The incidence of diabetes in Mauritius is amongst the highest in the world. Diabetic retinopathy (DR), a complication resulting from the disease, can lead to blindness if not detected early. The aim of this work was to investigate the use of transfer learning and data augmentation for the classification of fundus images into five different stages of diabetic retinopathy. The five stages are No DR, Mild nonproliferative DR, Moderate nonproliferative DR, Severe nonproliferative DR and Proliferative. To this end, deep transfer learning and three pre-trained models, VGG16, ResNet50 and DenseNet169, were used to classify the APTOS dataset. The preliminary experiments resulted in low training and validation accuracies, and hence, the APTOS dataset was augmented while ensuring a balance between the five classes. This dataset was then used to train the three models, and the best three models were used to classify a blind Mauritian test datum. We found that the ResNet50 model produced the best results out of the three models and also achieved very good accuracies for the five classes. The classification of class-4 Mauritian fundus images, severe cases, produced some unexpected results, with some images being classified as mild, and therefore needs to be further investigated.

Over the years, machine learning and deep learning have shown tremendous potential and good accuracy in the detection of Chronic Kidney Disease (CKD). Yet, these models are being criticised for their opaque decision-making process. To date, few studies have been conducted to look into which variables are more relevant in determining CKD. In this study, we first used two (2) ensemble approaches: XGBoost, AdaBoost. for detecting CKD. Next, we propose a model that uses deep neural networks (DNN) to extract and represent the features for the prognosis of CKD. The DNN was enhanced by fine-tuning the hyperparameters and it outperformed the ensemble methods with an accuracy of 98.8%. Finally, SHapley Additive exPlanations (SHAP), which is a technique that has attracted attention in the field of explainable artificial intelligence (XAI), was used to investigate which features contribute to predicting CKD and the results confirmed the medical explanations.