EuroVisShort

Permanent URI for this community

https://diglib.eg.org/handle/10.2312/987

Browse

Now showing 1 - 11 of 11

CellTrackVis: Analyzing the Performance of Cell Tracking Algorithms
(The Eurographics Association, 2022) Li, Weimin; Zhang, Xiang; Stern, Alan; Birtwistle, Marc; Iuricich, Federico; Agus, Marco; Aigner, Wolfgang; Hoellt, Thomas
Live-cell imaging is a common data acquisition technique used by biologists to analyze cell behavior. Since manually tracking cells in a video sequence is extremely time-consuming, many automatic algorithms have been developed in the last twenty years to accomplish the task. However, none of these algorithms can yet claim robust tracking performance at the varying of acquisition conditions (e.g., cell type, acquisition device, cell treatments). While many visualization tools exist to help with cell behavior analysis, there are no tools to help with the algorithm's validation. This paper proposes CellTrackVis, a new visualization tool for evaluating cell tracking algorithms. CellTrackVis allows comparing automatically generated cell tracks with ground truth data to help biologists select the best-suited algorithm for their experimental pipeline. Moreover, CellTackVis can be used as a debugging tool while developing a new cell tracking algorithm to investigate where, when, and why each tracking error occurred.
ChemoExplorer: A Dashboard for the Visual Analysis of Chemotherapy Response in Breast Cancer Patients
(The Eurographics Association, 2018) Karall, Nikolaus; Gröller, Eduard; Raidou, Renata Georgia; Jimmy Johansson and Filip Sadlo and Tobias Schreck
In breast cancer chemotherapy treatment, different alternative strategies can be employed. Clinical researchers working on the optimization of chemotherapy strategies need to analyze the progress of the treatment and to understand how different groups of patients respond to selected therapies. This is a challenging task, because of the multitude of imaging and non-imaging health record data involved. We, hereby, introduce a web-based dashboard that facilitates the comparison and analysis of publicly available breast cancer chemotherapy response data, consisting of a follow-up study of 63 patients. Each patient received one of two available therapeutic strategies and their treatment response was documented. Our dashboard provides an initial basis for clinical researchers working on chemotherapy optimization, to analyze the progress of treatment and to compare the response of (groups of) patients with distinct treatment characteristics. Our approach consists of multiple linked representations that provide interactive views on different aspects of the available imaging and non-imaging data. To illustrate the functionality of the ChemoExplorer, we conducted a usage scenario that shows the initial results of our work.
Comparative Visual Analysis of Pelvic Organ Segmentations
(The Eurographics Association, 2018) Reiter, Oliver; Breeuwer, Marcel; Gröller, Eduard; Raidou, Renata Georgia; Jimmy Johansson and Filip Sadlo and Tobias Schreck
In prostate cancer treatment, automatic segmentations of the pelvic organs are often used as input to radiotherapy planning systems. However, natural anatomical variability of the involved organs is a common reason, for which segmentation algorithms fail, introducing errors in the radiotherapy treatment procedure, as well. Understanding how the shape and size of these organs affect the accuracy of segmentation is of major importance for developers of segmentation algorithms. However, current means of exploration and analysis provide limited insight. In this work, we discuss the design and implementation of a web-based framework, which enables easy exploration and detailed analysis of shape variability, and allows the intended users - i.e., segmentation experts - to generate hypotheses in relation to the performance of the involved algorithms. Our proposed approach was tested with segmentation meshes from a small cohort of 17 patients. Each mesh consists of four pelvic organs and two organ interfaces, which are labeled and have per-triangle correspondences. A usage scenario and an initial informal evaluation with a segmentation expert demonstrate that our framework allows the developers of the algorithms to quickly identify inaccurately segmented organs and to deliberate about the relation of variability to anatomical features and segmentation quality.
Comprehensive Visualization of Longitudinal Patient Data for the Dermatological Oncological Tumor Board
(The Eurographics Association, 2020) Steinhauer, Nastasja; Hörbrugger, Marc; Braun, Andreas Dominik; Tüting, Thomas; Oeltze-Jafra, Steffen; Müller, Juliane; Kerren, Andreas and Garth, Christoph and Marai, G. Elisabeta
