Related Experiment Video
Updated: Mar 27, 2026

Full-Circle Cauterization of Limbal Vascular Plexus for Surgically Induced Glaucoma in Rodents
Published on: February 15, 2022
Predicting Intraocular Pressure From Glaucoma Patients Receiving Medication Treatment Using Explainable Machine
Robert T James1, Wenke Liu2, Gadi Wollstein1,3,4
1Departments of Ophthalmology and Radiology, Tech4Health Institute and Neuroscience Institute, New York University Grossman School of Medicine, NYU Langone Health, New York University, New York, New York, USA, nyu.edu.
Explainable AI predicts glaucoma treatment success by analyzing patient data. Key factors like Insulin-like Growth Factor 1 and LDL cholesterol influence intraocular pressure outcomes.
Area of Science:
- Ophthalmology
- Medical Informatics
- Computational Biology
Background:
- Glaucoma is a neurodegenerative disease impacting vision, with treatment aiming to reduce intraocular pressure (IOP).
- Some patients do not respond effectively to IOP-lowering treatments, risking progressive vision loss.
- Explainable machine learning (EML) offers tools for predicting treatment outcomes and identifying influential factors.
Purpose of the Study:
- To utilize EML to predict IOP in glaucoma patients undergoing medication treatment.
- To identify key features impacting treatment success using EML.
Main Methods:
- The study analyzed data from 161 glaucoma patients in the UK Biobank.
- eXtreme Gradient Boosting (XGBoost) was employed for predicting IOP, with feature sets including demographics, physiometabolic data, and medications.
- SHapley Additive exPlanation (SHAP) values were calculated to determine feature importance and interactions.
Main Results:
- XGBoost achieved an AUC of 0.708 when using a combined feature set.
- Insulin-like Growth Factor 1 (IGF-1), low-density lipoprotein (LDL), and lymphocyte count were the most significant predictors of IOP.
- Strong interactions were observed between LDL and IGF-1 in influencing treatment outcomes.
Conclusions:
- EML, specifically XGBoost, can effectively predict IOP outcomes in glaucoma patients.
- Blood levels of LDL and IGF-1 are crucial factors influencing the effectiveness of IOP-lowering treatments.
- This approach aids in understanding individual prognoses and identifying potential therapeutic targets.
Related Concept Videos
Open Angle Glaucoma: Treatment
Drugs such as carbonic anhydrase inhibitors, α2- and...
Glaucoma: Overview
Angle Closure Glaucoma: Treatment
Ophthalmic Drug Delivery Systems

