Presentation 2003/3/8
High-order Visualization of Medical Test Data with PrototypeLines
Takeshi WATANABE, Einoshin SUZUKI, Hideto YOKOI, Katsuhiko TAKABAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a method which visualizes irregular multi-dimensional time-series data as a sequence of probabilistic prototypes, and apply it to medical data. Chronic hepatitis data, which represent typical medical test data, pose various challenges to conventional machine learning methods. We have proposed a method which obtains prototypes from the data, and the prototypes were admitted valid and comprehensive by physicians. This paper proposes PrototypeLines which represent a visualization method based on the prototyping method. PrototypeLines consider information criterion and hue recognition, and display medical data of each patient in a comprehensive format. Promising results were obtained by experiments in which we visualized patients based on their degrees of fibrosis.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) PrototypeLines / Visualization / Estimation of Mixture Probabilistic Models / Time-Series Data / Medical Test Data
Paper # AI2002-84
Date of Issue

Conference Information
Committee AI
Conference Date 2003/3/8(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) High-order Visualization of Medical Test Data with PrototypeLines
Sub Title (in English)
Keyword(1) PrototypeLines
Keyword(2) Visualization
Keyword(3) Estimation of Mixture Probabilistic Models
Keyword(4) Time-Series Data
Keyword(5) Medical Test Data
1st Author's Name Takeshi WATANABE
1st Author's Affiliation Faculty of Engineering, Yokohama National University()
2nd Author's Name Einoshin SUZUKI
2nd Author's Affiliation Faculty of Engineering, Yokohama National University
3rd Author's Name Hideto YOKOI
3rd Author's Affiliation Division of Medical Informatics, Chiba University Hospital
4th Author's Name Katsuhiko TAKABAYASHI
4th Author's Affiliation Division of Medical Informatics, Chiba University Hospital
Date 2003/3/8
Paper # AI2002-84
Volume (vol) vol.102
Number (no) 711
Page pp.pp.-
#Pages 6
Date of Issue