BACKGROUND: Nonophthalmologist physicians do not confidently perform direct ophthalmoscopy. The use of artificial intelligence to detect papilledema and other optic-disk abnormalities from fundus photographs has not been well studied. METHODS: We trained, validated, and externally tested a deep-learning system to classify optic disks as being normal or having papilledema or other abnormalities from 15,846 retrospectively collected ocular fundus photographs that had been obtained with pharmacologic pupillary dilation and various digital cameras in persons from multiple ethnic populations. Of these photographs, 14,341 from 19 sites in 11 countries were used for training and validation, and 1505 photographs from 5 other sites were used for external testing. Performance at classifying the optic-disk appearance was evaluated by calculating the area under the receiver-operating-characteristic curve (AUC), sensitivity, and specificity, as compared with a reference standard of clinical diagnoses by neuro-ophthalmologists. RESULTS: The training and validation data sets from 6779 patients included 14,341 photographs: 9156 of normal disks, 2148 of disks with papilledema, and 3037 of disks with other abnormalities. The percentage classified as being normal ranged across sites from 9.8 to 100%; the percentage classified as having papilledema ranged across sites from zero to 59.5%. In the validation set, the system discriminated disks with papilledema from normal disks and disks with nonpapilledema abnormalities with an AUC of 0.99 (95% confidence interval [CI], 0.98 to 0.99) and normal from abnormal disks with an AUC of 0.99 (95% CI, 0.99 to 0.99). In the external-testing data set of 1505 photographs, the system had an AUC for the detection of papilledema of 0.96 (95% CI, 0.95 to 0.97), a sensitivity of 96.4% (95% CI, 93.9 to 98.3), and a specificity of 84.7% (95% CI, 82.3 to 87.1). CONCLUSIONS: A deep-learning system using fundus photographs with pharmacologically dilated pupils differentiated among optic disks with papilledema, normal disks, and disks with nonpapilledema abnormalities.
Artificial Intelligence to Detect Papilledema from Ocular Fundus Photographs / Milea, Dan; Najjar, Raymond P; Zhubo, Jiang; Ting, Daniel; Vasseneix, Caroline; Xu, Xinxing; Aghsaei Fard, Masoud; Fonseca, Pedro; Vanikieti, Kavin; Lagrèze, Wolf A; La Morgia, Chiara; Cheung, Carol Y; Hamann, Steffen; Chiquet, Christophe; Sanda, Nicolae; Yang, Hui; Mejico, Luis J; Rougier, Marie-Bénédicte; Kho, Richard; Thi Ha Chau, Tran; Singhal, Shweta; Gohier, Philippe; Clermont-Vignal, Catherine; Cheng, Ching-Yu; Jonas, Jost B; Yu-Wai-Man, Patrick; Fraser, Clare L; Chen, John J; Ambika, Selvakumar; Miller, Neil R; Liu, Yong; Newman, Nancy J; Wong, Tien Y; Biousse, Valérie; BONSAI Group; Amore, Giulia; Carelli, Valerio. - In: THE NEW ENGLAND JOURNAL OF MEDICINE. - ISSN 0028-4793. - ELETTRONICO. - 382:18(2020), pp. 1687-1695. [10.1056/NEJMoa1917130]
Artificial Intelligence to Detect Papilledema from Ocular Fundus Photographs
La Morgia, Chiara;Amore, GiuliaMembro del Collaboration Group
;Carelli, ValerioMembro del Collaboration Group
2020
Abstract
BACKGROUND: Nonophthalmologist physicians do not confidently perform direct ophthalmoscopy. The use of artificial intelligence to detect papilledema and other optic-disk abnormalities from fundus photographs has not been well studied. METHODS: We trained, validated, and externally tested a deep-learning system to classify optic disks as being normal or having papilledema or other abnormalities from 15,846 retrospectively collected ocular fundus photographs that had been obtained with pharmacologic pupillary dilation and various digital cameras in persons from multiple ethnic populations. Of these photographs, 14,341 from 19 sites in 11 countries were used for training and validation, and 1505 photographs from 5 other sites were used for external testing. Performance at classifying the optic-disk appearance was evaluated by calculating the area under the receiver-operating-characteristic curve (AUC), sensitivity, and specificity, as compared with a reference standard of clinical diagnoses by neuro-ophthalmologists. RESULTS: The training and validation data sets from 6779 patients included 14,341 photographs: 9156 of normal disks, 2148 of disks with papilledema, and 3037 of disks with other abnormalities. The percentage classified as being normal ranged across sites from 9.8 to 100%; the percentage classified as having papilledema ranged across sites from zero to 59.5%. In the validation set, the system discriminated disks with papilledema from normal disks and disks with nonpapilledema abnormalities with an AUC of 0.99 (95% confidence interval [CI], 0.98 to 0.99) and normal from abnormal disks with an AUC of 0.99 (95% CI, 0.99 to 0.99). In the external-testing data set of 1505 photographs, the system had an AUC for the detection of papilledema of 0.96 (95% CI, 0.95 to 0.97), a sensitivity of 96.4% (95% CI, 93.9 to 98.3), and a specificity of 84.7% (95% CI, 82.3 to 87.1). CONCLUSIONS: A deep-learning system using fundus photographs with pharmacologically dilated pupils differentiated among optic disks with papilledema, normal disks, and disks with nonpapilledema abnormalities.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
