Equity-enhanced glaucoma progression prediction from OCT with knowledge distillation

Sulaiman O. Afolabi, Leila Gheisi, Jing Shan, Lucy Q. Shen, Mengyu Wang, Min Shi

Introduction

Glaucoma is a leading cause of irreversible blindness worldwide. Predicting the rate of disease progression is vital for clinical intervention, yet current AI models face significant challenges. First, longitudinal data—required to track changes over years—is scarce compared to diagnostic "snapshots." Second, existing models often exhibit algorithmic bias, performing less accurately for minority groups due to anatomical variations and skewed training data.

Visual progression of glaucoma from normal vision to extreme tunnel vision.

To address these issues, researchers developed FairDist, a deep learning framework designed to predict glaucoma progression from a single baseline Optical Coherence Tomography (OCT) scan while ensuring equitable performance across diverse demographic groups.

The FairDist Framework

The FairDist architecture employs a two-stage approach to overcome data scarcity and bias:

  1. Stage 1: Equity-aware Teacher (EqEffNet): A large-scale model is first trained on a detection task (glaucoma vs. non-glaucoma) using 10,000 samples. It incorporates a Fairness-Aware Attention Layer that uses demographic data (race and gender) to weight anatomical features. This allows the model to learn group-specific markers, ensuring that "normal" variations in retinal thickness across different races do not lead to misdiagnosis.
  2. Stage 2: Knowledge Distillation (KD): The "wisdom" of the fair teacher model is transferred to a "student" model specialized in progression prediction. Because the progression dataset is small (500 samples), the student model is trained to mimic the internal feature representations of the teacher. This distillation process effectively bridges the data gap, allowing the progression model to inherit the fairness and robust feature extraction of the larger detection model.
Overview of longitudinal visual field tests and the two-stage FairDist framework involving EqEffNet and Knowledge Distillation.

Results and Equity Gains

The FairDist model was evaluated using the Equity-Scaled AUC (ES-AUC), a metric that penalizes performance gaps between subgroups. The results demonstrate that FairDist significantly outperforms standard architectures:

  • Predictive Accuracy: For Mean Deviation (MD) fast progression, FairDist achieved an overall AUC of 0.78, surpassing baseline models like EfficientNet.
  • Fairness Improvements: The model showed substantial gains in equity. For gender, the ES-AUC improved by 0.11 over traditional adversarial training.
  • Subgroup Performance: Notably, the Black subgroup saw a 0.14 AUC improvement compared to using the teacher model alone, suggesting that the distillation process actively enhances the student's ability to generalize across underrepresented groups.

Conclusion

FairDist represents a major step toward responsible AI in ophthalmology. By combining fairness-aware attention with knowledge distillation, the framework provides a viable path for training accurate, equitable models even when specialized longitudinal data is limited. Future work will focus on expanding the dataset size and integrating multimodal data, such as fundus photography and clinical notes, to further refine prediction accuracy.

Summary updated

Abstract

Abstract Purpose To develop an equitable deep learning model with knowledge distillation to enhance the demographic equity in glaucoma progression prediction. Methods We developed a novel deep learning model called FairDist which used baseline optical coherency tomography (OCT) scans to predict glaucoma progression. First, an equity-aware EfficientNet termed EqEffNet was trained for glaucoma detection. Next, the pretrained detection model was adapted for progression prediction using knowledge distillation which minimizes image and identity feature differences between the detection and progression models. Progression was defined based on longitudinal visual field maps from at least five visits up to six years. Model performance was measured by the area under the receiver operating characteristic curve (AUC), Sensitivity, Specificity, and equity was assessed using equity-scaled AUC (ES-AUC), which adjusts AUC by accounting for subgroup disparities, focusing on gender and racial groups. Results Two types of glaucoma progression including mean deviation (MD) Fast progression and Total Deviation (TD) Pointwise progression were explored. For MD Fast Progression, FairDist achieved the highest AUC and ES-AUC for gender (AUC: 0.738, ES-AUC: 0.693) and race (AUC: 0.778, ES-AUC: 0.677) compared with methods with and without integrating inequity mitigation strategies. For TD Pointwise progression, FairDist achieved the best AUC and ES-AUC for gender (AUC: 0.743, ES-AUC: 0.719) and race (AUC: 0.746, ES-AUC: 0.645) among all methods. Conclusions FairDist enhances both model performance and equity in glaucoma progression prediction after integrating the equity-aware learning and knowledge distillation components. The proposed deep learning model shows promise in improving glaucoma diagnosis while reducing disparities across demographic groups.
BibTeX
@article{afolabi2025equity,
  author = {Sulaiman O. Afolabi and Leila Gheisi and Jing Shan and Lucy Q. Shen and Mengyu Wang and Min Shi},
  title = {{Equity-enhanced glaucoma progression prediction from OCT with knowledge distillation}},
  journal = {Npj Digit. Medicine},
  year = {2025},
  doi = {10.1038/s41746-025-01884-9},
}
Figure