Browsing by Author "Liao, Hung-Chang"

Now showing 1 - 2 of 2

Predicting breast tumor via mining DNA viruses with decision tree
(2006)
Chen, Mu-Chen
;
Liao, Hung-Chang
;
Huang, Cheng-Lung
;
運輸與物流管理系註：原交通所+運管所
;
Department of Transportation and Logistics Management
Breast cancer is a serious problem, especially the young women in Taiwan. Until now, in the most medical researches, the reasons for suffering from breast tumor are unclear. However, most medical researches proved that DNA viruses are the high-risk factors closely related to human cancers. In recent years, hospitals and health organizations have been furnished with modern computerized medical equipment for data collection, monitoring and diagnosis. Additionally, these data are stored in large medical information systems for analysis purpose. Developing truthful and reliable classifiers for diagnosis and prognosis has become an essential task in medical and healthcare. It was reported with increasing confirmation that the machine learning algorithms can generate more accurate and transparent classifiers and decision rules for physicians than traditional methodologies. In the machine learning algorithms, decision trees have been already successfully used in the areas of medicine and healthcare. In this paper, an algorithm of decision trees, Chi-squared Automatic Interaction Detection (CHAID), is applied to build a classifier for predicting breast cancer and fibroadenoma. The results demonstrate that the decision tree technique is more favorably than logistic regression in terms of rule accuracy and knowledge transparency to physicians. Furthermore, the medical classifier can assist inexperienced physicians to prevent from misdiagnosis.
Prediction model building and feature selection with support vector machines in breast cancer diagnosis
(2008-01-01)
Huang, Cheng-Lung
;
Liao, Hung-Chang
;
Chen, Mu-Chen
;
運輸與物流管理系註：原交通所+運管所
;
Department of Transportation and Logistics Management
Breast cancer is a serious problem for the young women of Taiwan. Some medical researches have proved that DNA viruses are one of the high-risk factors closely related to human cancers. Five DNA viruses are studied in this research: specific types of HSV-1 (herpes simplex virus type 1), EBV (Epstein-Barr virus), CMV (cytomegalovirus), HPV (human. papillomavirus), and HHV-8 (human herpesvirus-8). The purposes of this study are to obtain the bioinformatics about breast tumor and DNA viruses, and to build an accurate diagnosis model about breast cancer and fibroadenoma. Research efforts have reported with increasing confirmation that the support vector machine (SVM) has a greater accurate diagnosis ability. Therefore, this study constructs a hybrid SVM-based strategy with feature selection to render a diagnosis between the breast cancer and fibroadenoma and to find the important risk factor for breast cancer. The results show that JHSV-l, HHV-8} or {HSV-1, HHV-8, CMV) are the most important features and that the diagnosis model achieved high classification accuracy, at 86% of average overall hit rate. A Linear discriminate analysis (LDA) diagnosis model is also constructed in this study. The LDA model shows that {HSV-1, HHV-8, EBV} or {HSV-1, HHV-8} are significant factors which are similar to that of the SVM-based classifier. However, the classificatory accuracy of the SVM-based classifier is slightly better than that of LDA in the negative hit ratio, positive hit ratio, and overall hit ratio. (c) 2006 Elsevier Ltd. All rights reserved.