Leveraging Artificial Intelligence and Clinical Laboratory Evidence to Advance Mobile Health Applications in Ophthalmology: Taking the Ocular Surface Disease as a Case Study DOI Creative Commons
Han Wang, Yi Pan, Xudong Jiang

et al.

iLABMED, Journal Year: 2025, Volume and Issue: unknown

Published: March 12, 2025

ABSTRACT Background The advent of mobile health (mHealth) applications has fundamentally transformed the healthcare landscape, particularly within field ophthalmology, by providing unprecedented opportunities for remote diagnosis, monitoring, and treatment. Ocular surface diseases, including dry eye disease (DED), are most common diseases that can be detected mHealth applications. However, artificial intelligence (AI) systems ocular detection predominantly based on self‐reported data collected through interviews, which lack rigor clinical evidence. These constraints underscore need to develop robust, evidence‐based AI frameworks incorporate objective indicators improve reliability utility Methods Two novel deep learning (DL) models, YoloTR YoloMBTR, were developed detect key (OSIs), tear meniscus height (TMH), non‐invasive Keratograph break‐up time (NIKBUT), redness, lipid layer, trichiasis. Additionally, back propagation neural networks (BPNN) universal network image segmentation (U‐Net) employed classification meibomian gland images predict Demodex mite infections. models trained a large dataset from high‐resolution devices, 5M various platforms (Huawei, Apple, Xiaomi). Results proposed DL YoloMBTR outperformed baseline you only look once (YOLO) (Yolov5n, Yolov6n, Yolov8n) across multiple performance metrics, test average precision (AP), validation AP, overall accuracy. two also exhibit superior compared machine plug‐in in KG5M when benchmarked against gold standard. Using Python's Matplotlib visualization SPSS statistical analysis, this study introduces an innovative proof‐of‐concept framework leveraging quantitative analysis address critical challenges ophthalmology. By integrating advanced offers robust approach detecting quantifying OSIs with high degree precision. This methodological advancement bridges gap between AI‐driven diagnostics ophthalmology translating complex into actionable insights. Conclusions Integrating laboratory holds significant potential advancing (MeHealth), OSIs. aims explore integration, focusing improving diagnostic accuracy accessibility. demonstrates tools ophthalmic diagnostics, paving way reliable, solutions patient monitoring continuous care. results contribute foundation AI‐powered extend beyond accessibility outcomes domains.

Language: Английский

Leveraging Artificial Intelligence and Clinical Laboratory Evidence to Advance Mobile Health Applications in Ophthalmology: Taking the Ocular Surface Disease as a Case Study DOI Creative Commons
Han Wang, Yi Pan, Xudong Jiang

et al.

iLABMED, Journal Year: 2025, Volume and Issue: unknown

Published: March 12, 2025

ABSTRACT Background The advent of mobile health (mHealth) applications has fundamentally transformed the healthcare landscape, particularly within field ophthalmology, by providing unprecedented opportunities for remote diagnosis, monitoring, and treatment. Ocular surface diseases, including dry eye disease (DED), are most common diseases that can be detected mHealth applications. However, artificial intelligence (AI) systems ocular detection predominantly based on self‐reported data collected through interviews, which lack rigor clinical evidence. These constraints underscore need to develop robust, evidence‐based AI frameworks incorporate objective indicators improve reliability utility Methods Two novel deep learning (DL) models, YoloTR YoloMBTR, were developed detect key (OSIs), tear meniscus height (TMH), non‐invasive Keratograph break‐up time (NIKBUT), redness, lipid layer, trichiasis. Additionally, back propagation neural networks (BPNN) universal network image segmentation (U‐Net) employed classification meibomian gland images predict Demodex mite infections. models trained a large dataset from high‐resolution devices, 5M various platforms (Huawei, Apple, Xiaomi). Results proposed DL YoloMBTR outperformed baseline you only look once (YOLO) (Yolov5n, Yolov6n, Yolov8n) across multiple performance metrics, test average precision (AP), validation AP, overall accuracy. two also exhibit superior compared machine plug‐in in KG5M when benchmarked against gold standard. Using Python's Matplotlib visualization SPSS statistical analysis, this study introduces an innovative proof‐of‐concept framework leveraging quantitative analysis address critical challenges ophthalmology. By integrating advanced offers robust approach detecting quantifying OSIs with high degree precision. This methodological advancement bridges gap between AI‐driven diagnostics ophthalmology translating complex into actionable insights. Conclusions Integrating laboratory holds significant potential advancing (MeHealth), OSIs. aims explore integration, focusing improving diagnostic accuracy accessibility. demonstrates tools ophthalmic diagnostics, paving way reliable, solutions patient monitoring continuous care. results contribute foundation AI‐powered extend beyond accessibility outcomes domains.

Language: Английский

Citations

0