A semi-supervised learning approach with consistency regularization for tumor histopathological images analysis

Abstract

IntroductionManual inspection of histopathological images is important in clinical cancer diagnosis. Pathologists implement pathological diagnosis and prognostic evaluation through the microscopic examination of histopathological slices. This entire process is time-consuming, laborious, and challenging for pathologists. The modern use of whole-slide imaging, which scans histopathology slides to digital slices, and analysis using computer-aided diagnosis is an essential problem.MethodsTo solve the problem of difficult labeling of histopathological data, and improve the flexibility of histopathological analysis in clinical applications, we herein propose a semi-supervised learning algorithm coupled with consistency regularization strategy, called“Semi- supervised Histopathology Analysis Network”(Semi-His-Net), for automated normal-versus-tumor and subtype classifications. Specifically, when inputted disturbing versions of the same image, the model should predict similar outputs. Based on this, the model itself can assign artificial labels to unlabeled data for subsequent model training, thereby effectively reducing the labeled data required for training.ResultsOur Semi-His-Net is able to classify patches from breast cancer histopathological images into normal tissue and three other different tumor subtypes, achieving an accuracy was 90%. The average AUC of cross-classification between tumors reached 0.893.DiscussionTo overcome the limitations of visual inspection by pathologists for histopathology images, such as long time and low repeatability, we have developed a deep learning-based framework (Semi-His-Net) for automatic classification subdivision of the subtypes contained in the whole pathological images. This learning-based framework has great potential to improve the efficiency and repeatability of histopathological image diagnosis.

Deep Transfer Learning Optimization Techniques for Medical Image Classifcation - A Survey

“…They are particularly useful for solving complex optimization problems, such as combinatorial or multiobjective optimization tasks, where traditional search methods might struggle to find high-quality solutions efficiently [56]. The suitability of each technique can depend on the architecture of the network and the characteristics of the data, and experimentation is often needed to determine the best choice for a particular task [49].…”

Deep Transfer Learning Optimization Techniques for Medical Image Classifcation - A Survey

“…The experiments conducted on two challenging medical imaging classification tasks demonstrate that the proposed approach outperforms state-ofthe-art baselines in terms of cross-domain generalization capability. This suggests that the learned feature space is more effective in capturing the underlying structure and variability of medical images [56], allowing the model to generalize better to new and unseen data.…”