Magnetic resonance imaging (MRI) is a medical imaging method that has become increasingly more important during the last 4 decades. This is partly because it allows us to acquire a 3D-representation of a part of the body without exposing patients to ionizing radiation. Furthermore, it also typically gives better contrast between soft tissues than x-ray based techniques such as CT. The image acquisition procedure of MRI is also much more flexible. One can vary the signal sequence, not only to change how different types of tissue map to different intensities, but also to measure flow, diffusion or even brain activity over time.  Machine learning has gained great impetus the last decade and a half. This is probably partly because of the work done on the mathematical foundations of machine learning done at the end of last century in conjunction with the availability of specialized massively parallel processors, originally developed as graphical processing units (GPUs), which are ideal for training or running machine learning models. The work presented in this thesis combines MRI and machine learning in order to leverage the large amounts of MRI-data available in open data sets, to address questions of clinical relevance about the brain.  The thesis comprises three studies. In the first one the subproblem which augmentation methods are useful in the larger context of classifying autism, was investigated. The second study is about predicting brain age. In particular it aims to construct light-weight models using the MRI volumes in a condensed form, so that the model can be trained in a short time and still reach good accuracy. The third study is a development of the previous that investigates other ways of condensing the brain volumes. / Magnetresonansavbildningar, ofta kallat MR eller MRI, är en bilddiagnostik-metod som har blivit allt viktigare under de senaste 40 åren. Detta på grund av att man kan erhålla 3D-bilder av kroppsdelar utan att utsätta patienter för joniserande strålning. Dessutom får man typiskt bättre kontraster mellan mjukdelar än man får med motsvarande genomlysningsmetod (CT, eller 3D röntgen). Själva bildinsamlingsförfarandet är också mera flexibelt med MR. Man kan genom att ändra program för utsända och registrerade signa-ler, inte bara ändra vad som framförallt framträder på bilden (t.ex. vatten, fett, H-densitet, o.s.v.) utan även mäta flöde och diffusion eller till och med hjärnaktivitet över tid. Maskininlärning har fått ett stort uppsving under 2010-talet, dels på grund av utveckling av teknologin för att träna och konstruera maskininlärningsmodeller dels på grund av tillgängligheten av massivt parallella specialprocessorer – initialt utvecklade för att generera datorgrafik. Detta arbete kombinerar MR med maskininlärning, för att dra nytta av de stora mängder MR data som finns samlad i öppna databaser, för att adressera frågor av kliniskt intresse angående hjärnan. Avhandlingen innehåller tre studier. I den första av dessa undersöks del-problemet vilken eller vilka metoder för att artificiellt utöka träningsdata som är bra vid klassificering om en person har autism. Det andra arbetet adresserar bedömning av så kallad "hjärn-ålder". Framför allt strävar arbetet efter att hitta lättviktsmodeller som använder en komprimerad form av varje hjärnvolym, och därmed snabbt kan tränas till att bedöma en persons ålder från en MR-volym av hjärnan. Det tredje arbetet utvecklar modellen från det föregående genom att undersöka andra typer av komprimering. / <p><strong>Funding:</strong> This research was supported by the Swedish research council (2017-04889), the ITEA/VINNOVA project ASSIST (Automation, Surgery Support and Intuitive 3D visualization to optimize workflow in IGT SysTems, 2021-01954), and the Åke Wiberg foundation (M20-0031, M21-0119, M22-0088).</p>

We develop efficient techniques for the non-rigid registration of medical images by using representations that adapt to the anatomy found in such images. Images of anatomical structures typically have uniform intensity interiors and smooth boundaries. We create methods to represent such regions compactly using tetrahedra. Unlike voxel-based representations, tetrahedra can accurately describe the expected smooth surfaces of medical objects. Furthermore, the interior of such objects can be represented using a small number of tetrahedra. Rather than describing a medical object using tens of thousands of voxels, our representations generally contain only a few thousand elements. Tetrahedra facilitate the creation of efficient non-rigid registration algorithms based on finite element methods (FEM). We create a fast, FEM-based method to non-rigidly register segmented anatomical structures from two subjects. Using our compact tetrahedral representations, this method generally requires less than one minute of processing time on a desktop PC. We also create a novel method for the non-rigid registration of gray scale images. To facilitate a fast method, we create a tetrahedral representation of a displacement field that automatically adapts to both the anatomy in an image and to the displacement field. The resulting algorithm has a computational cost that is dominated by the number of nodes in the mesh (about 10,000), rather than the number of voxels in an image (nearly 10,000,000). For many non-rigid registration problems, we can find a transformation from one image to another in five minutes. This speed is important as it allows use of the algorithm during surgery. We apply our algorithms to find correlations between the shape of anatomical structures and the presence of schizophrenia. We show that a study based on our representations outperforms studies based on other representations. We also use the results of our non-rigid registration algorithm as the basis of a segmentation algorithm. That algorithm also outperforms other methods in our tests, producing smoother segmentations and more accurately reproducing manual segmentations.

