Publications
Detailed Information
An Interactive Visual Analytics Framework for Diagnostic Gaze Data on Volumetric Medical Images : 3차원 의료 영상 판독 시선 정보의 대화형 시각적 분석 프레임워크
Cited 0 time in
Web of Science
Cited 0 time in Scopus
- Authors
- Advisor
- 서진욱
- Major
- 공과대학 전기·컴퓨터공학부
- Issue Date
- 2016-02
- Publisher
- 서울대학교 대학원
- Keywords
- Eye tracking ; Gaze visualization ; Gaze pattern comparison ; Volumetric medical images ; Context-embedded interactive scatterplot ; Interactive temporal chart
- Description
- 학위논문 (박사)-- 서울대학교 대학원 : 전기·컴퓨터공학부, 2016. 2. 서진욱.
- Abstract
- We propose an interactive visual analytics framework for diagnostic gaze data on volumetric medical images. The framework is designed to compare gaze data from multiple readers with effective visualizations, which are tailored for volumetric gaze data with additional contextual information. Gaze pattern comparison is essential to understand how radiologists examine medical images and to identify factors influencing the examination. However, prior work on diagnostic gaze data using the medical images acquired from volumetric imaging systems (e.g., computed tomography or magnetic resonance imaging) showed a number of limitations in comparative analysis. In the diagnosis, radiologists scroll through a stack of images to get a 3D cognition of organs and lesions that resulting gaze patterns contain additional depth information compared to the gaze tracking study with 2D stimuli. As a result, the additional spatial dimension aggravated the complexity on visual representation of gaze data. A recent work proposed a visualization design based on direct volume rendering (DVR) for gaze patterns in volumetric images
however, effective and comprehensive gaze pattern comparison is still challenging due to lack of interactive visualization tools for comparative gaze analysis.
In this paper, we first present an effective visual representation, and propose an interactive analytics framework for multiple volumetric gaze data. We also take the challenge integrating crucial contextual information such as pupil size and windowing (i.e., adjusting brightness and contrast of image) into the analysis process for more in-depth and ecologically valid findings. Among the interactive visualization components, a context-embedded interactive scatterplot (CIS) is especially designed to help users to examine abstract gaze data in diverse contexts by embedding medical imaging representations well-known to radiologists in it. We also present the results from case studies with chest and abdominal radiologists
- Language
- English
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.