SparseConvMIL: Sparse Convolutional Context-Aware Multiple Instance Learning for Whole Slide Image Classification
This work addresses the classification of cancer subtypes in medical imaging, representing an incremental improvement by incorporating spatial context into existing multiple instance learning methods.
The paper tackled the problem of whole slide image classification by exploiting spatial relationships between tiles, achieving state-of-the-art performance on a dataset of 10,000 images for cancer subtype classification.
Multiple instance learning (MIL) is the preferred approach for whole slide image classification. However, most MIL approaches do not exploit the interdependencies of tiles extracted from a whole slide image, which could provide valuable cues for classification. This paper presents a novel MIL approach that exploits the spatial relationship of tiles for classifying whole slide images. To do so, a sparse map is built from tiles embeddings, and is then classified by a sparse-input CNN. It obtained state-of-the-art performance over popular MIL approaches on the classification of cancer subtype involving 10000 whole slide images. Our results suggest that the proposed approach might (i) improve the representation learning of instances and (ii) exploit the context of instance embeddings to enhance the classification performance. The code of this work is open-source at {github censored for review}.