Thesis (Ph.D.)--University of Western Australia, 2004.

During resting state the brain exhibits synchronized activity within all major brain networks. Using blood oxygen level dependent (BOLD) resting state functional magnetic resonance imaging (fMRI) based detection it is possible to quantify the degree of correlation, connectivity, between regions of interest and assess information regarding the integrity of the inter-regional functional integration. A newly available multiband echo planar imaging (EPI) fMRI sequence allows for faster scan times which possibly allows us to better examine large-scale networks and increase the understanding of brain function/dysfunction. This thesis will assess how the newly developed sequence compares to a conventional EPI sequence for detecting resting state connectivity of canonical brain networks. The data acquisitions were made on a 3 Tesla scanner using a 32 channel head coil. The hypothesis was that the multiband sequence would produce a better result since it has faster sampling rate, thus more data points in its time-series to support the statistical analyses. Using Pearson’s linear correlation between the average time-series (approximately 12 minutes long) within a seed-region and all voxels contained in the image volume, correlation maps where created for each of the eight participants using data normalized to Montreal Neurological Institute (MNI) space. The resting state networks (RSN) were then found by performing a one sample T-test on group level. Six seed-coordinates, based on literature, where used revealing the the homotopic connections in anterior Hippocampus, Motor cortex, Dorsal attention, Visual and the Default mode network (DMN) as well for an anterior-posterior connection in the DMN. By comparing the maximum T-values within the regions for the RSN no systematic difference could be found between the multiband and conventional fMRI data. Further tests were conducted to evaluate if the sequences would differentiate in their results if the acquisition time was shortened, i.e shortening the time-series in the voxels. However no such difference could be established.Importantly, the results are specific to the 32 channel head coil used in the current study. Presumably recently available and improved coil designs could better exploit the multiband technique.

No description available.

Radiologists often have to look through many different patients and examinations in quick succession, and to aid in the workflow the different types of images should be presented for the radiologist in the same manner and order between each new examination. Thus decreasing the time needed for the radiologist to either find the correct image or rearrange the images to their liking. A step in thisprocess requires a comparison between two images to be made and produce a score between 0-1 describing how similar the images are. A similar algorithm already exists at Sectra, but that algorithm only uses the metadata from the images without considering the actual pixel data. The aim of this thesis were to explore different methods of doing the same comparison as the previous algorithm but only using the pixel data. Considering only 3D volumes from CT examinations of the abdomen and thorax region, this thesis explores the possibility of using SSIM, SIFT and SIFT together with a histogram comparison using the Bhattacharyya distance for this task. It was deemed very important that the ranking produced when ordering the images in terms of similarity to one reference image followed a specific order. This order was determined by consulting personnel at Sectra that works closely with the clinical side of radiology. SSIM were able to differentiate between different plane orientations since they usually had large resolution differences in each led, but it could not be made tofollow the desired ranking and was thus disregarded as a reliable option for this problem. The method using SIFT followed the desired ranking better, but struggled a lot with differentiating between the different contrast phases. A histogram component were also added to this method, which increased the accuracy and improved the ranking. Although, further development is still needed for thismethod to be a reliable option that could be used in a clinical setting.

There is a need for quantitative CT data in radiation therapy. Currently there are only few algorithms that address this issue, for instance the commercial DirectDensity algorithm. In scientific literature, an example of such an algorithm is DIRA. DIRA is an iterative model-based reconstruction method for dual-energy CT whose goal is to determine the material composition of the patient from accurate linear attenuation coefficients. It has been implemented in a two dimensional geometry, i.e., it could process axial scans only.  There was a need to extend DIRA so that it could process projection data generated in helical scanning geometries. The newly developed algorithm (DIRA-3D) implemented (i) polyenergetic semi-parallel projection generation, (ii) mono-energetic parallel projection generation and (iii) the PI-method for image reconstruction. The computation experiments showed that the accuracies of the resulting LAC and mass fractions were comparable to the ones of the original DIRA. The results converged after 10 iterations.

