Glioma subtype classification from histopathological images using in-domain and out-of-domain transfer learning: An experimental study
This work addresses the problem of accurate and efficient glioma classification for pathologists, though it is incremental as it builds on existing transfer learning methods.
The study tackled glioma subtype classification from histopathological images by comparing transfer learning strategies and proposing a semi-supervised approach, achieving a balanced accuracy of 96.91% and F1-score of 97.07% while reducing annotation effort.
We provide in this paper a comprehensive comparison of various transfer learning strategies and deep learning architectures for computer-aided classification of adult-type diffuse gliomas. We evaluate the generalizability of out-of-domain ImageNet representations for a target domain of histopathological images, and study the impact of in-domain adaptation using self-supervised and multi-task learning approaches for pretraining the models using the medium-to-large scale datasets of histopathological images. A semi-supervised learning approach is furthermore proposed, where the fine-tuned models are utilized to predict the labels of unannotated regions of the whole slide images (WSI). The models are subsequently retrained using the ground-truth labels and weak labels determined in the previous step, providing superior performance in comparison to standard in-domain transfer learning with balanced accuracy of 96.91% and F1-score 97.07%, and minimizing the pathologist's efforts for annotation. Finally, we provide a visualization tool working at WSI level which generates heatmaps that highlight tumor areas; thus, providing insights to pathologists concerning the most informative parts of the WSI.