Deep learning algorithms for automatic detection of pterygium using anterior segment photographs from slit-lamp and hand-held cameras

被引:25
作者
Fang, Xiaoling [1 ,2 ]
Deshmukh, Mihir [1 ]
Chee, Miao Li [1 ]
Soh, Zhi-Da [1 ]
Teo, Zhen Ling [1 ]
Thakur, Sahil [1 ]
Goh, Jocelyn Hui Lin [1 ]
Liu, Yu-Chi [1 ,3 ]
Husain, Rahat [1 ,3 ]
Mehta, Jodhbir [1 ,3 ]
Wong, Tien Yin [1 ,3 ,4 ,5 ]
Cheng, Ching-Yu [1 ,3 ,4 ,5 ]
Rim, Tyler Hyungtaek [1 ,3 ]
Tham, Yih-Chung [1 ,3 ]
机构
[1] Singapore Natl Eye Ctr, Singapore Eye Res Inst, Singapore, Singapore
[2] Shanghai Eye Hosp, Shanghai Eye Dis Prevent & Treatment Ctr, Dept Ophthalmol, Shanghai, Peoples R China
[3] Duke NUS Med Sch, Ophthalmol & Visual Sci Acad Clin Program, Singapore, Singapore
[4] Natl Univ Singapore, Yong Loo Lin Sch Med, Dept Ophthalmol, Singapore, Singapore
[5] Natl Univ Hlth Syst, Singapore, Singapore
基金
英国医学研究理事会; 上海市自然科学基金;
关键词
imaging; ocular surface; RISK-FACTORS; POPULATION; EPIDEMIOLOGY; PREVALENCE; RECURRENCE; SIZE; EYE;
D O I
10.1136/bjophthalmol-2021-318866
中图分类号
R77 [眼科学];
学科分类号
100212 ;
摘要
Background/aims To evaluate the performances of deep learning (DL) algorithms for detection of presence and extent pterygium, based on colour anterior segment photographs (ASPs) taken from slit-lamp and hand-held cameras. Methods Referable pterygium was defined as having extension towards the cornea from the limbus of >2.50 mm or base width at the limbus of >5.00 mm. 2503 images from the Singapore Epidemiology of Eye Diseases (SEED) study were used as the development set. Algorithms were validated on an internal set from the SEED cohort (629 images (55.3% pterygium, 8.4% referable pterygium)), and tested on two external clinic-based sets (set 1 with 2610 images (2.8% pterygium, 0.7% referable pterygium, from slit-lamp ASP); and set 2 with 3701 images, 2.5% pterygium, 0.9% referable pterygium, from hand-held ASP). Results The algorithm's area under the receiver operating characteristic curve (AUROC) for detection of any pterygium was 99.5%(sensitivity=98.6%; specificity=99.0%) in internal test set, 99.1% (sensitivity=95.9%, specificity=98.5%) in external test set 1 and 99.7% (sensitivity=100.0%; specificity=88.3%) in external test set 2. For referable pterygium, the algorithm's AUROC was 98.5% (sensitivity=94.0%; specificity=95.3%) in internal test set, 99.7% (sensitivity=87.2%; specificity=99.4%) in external set 1 and 99.0% (sensitivity=94.3%; specificity=98.0%) in external set 2. Conclusion DL algorithms based on ASPs can detect presence of and referable-level pterygium with optimal sensitivity and specificity. These algorithms, particularly if used with a handheld camera, may potentially be used as a simple screening tool for detection of referable pterygium. Further validation in community setting is warranted. Synopsis/precis DL algorithms based on ASPs can detect presence of and referable-level pterygium optimally, and may be used as a simple screening tool for the detection of referable pterygium in community screenings.
引用
收藏
页码:1642 / 1647
页数:6
相关论文
共 34 条
[1]  
[Anonymous], 2001, ADV NEURAL INFORM PR
[2]  
Avisar R, 2000, ISRAEL MED ASSOC J, V2, P14
[3]   Optimal classifier for imbalanced data using Matthews Correlation Coefficient metric [J].
Boughorbel, Sabri ;
Jarray, Fethi ;
El-Anbari, Mohammed .
PLOS ONE, 2017, 12 (06)
[4]   The prevalence of and risk factors for pterygium in an urban Malay population: The Singapore Malay Eye Study (SiMES) [J].
Cajucom-Uy, H. ;
Tong, L. ;
Wong, T. Y. ;
Tay, W. T. ;
Saw, S. M. .
BRITISH JOURNAL OF OPHTHALMOLOGY, 2010, 94 (08) :977-981
[5]   Pterygium: new insights [J].
Chu, Wai Kit ;
Choi, Hiu Lam ;
Bhat, Amar K. ;
Jhanji, Vishal .
EYE, 2020, 34 (06) :1047-1050
[6]   The prevalence, severity and risk factors for pterygium in central Myanmar: the Meiktila Eye Study [J].
Durkin, S. R. ;
Abhary, S. ;
Newland, H. S. ;
Selva, D. ;
Aung, T. ;
Casson, R. J. .
BRITISH JOURNAL OF OPHTHALMOLOGY, 2008, 92 (01) :25-29
[7]   Ethnic differences in the incidence of pterygium in a multi-ethnic Asian population: the Singapore Epidemiology of Eye Diseases Study [J].
Fang, Xiao Ling ;
Chong, Crystal Chun Yuen ;
Thakur, Sahil ;
Soh, Zhi Da ;
Teo, Zhen Ling ;
Majithia, Shivani ;
Lim, Zhi Wei ;
Rim, Tyler Hyungtaek ;
Sabanayagam, Charumathi ;
Wong, Tien Yin ;
Cheng, Ching-Yu ;
Tham, Yih-Chung .
SCIENTIFIC REPORTS, 2021, 11 (01)
[8]   Comparison of Different Measurement Tools and Dimensional Parameters of Pterygium to Investigate its Impact on Refractive Indices and Ocular Aberrations [J].
Gumus, Koray ;
Guven, Aysegul ;
Altinkaynak, Miray ;
Keskin, Merve .
EYE & CONTACT LENS-SCIENCE AND CLINICAL PRACTICE, 2018, 44 (02) :118-124
[9]   Effect of Pterygia on Refractive Indices, Corneal Topography, and Ocular Aberrations [J].
Gumus, Koray ;
Erkilic, Kuddusi ;
Topaktas, Duygu ;
Colin, Joseph .
CORNEA, 2011, 30 (01) :24-29
[10]   The treatment of pterygium [J].
Hirst, LW .
SURVEY OF OPHTHALMOLOGY, 2003, 48 (02) :145-180