In multidisciplinary oncological team meetings for patient-specific treatment decision-making, so-called tumor boards, usually one physician introduces a patient case verbally and proposes an initial therapy recommendation. This is followed by a short collaborative discussion of the recommendation's suitability. While patient-related image data, such as CT and MR scans, are displayed during the discussion, clinical patient data must be memorized from the introduction or repeatedly inquired by the participating domain experts. To support physicians in this concern, we propose a comprehensive visualization of longitudinal patient-specific information entities during case introduction and discussion. Our visual approach advances over existing work by simultaneously providing an overview of the current patient status as well as of previous therapy measures and their effects on the status. The latter assists in relating the currently proposed recommendation to the previous treatment measures and the related patient status. The visualization has been designed in close collaboration with dermatologists and oncologists aiming at a comprehensive yet easily comprehensible presentation of relevant patient-data and minimal user interaction. The usability and clinical relevance of the prototypical implementation of our visual approach have been evaluated in a qualitative user study with five domain experts based on real anonymized data of melanoma patients.
The Design Space of SparkWords
(The Eurographics Association, 2019) Brath, Richard; MacMurchy, Peter; Banissi, Ebad; Johansson, Jimmy and Sadlo, Filip and Marai, G. Elisabeta
The design space of SparkWords is consistently-sized words; embedded in sequential text (e.g. prose, lists); embellished by adding data, including categoric, ordered or quantitative data, that is encoded by a variety of attributes (singular or multiple) applied to words or letters. The breadth of the design space is illustrated with historic examples and novel implementations.
MOOCad: Visual Analysis of Anomalous Learning Activities in Massive Open Online Courses
(The Eurographics Association, 2019) Mu, Xing; Xu, Ke; Chen, Qing; Du, Fan; Wang, Yun; Qu, Huamin; Johansson, Jimmy and Sadlo, Filip and Marai, G. Elisabeta
The research on Massive Open Online Course (MOOC) has mushroomed worldwide due to the technical revolution and its unprecedented enrollments. Existing work mainly focuses on performance prediction, content recommendation, and learning behavior summarization. However, finding anomalous learning activities in MOOC data has posed special challenges and requires providing a clear definition of anomalous behavior, analyzing the multifaceted learning sequence data, and interpreting anomalies at different scales. In this paper, we present a novel visual analytics system, MOOCad, for exploring anomalous learning patterns and their clustering in MOOC data. The system integrates an anomaly detection algorithm to cluster learning sequences of MOOC learners into staged-based groups. Moreover, it allows interactive anomaly detection between and within groups on the basis of semantic and interpretable group-wise data summaries. We demonstrate the effectiveness of MOOCad via an in-depth interview with a MOOC lecturer with real-world course data.
Pressure-based Vortex Extraction in Cardiac 4D PC-MRI Blood Flow Data
(The Eurographics Association, 2018) Köhler, Benjamin; Grothoff, Matthias; Gutberlet, Matthias; Preim, Bernhard; Jimmy Johansson and Filip Sadlo and Tobias Schreck
We propose a technique for vortex extraction in cardiac 4D PC-MRI blood flow data that employs an intravascular, relative pressure calculation. The method is easy to implement, runs fully automatically, and requires no user-defined parameters. We qualitatively evaluated 100+ datasets of the aorta, pulmonary artery, or left ventricle from healthy volunteers as well as from patients acquired with different MR scanners. In all cases, the results suffer from significantly less noise than comparable approaches using the common λ 2 vortex criterion.
RiskFix: Supporting Expert Validation of Predictive Timeseries Models in High-Intensity Settings
(The Eurographics Association, 2023) Morgenshtern, Gabriela; Verma, Arnav; Tonekaboni, Sana; Greer, Robert; Bernard, Jürgen; Mazwi, Mjaye; Goldenberg, Anna; Chevalier, Fanny; Hoellt, Thomas; Aigner, Wolfgang; Wang, Bei
