Generative-adversarial-learning-based biomarker activation map for improving the interpretability of deep-learning-aided diabetic retinopathy screening

Abstract

Purpose : Deep learning (DL) can assist the diagnosis of diabetic retinopathy (DR) based on optical coherence tomography angiography. However, it is unclear how deep learning classification models arrive at their results, limiting discovery of DR biomarkers. To improve interpretability, a novel biomarker activation map (BAM) generation framework based on generative adversarial learning (GAL) is proposed.

Methods : 50 healthy participants and 305 patients with diabetes were recruited in this study. Masked trained retina specialists graded each eye based on 7-field fundus photography using early treatment of diabetic retinopathy study (ETDRS) criteria as either non-referable (nrDR; ETDRS score < 35) or referable (rDR; ETDRS score ≥ 35 or macular edema). Macular 3×3-mm scans for one or both eyes of each participant were acquired using a commercial 70-kHz spectral-domain OCT system. 456 superficial vascular complex (SVC) en face OCTA images were collected. The data set was divided into training (60%), validation (20%), and testing (20%) sets and used to train a referable/non-referable DR classifier that used the en face SVC angiograms as input. Two U-shaped networks were also trained as generators. The main generator was trained to produce an angiogram that the classifier would classify as nrDR from an input angiogram, while the assistant generator produces angiograms that would be classified as rDR (Fig. 1). The BAM was finally constructed as the absolute difference image between the input angiogram and output of the main generator.

Results : The diagnosis of rDR achieved an overall accuracy of 92%. Compared to the traditional class activation maps (CAMs) (Fig. 2B), the generated BAMs explicitly highlighted most of the pathological nonperfusion area, which represents the most important DR biomarker visible in the input angiogram (Fig. 2E). Unlike traditional CAMs, the BAMs ignored non-pathological features like the foveal avascular zone while emphasizing real pathological features that contribute the diagnostic decision making.

Conclusions : The generated BAMs using GAL method could provide sufficient interpretability to help clinicians utilize DL-aided referable DR screening, and help to quickly discern the DR-related pathologies. This innovation may also facilitate the new biomarker discovery on OCTA used for diagnosis of retinal vascular diseases.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.