{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deepseenet-a-deep-learning-model-for","title":"DeepSeeNet: A deep learning model for automated classification of patient-based age-related macular degeneration severity from color fundus photographs","arxiv_id":"1811.07492","date":"2018-11-19","proceeding":null,"authors":["Yifan Peng","Shazia Dharssi","Qingyu Chen","Tiarnan D. Keenan","Elvira Agrón","Wai T. Wong","Emily Y. Chew","Zhiyong Lu"],"abstract":"In assessing the severity of age-related macular degeneration (AMD), the\nAge-Related Eye Disease Study (AREDS) Simplified Severity Scale predicts the\nrisk of progression to late AMD. However, its manual use requires the\ntime-consuming participation of expert practitioners. Although several\nautomated deep learning systems have been developed for classifying color\nfundus photographs (CFP) of individual eyes by AREDS severity score, none to\ndate has used a patient-based scoring system that uses images from both eyes to\nassign a severity score. DeepSeeNet, a deep learning model, was developed to\nclassify patients automatically by the AREDS Simplified Severity Scale (score\n0-5) using bilateral CFP. DeepSeeNet was trained on 58,402 and tested on 900\nimages from the longitudinal follow-up of 4549 participants from AREDS. Gold\nstandard labels were obtained using reading center grades. DeepSeeNet simulates\nthe human grading process by first detecting individual AMD risk factors\n(drusen size, pigmentary abnormalities) for each eye and then calculating a\npatient-based AMD severity score using the AREDS Simplified Severity Scale.\nDeepSeeNet performed better on patient-based classification (accuracy = 0.671;\nkappa = 0.558) than retinal specialists (accuracy = 0.599; kappa = 0.467) with\nhigh AUC in the detection of large drusen (0.94), pigmentary abnormalities\n(0.93), and late AMD (0.97). DeepSeeNet demonstrated high accuracy with\nincreased transparency in the automated assignment of individual patients to\nAMD risk categories based on the AREDS Simplified Severity Scale. These results\nhighlight the potential of deep learning to assist and enhance clinical\ndecision-making in patients with AMD, such as early AMD detection and risk\nprediction for developing late AMD. DeepSeeNet is publicly available on\nhttps://github.com/ncbi-nlp/DeepSeeNet.","url_abs":"http://arxiv.org/abs/1811.07492v2","url_pdf":"http://arxiv.org/pdf/1811.07492v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deepseenet-a-deep-learning-model-for","repo_url":"https://github.com/ncbi-nlp/DeepSeeNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}