Automatic Diagnosis of Pathological Myopia from Heterogeneous Biomedical Data

Dana C Crawford,Zhuo Zhang,Damon Wing Kee Wong,Yanwu Xu,Seang-Mei Saw,Jiang Liu,Tien Yin Wong,Chee Keong Kwoh

doi:10.1371/journal.pone.0065736

Abstract

Pathological myopia is one of the leading causes of blindness worldwide. The condition is particularly prevalent in Asia. Unlike myopia, pathological myopia is accompanied by degenerative changes in the retina, which if left untreated can lead to irrecoverable vision loss. The accurate diagnosis of pathological myopia will enable timely intervention and facilitate better disease management to slow down the progression of the disease. Current methods of assessment typically consider only one type of data, such as that from retinal imaging. However, different kinds of data, including that of genetic, demographic and clinical information, may contain different and independent information, which can provide different perspectives on the visually observable, genetic or environmental mechanisms for the disease. The combination of these potentially complementary pieces of information can enhance the understanding of the disease, providing a holistic appreciation of the multiple risks factors as well as improving the detection outcomes. In this study, we propose a computer-aided diagnosis framework for Pathological Myopia diagnosis through Biomedical and Image Informatics(PM-BMII). Through the use of multiple kernel learning (MKL) methods, PM-BMII intelligently fuses heterogeneous biomedical information to improve the accuracy of disease diagnosis. Data from 2,258 subjects of a population-based study, in which demographic and clinical information, retinal fundus imaging data and genotyping data were collected, are used to evaluate the proposed framework. The experimental results show that PM-BMII achieves an AUC of 0.888, outperforming the detection results from the use of demographic and clinical information 0.607 (increase , ), genotyping data 0.774 (increase , ) or imaging data 0.852 (increase , ) alone. The accuracy of the results obtained demonstrates the feasibility of using heterogeneous data for improved disease diagnosis through our proposed PM-BMII framework.

Highlights

Pathological Myopia Pathological myopia (PM) is one of the leading causes of visual impairment worldwide[1,2,3] and is the most frequent cause of visual impairment in Asian countries [4]
Experimental Methods for PM-BMII To demonstrate that the combination of multiple data sources can enhance detection accuracy in our PM-BMII framework, we report and compare the diagnosis performance of 7 methods using the following different features and their combinations: 1. Demographic/clinical data only
The mean and standard variation (SD) values of area under the ROC curve (AUC) of each method were calculated based on the results obtained from the 20 sets of cross validation testing as described in the above Methods section

Summary

Introduction

Pathological Myopia Pathological myopia (PM) is one of the leading causes of visual impairment worldwide[1,2,3] and is the most frequent cause of visual impairment in Asian countries [4]. Known as high myopia or degenerative myopia, pathological myopia is a type of severe and progressive nearsightedness characterized by changes in the fundus of the eye, due to posterior staphyloma and deficient corrected acuity. Current clinical practice in detecting pathological myopia relies heavily on the manual screening and efforts of the clinicians, where a complete eye exam usually takes up to 60 minutes. Such eye exams include questions on the subject’s medical history and a physical eye examination which includes tests for visual acuity, visual field and refraction. A slit lamp exam evaluates the anterior sections and lens of the eye using microscope optics; tonometry measures the pressure inside the eye; and ophthalmoscopy allows observation of the back of the eye

Objectives

Methods

Results

Conclusion