Predicting 24-hour intraocular pressure peaks and averages with machine learning

Aug 4, 2026

Upload by Liam

Predicting 24-hour peak and average intraocular pressure (IOP) is crucial for diagnosing and managing glaucoma, where IOP varies substantially across the day and many peaks occur outside clinic hours. This retrospective study developed and evaluated machine learning models using electronic medical records from January 2014 to May 2024, incorporating 24-hour IOP monitoring data and patient characteristics. Five algorithms and multiple time-point combinations were tested, with feature importance interpreted using Shapley Additive Explanations.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Methods

Results

Conclusion

Keywords

OPEN ACCESS EDITED BY Flora Hui, Centre for Eye Research Australia, Australia REVIEWED BY Xianlong Zeng, Ohio University, United States Pan Sai, Chinese PLA General Hospital, China *CORRESPONDENCE Yanlong Bi [biyanlong@tongji.edu.cn](biyanlong@tongji.edu.cn)[ ](biyanlong@tongji.edu.cn)Haohao Zhu [zhuhaohao@fudan.edu.cn](zhuhaohao@fudan.edu.cn)[ ](zhuhaohao@fudan.edu.cn)RECEIVED 04 July 2024 ACCEPTED 19 September 2024 PUBLISHED 07 October 2024 CITATION Chen R, Lei J, Liao Y, Jin Y, Wang X, Li X, Wu D, Li H, Bi Y and Zhu H (2024) Predicting 24-hour intraocular pressure peaks and averages with machine learning. Front. Med. 11:1459629. doi: 10.3389/fmed.2024.1459629 COPYRIGHT © 2024 Chen, Lei, Liao, Jin, Wang, Li, Wu, Li, Bi and Zhu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. TYPE Original Research PUBLISHED 07 October 2024 DOI 10.3389/fmed.2024.1459629 Predicting 24-hour intraocular pressure peaks and averages with machine learning Ranran Chen 1, Jinming Lei 2, Yujie Liao 1, Yiping Jin 1, Xue Wang3, Xiaomei Li 1, Danping Wu 1, Hong Li 1, Yanlong Bi4* and Haohao Zhu 1* 1 Department of Ophthalmology, Shanghai Fifth People's Hospital, Fudan University, Shanghai, China, 2Software Engineering, Shenzhen Yishi Huolala Technology Company Limited, Shenzhen, China, 3 Department of Ophthalmology, Shanghai East Hospital, School of Medicine, Tongji University, Shanghai, China, 4 Department of Ophthalmology, Tongji Eye Institute, Tongji Hospital, School of Medicine, Tongji University, Shanghai, China Purpose: Predicting 24-hour peak and average intraocular pressure (IOP) is essential for the diagnosis and management of glaucoma. This study aimed to develop and assess a machine learning model for predicting 24-hour peak and average IOP, leveraging advanced techniques to enhance prediction accuracy. We also aimed to identify relevant features and provide insights into the prediction results to better inform clinical practice. Methods: In this retrospective study, electronic medical records from January 2014 to May 2024 were analyzed, incorporating 24-hour IOP monitoring data and patient characteristics. Predictive models based on five machine learning algorithms were trained and evaluated. Five time points (10:00 AM, 12:00 PM, 2:00 PM, 4:00 PM, and 6:00 PM) were tested to optimize prediction accuracy using their combinations. The model with the highest performance was selected, and feature importance was assessed using Shapley Additive Explanations. Results: This study included data from 517 patients (1,034 eyes) . For predicting 24-hour peak IOP, the Random Forest Regression (RFR) model utilizing IOP values at 10:00 AM, 12:00 PM, 2:00 PM, and 4:00 PM achieved optimal performance: MSE 5. 248, RMSE 2. 291, MAE 1.694, and R2 0.823. For predicting 24-hour average IOP, the RFR model using IOP values at 10:00 AM, 12:00 PM, 4:00 PM, and 6:00 PM performed best: MSE 1.374, RMSE 1. 172, MAE 0. 869, and R2 0.918. Conclusion: The study developed machine learning models that predict 24-hour peak and average IOP. Specific time point combinations and the RFR algorithm were identified, which improved the accuracy of predicting 24-hour peak and average intraocular pressure. These findings provide the potential for more effective management and treatment strategies for glaucoma patients. KEYWORDS intraocular pressure, 24-hour, measurement, nocturnal, machine learning, glaucoma Frontiers in Medicine 01 [frontiersin.org](frontiersin.org) 1 Introduction Glaucoma, a leading cause of global blindness (

FAQs

What problem does the study address in glaucoma care?

The study targets prediction of 24-hour IOP peak and average values, which are important for diagnosis and management of glaucoma where IOP fluctuates over the day.

How were the machine learning models built and evaluated?

Models were trained and tested in a retrospective dataset using 24-hour IOP monitoring data and patient characteristics, comparing five machine learning algorithms and multiple combinations of five time points.

Which model performed best for predicting peak and average IOP?

For 24-hour peak IOP, the Random Forest Regression model using IOP at 10:00 AM, 12:00 PM, 2:00 PM, and 4:00 PM performed best. For 24-hour average IOP, the best result came from the Random Forest Regression model using IOP at 10:00 AM, 12:00 PM, 4:00 PM, and 6:00 PM.

Aug 4, 2026

Upload by Liam