Multimodal Distillation-Driven Ensemble Learning for Long-Tailed Histopathology Whole Slide Images Analysis
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929738699767808 |
|---|---|
| author | Ling, Xitong Ping, Yifeng Li, Jiawen Peng, Jing Chen, Yuxuan Ouyang, Minxi Wang, Yizhi He, Yonghong Guan, Tian Liu, Xiaoping Zhu, Lianghui |
| author_facet | Ling, Xitong Ping, Yifeng Li, Jiawen Peng, Jing Chen, Yuxuan Ouyang, Minxi Wang, Yizhi He, Yonghong Guan, Tian Liu, Xiaoping Zhu, Lianghui |
| contents | Multiple Instance Learning (MIL) plays a significant role in computational pathology, enabling weakly supervised analysis of Whole Slide Image (WSI) datasets. The field of WSI analysis is confronted with a severe long-tailed distribution problem, which significantly impacts the performance of classifiers. Long-tailed distributions lead to class imbalance, where some classes have sparse samples while others are abundant, making it difficult for classifiers to accurately identify minority class samples. To address this issue, we propose an ensemble learning method based on MIL, which employs expert decoders with shared aggregators and consistency constraints to learn diverse distributions and reduce the impact of class imbalance on classifier performance. Moreover, we introduce a multimodal distillation framework that leverages text encoders pre-trained on pathology-text pairs to distill knowledge and guide the MIL aggregator in capturing stronger semantic features relevant to class information. To ensure flexibility, we use learnable prompts to guide the distillation process of the pre-trained text encoder, avoiding limitations imposed by specific prompts. Our method, MDE-MIL, integrates multiple expert branches focusing on specific data distributions to address long-tailed issues. Consistency control ensures generalization across classes. Multimodal distillation enhances feature extraction. Experiments on Camelyon+-LT and PANDA-LT datasets show it outperforms state-of-the-art methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_00915 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multimodal Distillation-Driven Ensemble Learning for Long-Tailed Histopathology Whole Slide Images Analysis Ling, Xitong Ping, Yifeng Li, Jiawen Peng, Jing Chen, Yuxuan Ouyang, Minxi Wang, Yizhi He, Yonghong Guan, Tian Liu, Xiaoping Zhu, Lianghui Computer Vision and Pattern Recognition Artificial Intelligence Multiple Instance Learning (MIL) plays a significant role in computational pathology, enabling weakly supervised analysis of Whole Slide Image (WSI) datasets. The field of WSI analysis is confronted with a severe long-tailed distribution problem, which significantly impacts the performance of classifiers. Long-tailed distributions lead to class imbalance, where some classes have sparse samples while others are abundant, making it difficult for classifiers to accurately identify minority class samples. To address this issue, we propose an ensemble learning method based on MIL, which employs expert decoders with shared aggregators and consistency constraints to learn diverse distributions and reduce the impact of class imbalance on classifier performance. Moreover, we introduce a multimodal distillation framework that leverages text encoders pre-trained on pathology-text pairs to distill knowledge and guide the MIL aggregator in capturing stronger semantic features relevant to class information. To ensure flexibility, we use learnable prompts to guide the distillation process of the pre-trained text encoder, avoiding limitations imposed by specific prompts. Our method, MDE-MIL, integrates multiple expert branches focusing on specific data distributions to address long-tailed issues. Consistency control ensures generalization across classes. Multimodal distillation enhances feature extraction. Experiments on Camelyon+-LT and PANDA-LT datasets show it outperforms state-of-the-art methods. |
| title | Multimodal Distillation-Driven Ensemble Learning for Long-Tailed Histopathology Whole Slide Images Analysis |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.00915 |