Back to Search
Start Over
HSG-MGAF Net: Heterogeneous subgraph-guided multiscale graph attention fusion network for interpretable prediction of whole-slide image.
- Source :
-
Computer Methods & Programs in Biomedicine . Apr2024, Vol. 247, pN.PAG-N.PAG. 1p. - Publication Year :
- 2024
-
Abstract
- • Developed a HSG-MGAF net to adaptively establish low-order and potential high-order spatial relationships of patches for interpretable prediction of giga-pixel WSI. • Embeds a Top-2k heterogeneous subgraph to guide hierarchical multi-scale graph framework for adaptive slide prediction and ROI localization. • Self-supervised contrastive learning constraint is introduced to maximize the consistency of predictions at heterogeneous scales in the optimization. • Interpretable heatmaps with complementary features can be obtained by using heterogeneous relationships of the top ranked patches at two scales. Pathological whole slide image (WSI) prediction and region of interest (ROI) localization are important issues in computer-aided diagnosis and postoperative analysis in clinical applications. Existing computer-aided methods for predicting WSI are mainly based on multiple instance learning (MIL) and its variants. However, most of the methods are based on instance independence and identical distribution assumption and performed at a single scale, which not fully exploit the hierarchical multiscale heterogeneous information contained in WSI. Heterogeneous Subgraph-Guided Multiscale Graph Attention Fusion Network (HSG-MGAF Net) is proposed to build the topology of critical image patches at two scales for adaptive WSI prediction and lesion localization. The HSG-MGAF Net simulates the hierarchical heterogeneous information of WSI through graph and hypergraph at two scales, respectively. This framework not only fully exploits the low-order and potential high-order correlations of image patches at each scale, but also leverages the heterogeneous information of the two scales for adaptive WSI prediction. We validate the superiority of the proposed method on the CAMELYON16 and the TCGA- NSCLC, and the results show that HSG-MGAF Net outperforms the state-of-the-art method on both datasets. The average ACC, AUC and F 1 score of HSG-MGAF Net can reach 92.7 %/0.951/0.892 and 92.2 %/0.957/0.919, respectively. The obtained heatmaps can also localize the positive regions more accurately, which have great consistency with the pixel-level labels. The results demonstrate that HSG-MGAF Net outperforms existing weakly supervised learning methods by introducing critical heterogeneous information between the two scales. This approach paves the way for further research on light weighted heterogeneous graph-based WSI prediction and ROI localization. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 01692607
- Volume :
- 247
- Database :
- Academic Search Index
- Journal :
- Computer Methods & Programs in Biomedicine
- Publication Type :
- Academic Journal
- Accession number :
- 176008083
- Full Text :
- https://doi.org/10.1016/j.cmpb.2024.108099