ScholarDeck

Cohort studyPhysics in medicine and biology · 2026

Modeling variable interactions using Bayesian networks to identify direct predictors of radiation-induced optic neuropathy after proton therapy: implications for personalized toxicity risk stratification.

Thao-Nguyen Pham, Ibrahim Chamseddine, Dorothee Lebhertz … Juliette Thariat

OphthalmologyCohort study

Vascular risk and baseline vision predicted optic nerve damage after proton therapy

Modeling variable interactions using Bayesian networks to identify direct predictors of radiation-induced optic neuropathy after proton therapy: implications for personalized toxicity risk stratification.

Thao-Nguyen Pham … Juliette Thariat

Physics in medicine and biology · 2026

Background

Introduction.

Methods

Demographic, ophthalmological and treatment data were prospectively collected from patients undergoing scanned proton beams between 2018 and 2024.

n = 179 eyes

0.75

Results

0.75

higher score means better prediction of who develops eye nerve damage

n = 179 eyes

More results

Even with modern proton therapy and adherence to dose constraints, about 6% of patients develop near-blindness from radiation-induced optic neuropathy.

Visual field deficit served as surrogate for grade ⩾2 clinically-relevant radiation-induced optic neuropathy.

Of 179 eyes (105 patients), 24% developed radiation-induced optic neuropathy.

More results

The bayesian network identified 15 directional associations, 12 were clinically valid, 2 conceivable and 1 weak.

“

Conclusion

This contrast with photon models is hypothesis-generating and favors proton-specific models for radiation-induced optic neuropathy prediction.

Post

Modeling variable interactions using Bayesian networks to identify direct predictors of radiation-induced optic neuropathy after proton therapy: implications for personalized toxicity risk stratification.

Vascular risk and baseline vision predicted optic nerve damage after proton therapy

Related papers