HydraMix-Net: A Deep Multi-task Semi-supervised Learning Approach for Cell Detection and Classification
This addresses the challenge of costly labeling in medical imaging, offering an incremental improvement for researchers and practitioners in that domain.
The paper tackles the problem of cell detection and classification in medical imaging with limited labeled data by proposing HydraMix-Net, a semi-supervised multi-task learning approach, achieving 80% accuracy compared to 70% for a simple CNN with only 100 labeled examples.
Semi-supervised techniques have removed the barriers of large scale labelled set by exploiting unlabelled data to improve the performance of a model. In this paper, we propose a semi-supervised deep multi-task classification and localization approach HydraMix-Net in the field of medical imagining where labelling is time consuming and costly. Firstly, the pseudo labels are generated using the model's prediction on the augmented set of unlabelled image with averaging. The high entropy predictions are further sharpened to reduced the entropy and are then mixed with the labelled set for training. The model is trained in multi-task learning manner with noise tolerant joint loss for classification localization and achieves better performance when given limited data in contrast to a simple deep model. On DLBCL data it achieves 80\% accuracy in contrast to simple CNN achieving 70\% accuracy when given only 100 labelled examples.