Logo des Repositoriums

Tagungsband MuC 2023

https://dl.gi.de/handle/20.500.12116/41975

Der Tagungsband der MuC 2023 ist in der DL der ACM veröffentlicht worden - unter https://dl.acm.org/doi/proceedings/10.1145/3603555

Auflistung nach:

Auflistung Tagungsband MuC 2023 nach Schlagwort "-"

1 - 3 von 3
  • Konferenzbeitrag
    Appealing but Potentially Biasing - Investigation of the Visual Representation of Segmentation Predictions by AI Recommender Systems for Medical Decision Making
    (Mensch und Computer 2023 - Tagungsband, 2023) Ammeling, Jonas; Manger, Carina; Kwaka, Elias; Krügel, Sebastian; Uhl, Matthias; Kießig, Angelika; Fritz, Alexis; Ganz, Jonathan; Riener, Andreas; Bertram, Christof A.; Breininger, Katharina; Aubreville, Marc
    Artificial intelligence (AI)-based recommender systems can help to improve efficiency and accuracy in medical decision making. Yet, it has been shown that a recommendation given by an algorithm can influence the human expert responsible for the decision. The strength and direction of this bias, induced by a computer-aided diagnosis workflow, can be influenced by the visual representation of the results. This study focuses on evaluating four frequently used visualization types (bounding box, segmentation mask, segmentation contour, and heatmap) for displaying segmentation results of medical data. A group of 24 medical experts specializing in pathology and radiology participated in the evaluation, assessing the subjective appeal of these visualizations. The study evaluated the pragmatic and hedonic quality of the visualizations based on a standardized questionnaire and specific criteria relevant to medical decision making. The findings indicate that the heatmap received the highest ratings for non-task-oriented aspects of the user experience. However, it exhibited significant inconsistencies among experts concerning task-oriented aspects and was perceived as the most biasing visualization type. On the other hand, the segmentation contour consistently received high ratings across various subscales. The results of the study contribute to better alignment between visualization techniques and user requirements for the development of future AI-based recommender systems.
  • Konferenzbeitrag
    CoShare: a Multi-Pointer Collaborative Screen Sharing Tool
    (Mensch und Computer 2023 - Tagungsband, 2023) Emmert, Martina; Schmid, Andreas; Wimmer, Raphael; Henze, Niels
    Existing tools for screen sharing and remote control only allow a single user to interact with a system while others are watching. Collaborative editors and whiteboards allow multiple users to work simultaneously, but only offer a limited set of tools. With CoShare, we combine both concepts into a screen sharing tool that gives remote viewers a mouse pointer and a text cursor so that they can seamlessly collaborate within the same desktop environment. We have developed a proof-of-concept implementation that leverages Linux’ multi-pointer support so users can control applications in parallel. It also allows limited sharing of clipboard and dragging files from the remote viewer’s desktop into the video-streamed desktop. In focus groups we gathered user requirements regarding privacy, control, and communication. A qualitative lab study identified further areas for improvement and demonstrated CoShare’s utility.
  • Konferenzbeitrag
    Enhancing the Supervision of Out-of-View Robots: A Study on Multimodal Feedback and Monitoring Screens
    (Mensch und Computer 2023 - Tagungsband, 2023) Kassem, Khaled; Shahu, Ambika; Tüchler, Christina; Wintersberger, Philipp; Michahelles, Florian
    Objective: investigating the effect of two support methods (multimodal feedback, monitoring screens, and a combination of both) on human dual-task performance, cognitive workload, and user experience when supervising an out-of-sight autonomous robot. Method: A 2x2 within-group user study was conducted in VR with 26 participants involving a cognitive-cognitive dual-task setting. Participants had to simultaneously solve math problems and supervise the robot. Different support methods were provided: multimodal feedback, a screen showing real-time robot activity, and a combination of both. Objective performance metrics and subjective feedback on cognitive load and user experience were collected using standard questionnaires. Data were statistically analyzed, and thematic analysis was performed on post-study debriefing interviews. Results: The support methods improved overall user experience and positively impacted robot collaboration performance while decreasing math task performance. Cognitive load was unaffected. Multimodal feedback with a monitoring screen was perceived as the most helpful. Conclusion: The results suggest that multimodal feedback can improve user experience and improve supervision, but may partially decrease primary task performance. The findings highlight the importance of examining the effect of support methods in specific situations, depending on task priority.