ExplainabilityTransparencySafetycs.LG
AI to Identify Strain-sensitive Regions of the Optic Nerve Head Linked to Functional Loss in Glaucoma
Thanadet Chuangsuwanich, Monisha E. Nongpiur, Fabian A. Braeu, Tin A. Tun, Alexandre Thiery, Shamira Perera, Ching Lin Ho, Martin Buist, George Barbastathis, Tin Aung, Michaël J. A. Girard
June 9, 2025
arXiv.org
2 citations
Key Takeaways
-
AI can predict specific patterns of vision loss in glaucoma by analyzing how the optic nerve head deforms under pressure, with models achieving 77-88% accuracy when incorporating biomechanical strain data alongside structural imaging—suggesting that how tissues physically respond to eye pressure matters more than structure alone for understanding individual disease progression.
-
Explainable AI revealed that the inferior and inferotemporal regions of the optic nerve rim are particularly strain-sensitive and critical to predicting vision loss, providing clinicians with specific anatomical targets that could guide more personalized monitoring and treatment decisions rather than one-size-fits-all approaches.
-
This research demonstrates the value of combining biomechanical modeling with machine learning for precision medicine in ophthalmology, showing that understanding the mechanical stress on eye tissues—not just their appearance—can improve predictions of which patients will experience specific types of functional decline.
-
The study involved 237 glaucoma patients with expert classification of their vision loss patterns, providing real-world validation that AI predictions correspond to clinically meaningful outcomes that patients actually experience, rather than abstract statistical measures.
Abstract
Objective: (1) To assess whether ONH biomechanics improves prediction of three progressive visual field loss patterns in glaucoma; (2) to use explainable AI to identify strain-sensitive ONH regions contributing to these predictions. Methods: We recruited 237 glaucoma subjects. The ONH of one eye was imaged under two conditions: (1) primary gaze and (2) primary gaze with IOP elevated to ~35 mmHg via ophthalmo-dynamometry. Glaucoma experts classified the subjects into four categories based on the presence of specific visual field defects: (1) superior nasal step (N=26), (2) superior partial arcuate (N=62), (3) full superior hemifield defect (N=25), and (4) other/non-specific defects (N=124). Automatic ONH tissue segmentation and digital volume correlation were used to compute IOP-induced neural tissue and lamina cribrosa (LC) strains. Biomechanical and structural features were input to a Geometric Deep Learning model. Three classification tasks were performed to detect: (1) superior nasal step, (2) superior partial arcuate, (3) full superior hemifield defect. For each task, the data were split into 80% training and 20% testing sets. Area under the curve (AUC) was used to assess performance. Explainable AI techniques were employed to highlight the ONH regions most critical to each classification. Results: Models achieved high AUCs of 0.77-0.88, showing that ONH strain improved VF loss prediction beyond morphology alone. The inferior and inferotemporal rim were identified as key strain-sensitive regions, contributing most to visual field loss prediction and showing progressive expansion with increasing disease severity. Conclusion and Relevance: ONH strain enhances prediction of glaucomatous VF loss patterns. Neuroretinal rim, rather than the LC, was the most critical region contributing to model predictions.