Many real-world machine learning workflows exist in longitudinal, interactive machine learning (ML) settings. This longitudinal nature is often due to incremental increasing of data, e.g., in clinical settings, where observations about patients evolve over their care period. Additionally, experts may become a bottleneck in the workflow, as their limited availability, combined with their role as human oracles, often leads to a lack of ground truth data. In such cases where ground truth data is small, the validation of interactive machine learning workflows relies on domain experts. Only those humans can assess the validity of a model prediction, especially in new situations that have been covered only weakly by available training data. Based on our experiences working with domain experts of a pediatric hospital's intensive care unit, we derive requirements for the design of support interfaces for the validation of interactive ML workflows in fast-paced, high-intensity environments. We present RiskFix, a software package optimized for the validation workflow of domain experts of such contexts. RiskFix is adapted to the cognitive resources and needs of domain experts in validating and giving feedback to the model. Also, RiskFix supports data scientists in their model-building work, with appropriate data structuring for the re-calibration (and possible retraining) of ML models.
TapVis: A Data Visualization Approach for Assessment of Alternating Tapping Performance in Patients with Parkinson's Disease
(The Eurographics Association, 2018) Jusufi, Ilir; Memedi, Mevludin; Nyholm, Dag; Jimmy Johansson and Filip Sadlo and Tobias Schreck
Advancements in telemedicine have been helpful for frequent monitoring of patients with Parkinson's disease (PD) from remote locations and assessment of their individual symptoms and treatment-related complications. These data can be useful for helping clinicians to interpret symptom states and individually tailor the treatments by visualizing the physiological information collected by sensor-based systems. In this paper we present a visualization metaphor that represents symptom information of PD patients during tapping tests performed with a smartphone. The metaphor has been developed and evaluated with a clinician. It enabled the clinician to observe fine motor impairments and identify motor fluctuations regarding several movement aspects of patients that perform the tests from their homes.
Visualising Collocation for Close Writing
(The Eurographics Association, 2020) Roberts, Jonathan C.; Butcher, Peter W. S.; Lew, Robert; Rees, Geraint Paul; Sharma, Nirwan; Frankenberg-Garcia, Ana; Kerren, Andreas and Garth, Christoph and Marai, G. Elisabeta
We present how we have developed a visualisation tool and text editor to display collocations for the purpose of close writing. Collocations are words that combine together in a natural way. Our design study approach brought together a collaboration of experts in lexicography, language learning, and visualisation, starting with low-fidelity prototypes before developing fuller functional systems. We studied the challenge of how to visualise collocations, such to help language learners write more effectively. We have co-created (i) an expert-curated dataset of over 30,000 collocations, (ii) developed a text-editor which performs word analysis, and recommends collocations, and (iii) created several in-situ visualisations linked to the editor, to help users visualise and lookup collocations, and view example sentences. Every stage of development has been evaluated with language learners and other potential users, which has positively improved its design and functionality.
Viz-Blocks: Building Visualizations and Documents in the Browser
(The Eurographics Association, 2019) McNeill, Graham; Hale, Scott A.; Johansson, Jimmy and Sadlo, Filip and Marai, G. Elisabeta
Viz-Blocks is a simple browser-based UI for data exploration and document creation. It incorporates the Vega-Lite grammar of graphics for standard visualizations (including multiple views and interaction) whereas 'code blocks' provide the full power of the JavaScript ecosystem for creating custom visualizations and other bespoke content. Visualizations are treated as reusable 'blocks' that are easily created, modified and compared during exploration. When preparing results for dissemination, visualizations can be customized and combined with Markdown and image blocks to produce a single or multi-page HTML document that is easily styled and exported. Viz-blocks was designed in consultation with academics, students and policy makers to bridge the gap between visualization tools and more traditional document-authoring tools. The application is aimed at a wide audience: the lightweight, hybrid UI allows all users to access the core functionality, while experienced users can take advantage of code-blocks and the option to use advanced features of Vega-Lite via JSON/YAML snippets.

Browse

Browsing EuroVisShort by Subject "Applied computing"

Results Per Page

Sort Options