Unknown

Dataset Information

0

Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program.


ABSTRACT: Deep learning algorithms have been used to detect diabetic retinopathy (DR) with specialist-level accuracy. This study aims to validate one such algorithm on a large-scale clinical population, and compare the algorithm performance with that of human graders. A total of 25,326 gradable retinal images of patients with diabetes from the community-based, nationwide screening program of DR in Thailand were analyzed for DR severity and referable diabetic macular edema (DME). Grades adjudicated by a panel of international retinal specialists served as the reference standard. Relative to human graders, for detecting referable DR (moderate NPDR or worse), the deep learning algorithm had significantly higher sensitivity (0.97 vs. 0.74, p?p?p?p?

SUBMITTER: Raumviboonsuk P 

PROVIDER: S-EPMC6550283 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

altmetric image

Publications

Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program.

Raumviboonsuk Paisan P   Krause Jonathan J   Chotcomwongse Peranut P   Sayres Rory R   Raman Rajiv R   Widner Kasumi K   Campana Bilson J L BJL   Phene Sonia S   Hemarat Kornwipa K   Tadarati Mongkol M   Silpa-Archa Sukhum S   Limwattanayingyong Jirawut J   Rao Chetan C   Kuruvilla Oscar O   Jung Jesse J   Tan Jeffrey J   Orprayoon Surapong S   Kangwanwongpaisan Chawawat C   Sukumalpaiboon Ramase R   Luengchaichawang Chainarong C   Fuangkaew Jitumporn J   Kongsap Pipat P   Chualinpha Lamyong L   Saree Sarawuth S   Kawinpanitan Srirut S   Mitvongsa Korntip K   Lawanasakol Siriporn S   Thepchatri Chaiyasit C   Wongpichedchai Lalita L   Corrado Greg S GS   Peng Lily L   Webster Dale R DR  

NPJ digital medicine 20190410


Deep learning algorithms have been used to detect diabetic retinopathy (DR) with specialist-level accuracy. This study aims to validate one such algorithm on a large-scale clinical population, and compare the algorithm performance with that of human graders. A total of 25,326 gradable retinal images of patients with diabetes from the community-based, nationwide screening program of DR in Thailand were analyzed for DR severity and referable diabetic macular edema (DME). Grades adjudicated by a pa  ...[more]

Similar Datasets

| S-EPMC7758133 | biostudies-literature
| S-EPMC10011252 | biostudies-literature
| S-EPMC8899862 | biostudies-literature
| S-EPMC7311555 | biostudies-literature
| S-EPMC9472142 | biostudies-literature
| S-EPMC8184323 | biostudies-literature
| S-EPMC9249740 | biostudies-literature
| S-EPMC11358665 | biostudies-literature
| S-EPMC8656222 | biostudies-literature
| S-EPMC6094472 | biostudies-literature