VCBM 2022: Eurographics Workshop on Visual Computing for Biology and Medicine

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/2633211

Browse

Now showing 1 - 16 of 16

COMFIS - Comparative Visualization of Simulated Medical Flow Data
(The Eurographics Association, 2022) Meuschke, Monique; Voß, Samuel; Eulzer, Pepe; Janiga, Gabor; Arens, Christoph; Wickenhöfer, Ralph; Preim, Bernhard; Lawonn, Kai; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Simulations of human blood and airflow are playing an increasing role in personalized medicine. Comparing flow data of different treatment scenarios or before and after an intervention is important to assess treatment options and success. However, existing visualization tools are either designed for the evaluation of a single data set or limit the comparison to a few partial aspects such as scalar fields defined on the vessel wall or internal flow patterns. Therefore, we present COMFIS, a system for the comparative visual analysis of two simulated medical flow data sets, e.g. before and after an intervention. We combine various visualization and interaction methods for comparing different aspects of the underlying, often time-dependent data. These include comparative views of different scalar fields defined on the vessel/mucous wall, comparative depictions of the underlying volume data, and comparisons of flow patterns. We evaluated COMFIS with CFD engineers and medical experts, who were able to efficiently find interesting data insights that help to assess treatment options.
Distance Visualizations for Vascular Structures in Desktop and VR: Overview and Implementation
(The Eurographics Association, 2022) Hombeck, Jan; Meuschke, Monique; Lieb, Simon; Lichtenberg, Nils; Datta, Rabi; Krone, Michael; Hansen, Christian; Preim, Bernhard; Lawonn, Kai; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
The role of expressive surface visualizations in rendering vascular structures has seen an increased impact over the last years. Surface visualizations provide an overview of complex anatomical structures and support treatment planning as well as medical education. To support decision-making, physicians need visualizations that depict anatomical structures and their spatial relations to each other, i.e., well perceivable visual encodings of egocentric and endocentric distances. We give an overview of common techniques for encoding distance information of 3D vessel surfaces. We also provide an implementation of all the visualizations presented as a starting point for other researchers. Therefore, we provide a Unity environment for each visualization, as well as implementation instructions. Thirteen different visualizations are included in this work, which can be divided into fundamental, surface-based, auxiliary and illustrative visualizations.
Eurographics Workshop on Visual Computing for Biology and Medicine: Frontmatter
(The Eurographics Association, 2022) Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
HistoContours: a Framework for Visual Annotation of Histopathology Whole Slide Images
(The Eurographics Association, 2022) Al-Thelaya, Khaled; Joad, Faaiz; Gilal, Nauman Ullah; Mifsud, William; Pintore, Giovanni; Gobbetti, Enrico; Agus, Marco; Schneider, Jens; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
We present an end-to-end framework for histopathological analysis of whole slide images (WSIs). Our framework uses deep learning-based localization & classification of cell nuclei followed by spatial data aggregation to propagate classes of sparsely distributed nuclei across the entire slide. We use YOLO (''You Only Look Once'') for localization instead of more costly segmentation approaches and show that using HistAuGAN boosts its performance. YOLO finds bounding boxes around nuclei at good accuracy, but the classification accuracy can be improved by other methods. To this end, we extract patches around nuclei from the WSI and consider models from the SqueezeNet, ResNet, and EfficientNet families for classification. Where we do not achieve a clear separation between highest and second-highest softmax activation of the classifier, we use YOLO's output as a secondary vote. The result is a sparse annotation of the WSI, which we turn dense by using kernel density estimation. The result is a full vector of per pixel probabilities for each class of nucleus we consider. This allows us to visualize our results using both color-coding and isocontouring, reducing visual clutter. Our novel nuclei-to-tissue coupling allows histopathologists to work at both the nucleus and the tissue level, a feature appreciated by domain experts in a qualitative user study.
Is there a Tornado in Alex's Blood Flow? A Case Study for Narrative Medical Visualization
(The Eurographics Association, 2022) Kleinau, Anna; Stupak, Evgenia; Mörth, Eric; Garrison, Laura A.; Mittenentzwei, Sarah; Smit, Noeska N.; Lawonn, Kai; Bruckner, Stefan; Gutberlet, Matthias; Preim, Bernhard; Meuschke, Monique; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Narrative visualization advantageously combines storytelling with new media formats and techniques, like interactivity, to create improved learning experiences. In medicine, it has the potential to improve patient understanding of diagnostic procedures and treatment options, promote confidence, reduce anxiety, and support informed decision-making. However, limited scientific research has been conducted regarding the use of narrative visualization in medicine. To explore the value of narrative visualization in this domain, we introduce a data-driven story to inform a broad audience about the usage of measured blood flow data to diagnose and treat cardiovascular diseases. The focus of the story is on blood flow vortices in the aorta, with which imaging technique they are examined, and why they can be dangerous. In an interdisciplinary team, we define the main contents of the story and the resulting design questions. We sketch the iterative design process and implement the story based on two genres. In a between-subject study, we evaluate the suitability and understandability of the story and the influence of different navigation concepts on user experience. Finally, we discuss reusable concepts for further narrative medical visualization projects.
