Open access
Oct 2026
HierTAC-Net: a robust and explainable hierarchical attention network for multi-configuration gastrointestinal endoscopic image classification
HierTAC-Net is proposed, a six-stage convolutional classifier that combines inverted residual blocks, triple attention at three resolution levels, and ConvNeXt-style refinement that is suitable for real-time deployment and remains stable under common artifacts and across two clinical collections.
M. Mia, A. Khandakar, Sawal Hamid Bin Mohd Ali et al.
· Scientific Reports · 0 citations