We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Multiple Instance Pathology Image Diagnosis Model based on Channel Attention and Data Augmentation.
- Authors
Wan, Tianjiang; Tian, Jingmin; Wei, Ping; Li, Junli
- Abstract
The application of machine learning in the medical field has resulted in significant advancements in computer-aided pathological diagnosis. Multiple instance learning (MIL) has emerged as a promising approach for pathological image classification, particularly in scenarios where local annotations are lacking. However, current MIL models often overlook the importance of feature weights in the channel dimension and struggle with imbalanced positive and negative data. To address these limitations, an integration of a channel attention (CA) module and an augmented data (AUG) mechanism into the MIL model is proposed, resulting in improved performance. The CA module dynamically assigns weights to example features in the channel dimension, enhancing or suppressing features adaptively. Additionally, the AUG mechanism effectively balances the distribution of positive and negative data, significantly reducing false negatives. Through ablation experiments, the contributions of the CA module and AUG mechanism in enhancing the overall model performance are analyzed. Experimental validations on the CAMELYON16/17 public pathological image datasets demonstrate that the proposed model and method outperform existing approaches, with particular emphasis on reducing false negatives. Highlights: While multiple instance learning (MIL) models address scarce local annotations in pathological images, existing methods often neglect feature weights in the channel dimension and struggle with imbalanced data. This study introduces a novel approach by proposing a channel attention (CA) module and data augmentation mechanism for integration into the MIL model, significantly improving its performance. By conducting carefully designed ablation experiments, this study probes into the precise contributions of the introduced CA module and data augmentation mechanism towards enhancing the performance of the MIL model. Through validation on public datasets reveals that the proposed method excels over the latest methods by notably reducing false negatives, thereby enhancing sensitivity metrics.
- Subjects
DATA augmentation; IMAGE recognition (Computer vision); COMPUTER-aided diagnosis; PATHOLOGY; DIAGNOSIS; COINTEGRATION; MACHINE learning
- Publication
Discover Applied Sciences, 2024, Vol 6, Issue 9, p1
- ISSN
3004-9261
- Publication type
Article
- DOI
10.1007/s42452-024-06093-9