Facial image medical analysis system using quantitative chromatic feature

Authors:
Xingzheng Wang;Bob Zhang;Zhenhua Guo;David Zhang
Affiliations:
Biometric Research Center, Dept. of Computing, The Hong Kong Polytechnic University, Hong Kong, China;Biometric Research Center, Dept. of Computing, The Hong Kong Polytechnic University, Hong Kong, China and Dept. of Electrical and Computer Engineering, University of Waterloo, Waterloo, ON, Canada;Shenzhen Key Laboratory of Broadband Network & Multimedia, Graduate School at Shenzhen, Tsinghua University, Shenzhen, China;Biometric Research Center, Dept. of Computing, The Hong Kong Polytechnic University, Hong Kong, China
Venue:
Expert Systems with Applications: An International Journal
Year:
2013

Citing 8
Cited 0

Face Detection in Color Images

IEEE Transactions on Pattern Analysis and Machine Intelligence
Gene Selection for Cancer Classification using Support Vector Machines

Machine Learning
An introduction to variable and feature selection

The Journal of Machine Learning Research
Variable selection using svm based criteria

The Journal of Machine Learning Research
Selection and Fusion of Color Models for Image Feature Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Pulse images recognition using fuzzy neural network

Expert Systems with Applications: An International Journal
An optimized tongue image color correction scheme

IEEE Transactions on Information Technology in Biomedicine
Color clustering and learning for image segmentation based on neural networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	12.05

Visualization

Abstract

In order to investigate whether the appearance of a human face can be utilized for diagnostic purposes, which have been practiced for thousands of years in Traditional Chinese Medicine (TCM), this paper aims to present a computerized facial image analysis system by using quantitative chromatic features for disease diagnosis applications. A face image acquisition device is dedicatedly designed to acquire image samples from volunteers who have three types of health conditions: normal health, icterohepatitis, and severe hepatitis. Then, after color calibration on the acquired images to remove noises caused by lighting fluctuations, quantitative dominant color features are extracted by fuzzy clustering method. In order to further improve the diagnosis accuracy, a feature selection procedure is involved to identify the most discriminative feature subset for the diagnostic classification. Lastly, based on these selected quantitative feature, each face image could be diagnosed into different health groups. Experiments are conducted based on a database which includes over 300 sample images, and the result shows that the overall diagnosis accuracy between healthy samples and other two diseases is higher than 88%. Hence the feasibility of disease diagnosis by inspecting the chromatic feature of human face could be verified.