Learning Anatomy through Shared Virtual Reality
(The Eurographics Association, 2022) Reyes-Cabrera, José Juan; Santana-Núñez, José Miguel; Trujillo-Pino, Agustín; Maynar, Manuel; Rodriguez-Florido, Miguel Angel; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Virtual reality (VR) is a powerful tool for educational purposes. In this work, we present a VR application for learning anatomy, focusing on the cardiac system in this early stage. Our application proposes that medical students put together parts of the human anatomy and check their performance at this task. The system also features a shared-VR mode, in which two or more students can work together, or can even be joined by a medical professor. In this paper, we briefly describe our new approach to medicine teaching and show promising results for further development. In addition, we have tested our application with students at the Medical School, and we are confident that this application will improve their training.
Multi-modal 3D Image Registration Using Interactive Voxel Grid Deformation and Rendering
(The Eurographics Association, 2022) Richard, Thomas; Chastagnier, Yan; Szabo, Vivien; Chalard, Kevin; Summa, Brian; Thiery, Jean-Marc; Boubekeur, Tamy; Faraj, Noura; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
We introduce a novel multi-modal 3D image registration framework based on 3D user-guided deformation of both volume's shape and intensity values. Being able to apply deformations in 3D gives access to a wide new range of interactions allowing for the registration of images from any acquisition method and of any organ, complete or partial. Our framework uses a state of the art 3D volume rendering method for real-time feedback on the registration accuracy as well as the image deformation. We propose a novel methodological variation to accurately display 3D segmented voxel grids, which is a requirement in a registration context for visualizing a segmented atlas. Our pipeline is implemented in an open-source software (available via GitHub) and was directly used by biologists for registration of mouse brain model autofluorescence acquisition on the Allen Brain Atlas. The latter mapping allows them to retrieve regions of interest properly identified on the segmented atlas in acquired brain datasets and therefore extract only high-resolution images of those areas, avoiding the creation of images too large to be processed.
MuSIC: Multi-Sequential Interactive Co-Registration for Cancer Imaging Data based on Segmentation Masks
(The Eurographics Association, 2022) Eichner, Tanja; Mörth, Eric; Wagner-Larsen, Kari S.; Lura, Njål; Haldorsen, Ingfrid S.; Gröller, Eduard; Bruckner, Stefan; Smit, Noeska N.; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
In gynecologic cancer imaging, multiple magnetic resonance imaging (MRI) sequences are acquired per patient to reveal different tissue characteristics. However, after image acquisition, the anatomical structures can be misaligned in the various sequences due to changing patient location in the scanner and organ movements. The co-registration process aims to align the sequences to allow for multi-sequential tumor imaging analysis. However, automatic co-registration often leads to unsatisfying results. To address this problem, we propose the web-based application MuSIC (Multi-Sequential Interactive Co-registration). The approach allows medical experts to co-register multiple sequences simultaneously based on a pre-defined segmentation mask generated for one of the sequences. Our contributions lie in our proposed workflow. First, a shape matching algorithm based on dual annealing searches for the tumor position in each sequence. The user can then interactively adapt the proposed segmentation positions if needed. During this procedure, we include a multi-modal magic lens visualization for visual quality assessment. Then, we register the volumes based on the segmentation mask positions. We allow for both rigid and deformable registration. Finally, we conducted a usability analysis with seven medical and machine learning experts to verify the utility of our approach. Our participants highly appreciate the multi-sequential setup and see themselves using MuSIC in the future.
Perceptual Evaluation of Common Line Variables for Displaying Uncertainty on Molecular Surfaces
(The Eurographics Association, 2022) Sterzik, Anna; Lichtenberg, Nils; Krone, Michael; Cunningham, Douglas W.; Lawonn, Kai; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Data are often subject to some degree of uncertainty, whether aleatory or epistemic. This applies both to experimental data acquired with sensors as well as to simulation data. Displaying these data and their uncertainty faithfully is crucial for gaining knowledge. Specifically, the effective communication of the uncertainty can influence the interpretation of the data and the users' trust in the visualization. However, uncertainty-aware visualization has gotten little attention in molecular visualization. When using the established molecular representations, the physicochemical attributes of the molecular data usually already occupy the common visual channels like shape, size, and color. Consequently, to encode uncertainty information, we need to open up another channel by using feature lines. Even though various line variables have been proposed for uncertainty visualizations, they have so far been primarily used for two-dimensional data and there has been little perceptual evaluation. Therefore, we conducted a perceptual study to determine the suitability of the line variables sketchiness, dashing, grayscale, and width for distinguishing several uncertainty values on molecular surfaces.