This thesis investigated whether a deep learning model could learn features of Chaos,from the Chaos &amp; Clues evaluation protocol, in a given dermatoscopic image data set. Asuccessful result could be of use in a future decision-support system for when dermatologists examine skin lesions for traces of melanoma (type of skin cancer). The chosen deep learning model (Inception V3) was trained to recognise four classesrelated to Chaos. Anonymous patient data was used, provided by the partnering companyGnosco. The data was partitioned into one or two classes depending on the symmetryproperties found in the corresponding image annotation. More than twenty differentmodel configurations was run to obtain the results in this thesis. The results indicate that the chosen model was not capable of learning features of Chaosfrom the dermatoscopical image data-set. Training the model to recognise features ofChaos resulted in an overfit system with low validation accuracy (close to 30%). The prediction target was changed to contrast the negative results from the Chaos classification. The chosen model was therefore configured to learn two classes, ’melanoma’ and’nevus’. This prediction target yielded a more positive result as the validation accuracywas close to 85%. However, the corresponding confusion matrix showed that these resultsare not trustworthy. It is inconclusive whether the negative results from the Chaos classification stem from thechosen approach or if the data set was insufficient for the task-difficulty. We propose adjustments to the data set for future work which could disclose if the outlined approach isviable or not.

Over 30 000 liver, abdomen, and thigh slices have been acquired by computed tomography for the SCAPIS and IGT study. To utilise the full potential of the large cohort and enable statistical pixel-wise body composition analysis and visualisation of associations with other biomarkers, a point-to-point correspondence between the scans is needed. For this purpose, an inter-subject image registration pipeline that combines the low-level information from CT images with high-level information from segmentation masks have been developed. It uses tissue-specific regularisation and processes images efficiently. The pipeline was used to deform 4603 CT scans of each slice into a respective common reference space in less than 30 hours. All but the ribs in the liver slices and the intra abdominal region of the abdomen were generally registered correctly. Vector and intensity magnitude errors indicating inverse consistency were on average less than 2.5 mm and 40 Hounsfield units respectively. The method may serve as a starting point for statistical pixel-wise body composition analysis, its association with non-imaging data, as well as saliency mapping analysis of the three-slice CT scans from the large SCAPIS and IGT cohorts.

Prosthetic heart valves (PHVs) are routinely used to replace defective native heart valves in patients suffering from valvular heart diseases. While PHVs are life-saving, they have limitations in performance and durability. Therefore, it is crucial to rigorously test and evaluate their designs before their implantation. PHVs are commonly examined using cardiovascular testing equipment that measures the hemodynamic characteristics of the valves, while also providing the opportunity for their visual assessment by collecting high-quality videos. Such visual data, obtained during mechanical simulations, are typically assessed by human experts, which is a tedious and error-prone task. Automatic assessment of PHVs from video data is possible, however, there are some challenges that need to be addressed. The evolution of the valve orifice area during one cardiac cycle is one of the key quality metrics for PHV visual assessment. Very fast motion of the valve’s leaflets is one of the challenges while dealing with the visual data. Nevertheless, the more important issue lies in the orifice being partly occluded by the inner side of the leaflets or inaccurately depicted due to its transparency. This issue has not been addressed in the literature. In the first part of the thesis, a novel orifice area segmentation algorithm is proposed for automatic quantitative performance analysis of PHVs, based on the leaflet free edges to accurately extract the actual orifice area. The video frames, recorded by a high-speed digital camera during in vitro simulations, are used to obtain an initial estimate of the orifice area using active contouring methods. This initial estimate is then refined to detect leaflet free edges via a curve extension scheme and considering brightness and smoothness criteria. Both of the developed algorithms are later modified for addressing challenges related to the fast motion of leaflets, automatic detection of the beginning of a cycle, and overly bright spots and narrow areas. Evaluation on several cases including three different PHVs and with different video qualities demonstrated the effectiveness of the proposed approach and adjustments in detecting valve leaflet free edges and extraction of the actual orifice area. The proposed method significantly outperforms a baseline algorithm both in terms of valve design and computer vision evaluation metrics. It can also cope with lower quality videos and is better at processing frames with a very small opening, which is a very crucial quality for determining the malfunctions related to improper closing of the valves. In the second part of the thesis, the above-mentioned segmented orifice area is used for the durability estimation of the prosthetic heart valves. More than 50% of PHVs encounter a structural failure within 15 years post-implantation mostly because of the excessive localized forces on some areas. We perform a computer vision (CV)-based analysis of the visual symmetry of valve leaflet motion and investigate its correlation with the functional symmetry of the valve. We hypothesize that an asymmetry in the valve leaflet motion will generate an asymmetry in the flow patterns, resulting in added local stress and forces on some of the leaflets. Two pair-wise leaflet symmetry scores are proposed based on diagonals of orthogonal projection matrices (DOPM) and dynamic time warping (DTW) techniques. The proposed symmetry score profiles are compared with fluid dynamic parameters (vorticity and velocity values) at the leaflet borders, obtained from valve-specific numerical simulations. Experiments on four cases including different tricuspid PHV designs yielded promising results, with DTW scores showing good coherence with respect to the simulations, which confirms our hypothesis. The established link between visual and functional symmetry opens the door for durability estimation of prosthetic heart valves using computer vision techniques. / Graduate