In modern neurosurgical practice, a surgeon can see a patient’s fiber tracts (nerve tracts) on a monitor in the operating room. This design study investigates the benefit of adding the uncertainty of the tracts and aims to improve the surgeon’s orientation while reducing visual clutter.

Based on an interview with a neurosurgeon and our experience during a surgery visit, we designed, implemented, and evaluated a new visualization method for intraoperative use. We present two new visualization techniques: a partial hull that appears when a surgeon approaches it and a distance meter that serves as a warning tool.

We conclude that there is a need for the intraoperative visualization of fiber uncertainty and recommend doing clinical trials using these methods.

A craniotomy is a procedure were a neurosurgeon has to open the skull to gain direct access to the brain. When a brain tumor has to be removed from a patient, the craniotomy position is of great importance. This mostly defines the access path from the skull surface to the tumor and thus also what healthy brain tissue will be removed to access to the tumor itself. To minimize the amount of important brain structures that are removed, the craniotomy has to be carefully planned. This is a complex procedure, where a neurosurgeon is required to mentally reconstruct spatial relations of important brain structures to avoid these as much as possible.

We propose a visualization using augmented reality which may assist in the planning of a craniotomy. In this visualization the goal is to show important brain structures aligned with the physical position of the patient. This should allow better perception of the spatial relations of these structures and thus assist the neurosurgeon. Additionally to the visualization of the structures, we created a heat map that is projected on top of the skull. This should give a quick overview of in which areas there are many important structures between the tumor and the skull surface, and should therefore be avoided.

User studies were conducted amongst neurosurgeons and surgeons from other fields to evaluate the proposed visualization. We found that many of the participants indeed thought that the visualization can assist in surgery. For the specific case of craniotomy planning, several improvements have to be made on the heat map before it can be useful. Nevertheless, the visualization of the structures in itself can assist neurosurgeons in the planning of a craniotomy. Although more work has be performed at practical aspects of the visualization to make it ready for clinical experiments.

Code reviews are a widely adopted practice in software engineering that is proven to increase the quality of the code. Despite its evolution in the last decade, it still presents a number of challenges, such as understanding the changeset in review. In this thesis we research the usage of Software Visualisation paradigms to aid reviewers in the change-understanding process with a tool-based approach. Based on a survey of the code change visualisation and code navigation research areas, we devise a set of candidate prototypes of a cognitive support review tool, which we iteratively refine involving developers in the process. Through an online survey, we select one of them and build CHANGEVIZ, the implementation of our novel code review environment. The effectiveness of our approach is validated with a preliminary experiment in which developers perform change-review tasks in our review environment. The results suggest that the features incorporated by our tool are valuable for reviewers.

Visualizing runners trajectory from video data is not straightforward because the video data does not contain the explicit information of which runners appear in the video. Only the visual information related to the runner, such as runner’s unique ID (called bib number), is available. To this end, we propose two automatic runner detection methods, i.e. scene text detection which identifies the runners by detecting their bib number and person re-identification which detects the runners based on their appearance. To evaluate the proposed methods, we create a ground truth database from the video dataset, which consists of video and frame interval information where the runners appear. The video dataset was recorded by nine cameras at different locations during the Campus Run 2018 event. The experimental evidence shows that the scene text recognition method achieves up to 74.05 for F1-score and person re identification achieves up to 87.76 for F1-score. To conclude, we find that the person re-identification method outperforms the scene text recognition method.

