonance imaging (MRI) scans. Computer vision algorithms and deep learning models capable of analyzing medical images can be leveraged for brain tumor segmentation. These algorithms and models have the potential to provide automated, reliable, and non-invasive screening for brain tumors, thereby enabling earlier and more effective treatment. For a considerable time, Convolutional Neural Networks (CNNs), including the U-Net, have served as the standard backbone architectures employed to address challenges in computer vision. In recent years, the Transformer architecture, which already has firmly established itself as the new state-of-the-art in the field of natural language processing (NLP), has been adapted to computer vision tasks. The Vision Transformer (ViT) and the Swin Transformer are two architectures derived from the original Transformer architecture that have been successfully employed for image analysis. The emergence of Transformer based architectures in the field of computer vision calls for an investigation whether CNNs can be rivaled as the de facto architecture in this field.  This thesis compares the performance of four model architectures, namely the Swin Transformer, the Vision Transformer, the 2D U-Net, and the 2D U-Net which is implemented with the nnU-Net framework. These model architectures are trained using increasing amounts of brain tumor images from the BraTS 2020 dataset and subsequently evaluated on the task of brain tumor segmentation for both HGG and LGG together, as well as HGG and LGG individually. The model architectures are compared on total training time, segmentation time, GPU memory usage, and on the evaluation metrics Dice Coefficient, Jaccard Index, precision, and recall. The 2D U-Net implemented using the nnU-Net framework performs the best in correctly segmenting HGG and LGG, followed by the Swin Transformer, 2D U-Net, and Vision Transformer. The Transformer based architectures improve the least when going from 50% to 100% of training data. Furthermore, when data augmentation is applied during training, the nnU-Net outperforms the other model architectures, followed by the Swin Transformer, 2D U-Net, and Vision Transformer. The nnU-net benefited the least from employing data augmentation during training, while the Transformer based architectures benefited the most.  In this thesis we were able to perform a successful comparative analysis effectively showcasing the distinct advantages of the four model architectures under discussion. Future comparisons could incorporate training the model architectures on a larger set of brain tumor images, such as the BraTS 2021 dataset. Additionally, it would be interesting to explore how Vision Transformers and Swin Transformers, pre-trained on either ImageNet- 21K or RadImageNet, compare to the model architectures of this thesis on brain tumor segmentation.

When optimizing a controlled dose in radiotherapy, high resolution spatial information about healthy organs in close proximity to the malignant cells are necessary in order to mitigate dispersion into these organs-at-risk. This information can be provided by deep volumetric segmentation networks, such as 3D U-Net. However, due to limitations of memory in modern graphical processing units, it is not feasible to train a volumetric segmentation network on full image volumes and subsampling the volume gives a too coarse segmentation. An alternative is to sample a region of interest from the image volume and train an organ-specific network. This approach requires knowledge of which region in the image volume that should be sampled and can be provided by a 3D object detection network. Typically the detection network will also be region specific, although a larger region such as the thorax region, and requires human assistance in choosing the appropriate network for a certain region in the body.  Instead, we propose a multi-site object detection network based onYOLOv3 trained on 43 different organs, which may operate on arbitrary chosen axial patches in the body. Our model identifies the organs present (whole or truncated) in the image volume and may automatically sample a region from the input and feed to the appropriate volumetric segmentation network. We train our model on four small (as low as 20 images) site-specific datasets in a weakly-supervised manner in order to handle the partially unlabeled nature of site-specific datasets. Our model is able to generate organ-specific regions of interests that enclose 92% of the organs present in the test set. / Vid optimering av en kontrollerad dos inom strålbehandling krävs det information om friska organ, så kallade riskorgan, i närheten av de maligna cellerna för att minimera strålningen i dessa organ. Denna information kan tillhandahållas av djupa volymetriskta segmenteringsnätverk, till exempel 3D U-Net. Begränsningar i minnesstorleken hos moderna grafikkort gör att det inte är möjligt att träna ett volymetriskt segmenteringsnätverk på hela bildvolymen utan att först nedsampla volymen. Detta leder dock till en lågupplöst segmentering av organen som inte är tillräckligt precis för att kunna användas vid optimeringen. Ett alternativ är att endast behandla en intresseregion som innesluter ett eller ett fåtal organ från bildvolymen och träna ett regionspecifikt nätverk på denna mindre volym. Detta tillvägagångssätt kräver dock information om vilket område i bildvolymen som ska skickas till det regionspecifika segmenteringsnätverket. Denna information kan tillhandahållas av ett 3Dobjektdetekteringsnätverk. I regel är även detta nätverk regionsspecifikt, till exempel thorax-regionen, och kräver mänsklig assistans för att välja rätt nätverk för en viss region i kroppen. Vi föreslår istället ett multiregions-detekteringsnätverk baserat påYOLOv3 som kan detektera 43 olika organ och fungerar på godtyckligt valda axiella fönster i kroppen. Vår modell identifierar närvarande organ (hela eller trunkerade) i bilden och kan automatiskt ge information om vilken region som ska behandlas av varje regionsspecifikt segmenteringsnätverk. Vi tränar vår modell på fyra små (så lågt som 20 bilder) platsspecifika datamängder med svag övervakning för att hantera den delvis icke-annoterade egenskapen hos datamängderna. Vår modell genererar en organ-specifik intresseregion för 92 % av organen som finns i testmängden.