HMARNET-A Hierarchical Multi-Attention Residual Network for Gleason scoring of prostate cancer

Manual delineation of prostate cancer (PCa) from whole slide images (WSIs) demands requires pathologists with adequate domain knowledge. This process is generally strenuous and may be subjected to poor inter-pathologist reproducibility. Accurate Gleason scoring is an important step in the computer-aided diagnosis of PCa. This work proposes a novel lightweight convolutional neural networks (CNN) to extract significant hierarchical features from the histopathology images. It learns meticulous attention-guided feature representations through the convolutional layers for precise scoring of Gleason grades. The Hierarchical Multi-Attention Residual (HMAR) block extracts attention-guided features and fuses the resulting feature maps from multiple levels with a reduced number of trainable parameters. We have also proposed a new lightweight channel-attention module, enhanced channel attention (ECA) to extract inter-channel features from different receptive fields. With the presence of different attention mechanisms, the network learns to focus on more significant features rather than unnecessary ones. Additionally, mixed FocalOHEM loss is proposed to optimize the CNN and efficiently minimize the error. While the focal loss helps to address the class imbalance present in the TCGA-PRAD dataset, the online hard example mining (OHEM) loss focuses on optimizing hard negative samples. It achieved an accuracy of 89.19% with a Kappa score of 86% on the TCGA-PRAD dataset.

Automated summary

This study proposes a novel lightweight convolutional neural network for accurate scoring of Gleason grades in prostate cancer. With multiple levels of attention mechanisms and a channel attention module, the network is able to focus on more significant features, thus improving the accuracy of scoring.