AUTOMATED DIAGNOSIS OF RETINOPATHY BY CONTENT-BASED IMAGE RETRIEVAL

被引:53
作者
Chaum, Edward [1 ,2 ,3 ,4 ]
Karnowski, Thomas P. [5 ]
Govindasamy, V. Priya [5 ]
Abdelrahman, Mohamed [6 ]
Tobin, Kenneth W. [5 ]
机构
[1] Univ Tennessee, Hlth Sci Ctr, Dept Ophthalmol, Memphis, TN USA
[2] Univ Tennessee, Hlth Sci Ctr, Dept Anat & Neurobiol, Memphis, TN USA
[3] Univ Tennessee, Hlth Sci Ctr, Dept Biomed Engn, Memphis, TN USA
[4] Univ Tennessee, Hlth Sci Ctr, Dept Pediat, Memphis, TN USA
[5] Oak Ridge Natl Lab, Image Sci & Machine Vis Grp, Oak Ridge, TN USA
[6] Tennessee Technol Univ, Cookeville, TN 38505 USA
来源
RETINA-THE JOURNAL OF RETINAL AND VITREOUS DISEASES | 2008年 / 28卷 / 10期
关键词
content-based image retrieval; computer-aided diagnosis; retina; retinopathy; diabetic retinopathy; age-related macular degeneration; image analysis;
D O I
10.1097/IAE.0b013e31818356dd
中图分类号
R77 [眼科学];
学科分类号
100212 ;
摘要
Purpose: To describe a novel computer-based image analysis method that is being developed to assist and automate the diagnosis of retinal disease. Methods: Content-based image retrieval is the process of retrieving related images from large database collections using their pictorial content. The content feature list becomes the index for storage, search, and retrieval of related images from a library based upon specific visual characteristics. Low-level analyses use feature description models and higher-level analyses use perceptual organization and spatial relationships, including clinical metadata, to extract semantic information. Results: We defined, extracted, and tested a large number of region- and lesion-based features from a dataset of 395 retinal images. Using a statistical hold-one-out method, independent queries for each image were submitted to the system and a diagnostic prediction was formulated. The diagnostic sensitivity for all stratified levels of age-related macular degeneration ranged from 75% to 100%. Similarly, the sensitivity of detection and accuracy for proliferative diabetic retinopathy ranged from 75% to 91.7% and for nonproliferative diabetic retinopathy, ranged from 75% to 94.7%. The overall purity of the diagnosis (specificity) for all disease states in the dataset was 91.3%. Conclusions: The probabilistic nature of content-based image retrieval permits us to make statistically relevant predictions regarding the presence, severity, and manifestations of common retinal diseases from digital images in an automated and deterministic manner.
引用
收藏
页码:1463 / 1477
页数:15
相关论文
共 59 条
[1]  
ABRAMOFF MD, 2006, 28 IEEE EMBS ANN C N
[2]   Automated segmentation of the optic disc from stereo color photographs using physiologically plausible features [J].
Abramoff, Michael D. ;
Alward, Wallace L. M. ;
Greenlee, Emily C. ;
Shuba, Lesya ;
Kim, Chan Y. ;
Fingert, John H. ;
Kwon, Young H. .
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE, 2007, 48 (04) :1665-1673
[3]  
Anand R, 2000, OPHTHALMOLOGY, V107, P2224
[4]  
[Anonymous], 1976, Am J Ophthalmol, V81, P383
[5]  
[Anonymous], 1985, ARCH OPHTHALMOL-CHIC, V103, P1796, DOI [DOI 10.1001/ARCHOPHT.1985.01050120030015, 10.1001/archopht.1985.01050120030015]
[6]  
[Anonymous], 2001, Exploratory Image Databases
[7]  
ANTANI S, 2004, MEDINFO, V11, P829
[8]   An optimal algorithm for approximate nearest neighbor searching in fixed dimensions [J].
Arya, S ;
Mount, DM ;
Netanyahu, NS ;
Silverman, R ;
Wu, AY .
JOURNAL OF THE ACM, 1998, 45 (06) :891-923
[9]   A new approach of geodesic reconstruction for drusen segmentation in eye fundus images [J].
Ben Sbeh, Z ;
Cohen, LD ;
Mimoun, G ;
Coscas, G .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2001, 20 (12) :1321-1333
[10]   Semiconductor sidewall shape estimation [J].
Bingham, PR ;
Price, JR ;
Tobin, KW ;
Karnowski, TP .
JOURNAL OF ELECTRONIC IMAGING, 2004, 13 (03) :474-485