Polyp-Cavity Segmentation of Cold-Water Corals guided by Ambient Occlusion and Ambient Curvature
(The Eurographics Association, 2022) Schmitt, Kira; Titschack, Jürgen; Baum, Daniel; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
The segmentation of cavities in three-dimensional images of arbitrary objects is a difficult problem since the cavities are usually connected to the outside of the object without any difference in image intensity. Hence, the information whether a voxel belongs to a cavity or the outside needs to be derived from the ambient space. If a voxel is enclosed by object material, it is very likely that this voxel belongs to a cavity. However, there are dense structures where a voxel might still belong to the outside even though it is surrounded to a large degree by the object. This is, for example, the case for coral colonies. Therefore, additional information needs to be considered to distinguish between those cases. In this paper, we introduce the notion of ambient curvature, present an efficient way to compute it, and use it to segment coral polyp cavities by integrating it into the ambient occlusion framework. Moreover, we combine the ambient curvature with other ambient information in a Gaussian mixture model, trained from a few user scribbles, resulting in a significantly improved cavity segmentation. We showcase the superiority of our approach using four coral colonies of very different morphological types. While in this paper we restrict ourselves to coral data, we believe that the concept of ambient curvature is also useful for other data. Furthermore, our approach is not restricted to curvature but can be easily extended to exploit any properties given on an object's surface, thereby adjusting it to specific applications.
Predicting, Analyzing and Communicating Outcomes of COVID-19 Hospitalizations with Medical Images and Clinical Data
(The Eurographics Association, 2022) Stritzel, Oliver; Raidou, Renata Georgia; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
We propose PACO, a visual analytics framework to support the prediction, analysis, and communication of COVID-19 hospitalization outcomes. Although several real-world data sets about COVID-19 are openly available, most of the current research focuses on the detection of the disease. Until now, no previous work exists on combining insights from medical image data with knowledge extracted from clinical data, predicting the likelihood of an intensive care unit (ICU) visit, ventilation, or decease. Moreover, available literature has not yet focused on communicating such results to the broader society. To support the prediction, analysis and communication of the outcomes of COVID-19 hospitalizations on the basis of a publicly available data set comprising both electronic health data and medical image data [SSP∗21], we conduct the following three steps: (1) automated segmentation of the available X-ray images and processing of clinical data, (2) development of a model for the prediction of disease outcomes and a comparison to state-of-the-art prediction scores for both data sources, i.e., medical images and clinical data, and (3) the communication of outcomes to two different groups (i.e., clinical experts and the general population) through interactive dashboards. Preliminary results indicate that the prediction, analysis and communication of hospitalization outcomes is a significant topic in the context of COVID-19 prevention.
A Stratification Matrix Viewer for Analysis of Neural Network Data
(The Eurographics Association, 2022) Harth, Philipp; Vohra, Sumit; Udvary, Daniel; Oberlaender, Marcel; Hege, Hans-Christian; Baum, Daniel; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
The analysis of brain networks is central to neurobiological research. In this context the following tasks often arise: (1) understand the cellular composition of a reconstructed neural tissue volume to determine the nodes of the brain network; (2) quantify connectivity features statistically; and (3) compare these to predictions of mathematical models. We present a framework for interactive, visually supported accomplishment of these tasks. Its central component, the stratification matrix viewer, allows users to visualize the distribution of cellular and/or connectional properties of neurons at different levels of aggregation. We demonstrate its use in four case studies analyzing neural network data from the rat barrel cortex and human temporal cortex.
Studying the Effect of Tissue Properties on Radiofrequency Ablation by Visual Simulation Ensemble Analysis
(The Eurographics Association, 2022) Heimes, Karl; Evers, Marina; Gerrits, Tim; Gyawali, Sandeep; Sinden, David; Preusser, Tobias; Linsen, Lars; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Radiofrequency ablation is a minimally invasive, needle-based medical treatment to ablate tumors by heating due to absorption of radiofrequency electromagnetic waves. To ensure the complete target volume is destroyed, radiofrequency ablation simulations are required for treatment planning. However, the choice of tissue properties used as parameters during simulation induce a high uncertainty, as the tissue properties are strongly patient-dependent. To capture this uncertainty, a simulation ensemble can be created. Understanding the dependency of the simulation outcome on the input parameters helps to create improved simulation ensembles by focusing on the main sources of uncertainty and, thus, reducing computation costs. We present an interactive visual analysis tool for radiofrequency ablation simulation ensembles to target this objective. Spatial 2D and 3D visualizations allow for the comparison of ablation results of individual simulation runs and for the quantification of differences. Simulation runs can be interactively selected based on a parallel coordinates visualization of the parameter space. A 3D parameter space visualization allows for the analysis of the ablation outcome when altering a selected tissue property for the three tissue types involved in the ablation process. We discuss our approach with domain experts working on the development of new simulation models and demonstrate the usefulness of our approach for analyzing the influence of different tissue properties on radiofrequency ablations.