Both MRI and CT imaging are commonly used and combined in medical imaging because of their complementary information about soft tissue and bone respectively. However, CT imaging relies on harmful ionizing radiation. Thus medical imaging scientists are working on transferring harmless MRI scans to CT-like images where using deep learning methods are the current state of the art. Most deep learning methods however require large amounts of paired data to train the transfer systems, which is why we try to perform the transfer using an unpaired method, CycleGANs. Using ex-vivo dog hip MRI and CT scan pairs we trained a CycleGAN and a U-Net using both registered and poorly registered data. We also further investigated the effect of discriminator’s receptive field size in a CycleGAN as well as introduced a novel NMI loss component to its generators. Performance of said systems was evaluated by calculating MAE and Dice scores. We show that CycleGAN's performance is sensitive to its discriminator's receptive field size and can even generate unwanted structrures in the CT-like images. Also, we show that adding NMI loss component with the right weight can improve the system’s performance. The results of comparing the pairwise and unpaired method show that a CycleGAN will only outperform the U-Net when transferring poorly registered data. This suggest that, whenever paired registered data is available, it is better to use a pairwise transfer approach. However, the results of the experiment where the NMI loss component was added suggest that transfer from MRI scans to CT-like images can become more accurate and encourage further research. Finally, this means in certain cases CT scans could be avoided, providing a more harmless medical imaging future.

Dimensionality reduction algorithms transform high-dimensional datasets with many attributes per observation into lower-dimensional representations (called embeddings) such that the structure of the dataset is maintained as well as possible. In this research, the use of Virtual Reality (VR) to analyse these embeddings has been evaluated and compared to the analysis on a desktop computer. The rationale for using VR is two-fold: three-dimensional embeddings generally better preserve the structure of a high-dimensional dataset than two-dimensional embeddings and the analysis of three-dimensional embeddings is difficult on desktop monitors. A user study (n=29) has been conducted in which participants performed the common analysis task of cluster identification. The task has been performed using a two-dimensional embedding on a desktop computer, a three-dimensional embedding on a desktop computer and a three-dimensional embedding in Virtual Reality. On average, participants that had at least used VR once before could better and more consistently identify clusters in the VR experiments compared to other methods. Participants found it easier to analyse a three-dimensional embedding in VR compared to analysing it on a desktop computer.

Perivascular spaces (PVS) visible on MRI are currently emerging as an important potential neuroimaging marker for several pathologies in the brain like Alzheimer’s disease and cerebral small vessel disease. PVS are fluid-filled spaces surrounding vessels as they enter the brain. Although PVS are normally not noticeable on MRI scans acquired at clinical field strengths, when these spaces increase in size they become increasingly visible and quantifiable. To study these spaces it is important to have a robust method for quantifying PVS. Manual quantification of PVS is challenging, time-consuming and subject to observer bias due to the difficulty of distinguishing PVS from mimics and the large number of PVS that can occur in MRI scans. Many promising (semi-)automated methods have been proposed recently to decrease annotation time and intra- and inter-observer variability while providing more information about EPVS. However there are still various limitations in the current methods that need to be overcome.
An important limitation is that most of the methods are based on elaborate preprocessing steps, feature extraction and heuristic fine-tuning of parameters, making the use of these methods on new datasets cumbersome. Furthermore the majority of the currently proposed methods have been evaluated on small sets of barely 30 images, as most of these methods aim to segment PVS and require voxel-wise annotations for evaluation. In this thesis we propose a method for automated detection of perivascular spaces that combines a convolutional neural network and geodesic distance transform (GDT). We propose to use dot annotations instead of voxel-wise segmentations as this is less time-consuming than fully segmenting PVS while still providing the location of PVS. This enables us to use a considerably larger dataset with ground truth locations than is used in all previously proposed (semi-)automatic methods that provide the location of PVS. We investigated two approaches of using geodesic distance transform to optimize the CNN to detect PVS. The first approach focuses on optimizing the CNN for voxel-wise regression of the geodesic distance map (GDM) computed from the dots and the intensity image. The second approach aims to predict segmentations of the PVS using a CNN that is trained on approximated segmentations obtained by thresholding GDMs. We use 1202 proton density-weighted (PDw) MRI scans to develop our methods and 1000 other scans are used to evaluate the performance of the methods. We show that our methods match human intra-rater performance on detecting PVS without the need for any user interaction. Additionally we show that GDMs are extremely useful for capturing complex morphologies when computed from dot annotations. Our experiments indicate that GDMs can be used to provide valuable additional information to CNNs during training.