A transductive few-shot learning approach for classification of digital histopathological slides from liver cancer (2311.17740v2)
Abstract: This paper presents a new approach for classifying 2D histopathology patches using few-shot learning. The method is designed to tackle a significant challenge in histopathology, which is the limited availability of labeled data. By applying a sliding window technique to histopathology slides, we illustrate the practical benefits of transductive learning (i.e., making joint predictions on patches) to achieve consistent and accurate classification. Our approach involves an optimization-based strategy that actively penalizes the prediction of a large number of distinct classes within each window. We conducted experiments on histopathological data to classify tissue classes in digital slides of liver cancer, specifically hepatocellular carcinoma. The initial results show the effectiveness of our method and its potential to enhance the process of automated cancer diagnosis and treatment, all while reducing the time and effort required for expert annotation.
- “Deep learning for Whole Slide Image analysis: An overview,” Clinical Orthopaedics and Related Research, vol. abs/1910.11097, 2019.
- “The overview of the deep learning integrated into the medical imaging of liver: A review,” Hepatology International, vol. 15, pp. 868–880, 2021.
- “Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning,” IEEE Transactions on Medical Imaging, vol. 35, no. 5, pp. 1285–1298, 2016.
- “Predicting survival after hepatocellular carcinoma resection using deep learning on histological slides,” Hepatology, vol. 72, no. 6, pp. 2000–2013, 2020.
- “Machine learning in computational histopathology: Challenges and opportunities,” Genes, Chromosomes and Cancer, 2023.
- Martin J Van Den Bent, “Interobserver variation of the histopathological diagnosis in clinical trials on glioma: a clinician’s perspective,” Acta Neuropathologica, vol. 120, no. 3, pp. 297–304, 2010.
- “Survey on deep learning with class imbalance,” Journal of Big Data, vol. 6, no. 1, pp. 1–54, 2019.
- “Towards practical few-shot query sets: Transductive minimum description length inference,” Advances in Neural Information Processing Systems, vol. 35, pp. 34677–34688, 2022.
- “Realistic evaluation of transductive few-shot learning,” Advances in Neural Information Processing Systems, vol. 34, 2021.
- “An overview of few-shot learning methods in analysis of histopathological images,” Advances in Smart Healthcare Paradigms and Applications, pp. 87–113, 2023.
- “A closer look at few-shot classification,” in International Conference on Learning Representations, 2019.
- “Tasknorm: Rethinking batch normalization for meta-learning,” in International Conference on Machine Learning. PMLR, 2020, pp. 1153–1164.
- “Leveraging the feature distribution in transfer-based few-shot learning,” in International Conference on Artificial Neural Networks. Springer, 2021, pp. 487–499.
- Thorsten Joachims, “Transductive inference for text classification using support vector machines,” in International Conference on Machine Learning, 1999, vol. 99, pp. 200–209.
- “Sparse inverse covariance estimation with the graphical lasso,” Biostatistics, vol. 9, no. 3, pp. 432–441, 2008.
- “Self supervised contrastive learning for digital histopathology,” Machine Learning with Applications, vol. 7, pp. 100198, 2022.
- “Color transfer between images,” IEEE Computer Graphics and Applications, vol. 21, no. 5, pp. 34–41, 2001.
- “Simpleshot: Revisiting nearest-neighbor classification for few-shot learning,” Computer Vision and Pattern Recognition, 2019.
