Hybrid Deep Learning Gaussian Process for Diabetic Retinopathy Diagnosis and Uncertainty Quantification
This work addresses diabetic retinopathy diagnosis for medical applications, but it is incremental as it builds on existing deep learning methods by adding uncertainty quantification.
The paper tackles diabetic retinopathy diagnosis by proposing a hybrid deep learning-Gaussian process method for grade estimation and uncertainty quantification, showing that uncertainty quantification improves interpretability as a diagnostic support tool.
Diabetic Retinopathy (DR) is one of the microvascular complications of Diabetes Mellitus, which remains as one of the leading causes of blindness worldwide. Computational models based on Convolutional Neural Networks represent the state of the art for the automatic detection of DR using eye fundus images. Most of the current work address this problem as a binary classification task. However, including the grade estimation and quantification of predictions uncertainty can potentially increase the robustness of the model. In this paper, a hybrid Deep Learning-Gaussian process method for DR diagnosis and uncertainty quantification is presented. This method combines the representational power of deep learning, with the ability to generalize from small datasets of Gaussian process models. The results show that uncertainty quantification in the predictions improves the interpretability of the method as a diagnostic support tool. The source code to replicate the experiments is publicly available at https://github.com/stoledoc/DLGP-DR-Diagnosis.