Understanding Graph Convolutional Networks to detect Brain Lesions from Stroke
(The Eurographics Association, 2022) Iporre-Rivas, Ariel; Scheuermann, Gerik; Gillmann, Christina; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Brain lesions derived from stroke episodes can result in disabilities for a patient. Therefore, the segmentation of brain lesions is an important task in neurology. Recently this task has been mainly tackled by machine learning approaches that demonstrated to be very successful. One of these approaches is Graph Convolutional Networks (GCN), where the input image is interpreted as a graph structure. As usual for neural networks, the interpretability is hard due to their black-box nature. We provide an interactive visualization of the activation inherent in the GCN, which is map from the original dataset. We visualize the activation values of the underlying graph network on top of the input image. We show the usability of our approach by applying it to a GCN that was trained on a real-world dataset.
Understanding the Impact of Statistical and Machine Learning Choices on Predictive Models for Radiotherapy
(The Eurographics Association, 2022) Böröndy, Ádám; Furmanová, Katarína; Raidou, Renata Georgia; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
During radiotherapy (RT) planning, an accurate description of the location and shape of the pelvic organs is a critical factor for the successful treatment of the patient. Yet, during treatment, the pelvis anatomy may differ significantly from the planning phase. A series of recent publications, such as PREVIS [FMCM∗21], have examined alternative approaches to analyzing and predicting pelvic organ variability of individual patients. These approaches are based on a combination of several statistical and machine learning methods, which have not been thoroughly and quantitatively evaluated within the scope of pelvic anatomical variability. Several of their design decisions could have an impact on the outcome of the predictive model. The goal of this work is to assess the impact of alternative choices, focusing mainly on the two key-aspects of shape description and clustering, to generate better predictions for new patients. The results of our assessment indicate that resolution-based descriptors provide more accurate and reliable organ representations than state-of-the-art approaches, while different clustering settings (distance metric and linkage) yield only slightly different clusters. Different clustering methods are able to provide comparable results, although when more shape variability is considered their results start to deviate. These results are valuable for understanding the impact of statistical and machine learning choices on the outcomes of predictive models for anatomical variability.
Visual Analytics to Assess Deep Learning Models for Cross-Modal Brain Tumor Segmentation
(The Eurographics Association, 2022) Magg, Caroline; Raidou, Renata Georgia; Renata G. Raidou; Björn Sommer; Torsten W. Kuhlen; Michael Krone; Thomas Schultz; Hsiang-Yun Wu
Accurate delineations of anatomically relevant structures are required for cancer treatment planning. Despite its accuracy, manual labeling is time-consuming and tedious-hence, the potential of automatic approaches, such as deep learning models, is being investigated. A promising trend in deep learning tumor segmentation is cross-modal domain adaptation, where knowledge learned on one source distribution (e.g., one modality) is transferred to another distribution. Yet, artificial intelligence (AI) engineers developing such models, need to thoroughly assess the robustness of their approaches, which demands a deep understanding of the model(s) behavior. In this paper, we propose a web-based visual analytics application that supports the visual assessment of the predictive performance of deep learning-based models built for cross-modal brain tumor segmentation. Our application supports the multi-level comparison of multiple models drilling from entire cohorts of patients down to individual slices, facilitates the analysis of the relationship between image-derived features and model performance, and enables the comparative exploration of the predictive outcomes of the models. All this is realized in an interactive interface with multiple linked views. We present three use cases, analyzing differences in deep learning segmentation approaches, the influence of the tumor size, and the relationship of other data set characteristics to the performance. From these scenarios, we discovered that the tumor size, i.e., both volumetric in 3D data and pixel count in 2D data, highly affects the model performance, as samples with small tumors often yield poorer results. Our approach is able to reveal the best algorithms and their optimal configurations to support AI engineers in obtaining more insights for the development of their segmentation models.

Browse

Browsing VCBM 2022: Eurographics Workshop on Visual Computing for Biology and Medicine by Title

Results Per Page

Sort Options