Inherently interpretable deep learning framework highlights clinically relevant image regions.

Similar Items

• Inherently interpretable deep learning framework highlights lesions more precisely than post-hoc techniques applied to a standard DNN.
  by: Kerol Djoumessi (21349992)
  Published: (2025)
• Overview of the development data and proposed inherently interpretable deep learning framework evaluated in this study.
  by: Kerol Djoumessi (21349992)
  Published: (2025)
• Providing AI-based clinical decision support based on the inherently interpretable deep learning model improves DR screening.
  by: Kerol Djoumessi (21349992)
  Published: (2025)
• Summary of the classification performance with confidence intervals (CIs) computed at 95% using bootstrapping (n=1000). “AUC” refer to the receiver-operating curve. “Loc Bag” and “Loc GBP” respectively refer to the localization precision of the sparse BagNet and Guided Backpropagation on ResNet-50 at localizing lesions from annotated images. For each dataset, the first row shows the performance of the interpretable sparse BagNet model, while the second row shows the performance of the baseline black-box ResNet-50 model. The Kaggle dataset (first row) is the internal dataset used to train and evaluate the model, while the other datasets were used for external validation to assess the generalization properties of the trained model. The low classification performance on the FCM-UNA and FGA-DR datasets can be explained by the relatively low quality of most images in the FCM-UNA dataset and the large intensity variation of the FGA-DR dataset (S
  by: Kerol Djoumessi (21349992)
  Published: (2025)
• Summary of the internal and external validation datasets used to evaluate the models. “Origin” refers to the country where the data was collected. “Lesion” refers to the number of images in the dataset with lesion annotations. The Kaggle dataset (first row, shaded in gray) is the internal dataset used to evaluate the model, while the other datasets were used for external validation to assess the generalization properties of the trained model.
  by: Kerol Djoumessi (21349992)
  Published: (2025)