Frontiers in physiology · 2026
Artificial intelligence for pathologic myopia classification based on the META-PM system: a systematic review and meta-analysis.
СтатьяСистематический обзорНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Hu Yuting, Wang Xiaoyan, Ji Yuke, Chen Yuming, Liang Ziqi, Wu Sheng, Lai Mingying, Yang Weihua
Аннотация
BACKGROUND: Artificial intelligence (AI) has shown considerable potential for pathologic myopia (PM) detection, yet its overall diagnostic performance under the META-PM classification framework remains uncertain.
METHODS: A comprehensive literature search was conducted across PubMed, Web of Science, IEEE Xplore, and Embase up to February 10, 2026. Studies applying deep learning models for PM classification using the META-PM system were included. Pooled sensitivity, specificity, and hierarchical summary receiver operating characteristic (HSROC) analyses were calculated. Twelve studies were eligible for the systematic review, of which ten were included in the meta-analysis.
RESULTS: For referable PM detection, the pooled sensitivity and specificity were 0.96 (95% CI: 0.93-0.97) and 0.98 (95% CI: 0.96-0.99), respectively. For PM classification, the pooled macro-area under the receiver operating characteristic curve (AUC) reached 0.99 (95% CI: 0.98-1.00), indicating excellent overall diagnostic performance. Fagan nomogram analysis demonstrated favorable post-test probabilities across different clinical scenarios. The meta-regression identified external validation and image input resolution as significant sources of heterogeneity across studies.
CONCLUSION: Overall, AI demonstrates outstanding performance for automated PM detection and grading under the META-PM framework. Future studies should focus on multicenter external validation, prospective clinical evaluation, and integration with emerging ultra-widefield imaging technologies to facilitate real- world implementation in primary eye-care screening and risk stratification systems.
SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD420261351134.
METHODS: A comprehensive literature search was conducted across PubMed, Web of Science, IEEE Xplore, and Embase up to February 10, 2026. Studies applying deep learning models for PM classification using the META-PM system were included. Pooled sensitivity, specificity, and hierarchical summary receiver operating characteristic (HSROC) analyses were calculated. Twelve studies were eligible for the systematic review, of which ten were included in the meta-analysis.
RESULTS: For referable PM detection, the pooled sensitivity and specificity were 0.96 (95% CI: 0.93-0.97) and 0.98 (95% CI: 0.96-0.99), respectively. For PM classification, the pooled macro-area under the receiver operating characteristic curve (AUC) reached 0.99 (95% CI: 0.98-1.00), indicating excellent overall diagnostic performance. Fagan nomogram analysis demonstrated favorable post-test probabilities across different clinical scenarios. The meta-regression identified external validation and image input resolution as significant sources of heterogeneity across studies.
CONCLUSION: Overall, AI demonstrates outstanding performance for automated PM detection and grading under the META-PM framework. Future studies should focus on multicenter external validation, prospective clinical evaluation, and integration with emerging ultra-widefield imaging technologies to facilitate real- world implementation in primary eye-care screening and risk stratification systems.
SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD420261351134.