WISE: Efficient WSI selection for active learning in histopathology

• Published in: Computerized medical imaging and graphics 2024-12, Vol.118, p.102455, Article 102455
• Main Authors: Kang, Hyeongu, Kim, Mujin, Ko, Young Sin, Cho, Yesung, Yi, Mun Yong
• Format: Article
• Language: English
• Subjects: Active learning; Cancer diagnosis; Deep Learning; Histopathology; Humans; Hybrid; Image Interpretation, Computer-Assisted - methods; Neural Networks, Computer; Problem-Based Learning; Whole-slide images; WSI-informative selection
• ISSN: 0895-6111; 1879-0771; 1879-0771
• DOI: 10.1016/j.compmedimag.2024.102455

Deep neural network (DNN) models have been applied to a wide variety of medical image analysis tasks, often with the successful performance outcomes that match those of medical doctors. However, given that even minor errors in a model can impact patients’ life, it is critical that these models are continuously improved. Hence, active learning (AL) has garnered attention as an effective and sustainable strategy for enhancing DNN models for the medical domain. Extant AL research in histopathology has primarily focused on patch datasets derived from whole-slide images (WSIs), a standard form of cancer diagnostic images obtained from a high-resolution scanner. However, this approach has failed to address the selection of WSIs, which can impede the performance improvement of deep learning models and increase the number of WSIs needed to achieve the target performance. This study introduces a WSI-level AL method, termed WSI-informative selection (WISE). WISE is designed to select informative WSIs using a newly formulated WSI-level class distance metric. This method aims to identify diverse and uncertain cases of WSIs, thereby contributing to model performance enhancement. WISE demonstrates state-of-the-art performance across the Colon and Stomach datasets, collected in the real world, as well as the public DigestPath dataset, significantly reducing the required number of WSIs by more than threefold compared to the one-pool dataset setting, which has been dominantly used in the field. •Active learning enhances deep learning in medicine.•Neglecting WSIs in active learning causes inefficiencies in train dataset construction.•WISE embeds WSIs in class distances.•K-means++ selects diverse cases in WISE.•WISE reduces the number of WSIs required threefold while improving the model.