Flat-Aware Cross-Stage Distilled Framework for Imbalanced Medical Image Classification

Li, Jinpeng; Chen, Guangyong; Mao, Hangyu; Deng, Danruo; Liu, Dong; Hao, Jianye; Dou, Qi; Heng, Pheng‐Ann

doi:10.1007/978-3-031-16437-8_21

Cited by 7 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ozturk et al proposed to decouple the learning of features and the classifier: this method uses the deep clustering method to obtain features with maximum class separation and then learns the classifier by keeping the class marginal [42]. Similarly, Li et al proposed cross-staged distilling method to prevent the classifier from being biased based on the learned features [67]. The attention mechanism was also leveraged to exploit class-agnostic global attention feature maps for the imbalanced medical data [68].…”

Section: Class Imbalancementioning

confidence: 99%

Uncertainty-Aware Health Diagnostics via Class-Balanced Evidential Deep Learning

Xia,

Dang,

Han

et al. 2024

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Uncertainty quantification is critical for ensuring the safety of deep learning-enabled health diagnostics, as it helps the model account for unknown factors and reduces the risk of misdiagnosis. However, existing uncertainty quantification studies often overlook the significant issue of class imbalance, which is common in medical data. In this paper, we propose a class-balanced evidential deep learning framework to achieve fair and reliable uncertainty estimates for health diagnostic models. This framework advances the state-ofthe-art uncertainty quantification method of evidential deep learning with two novel mechanisms to address the challenges posed by class imbalance. Specifically, we introduce a pooling loss that enables the model to learn less biased evidence among classes and a learnable prior to regularize the posterior distribution that accounts for the quality of uncertainty estimates. Extensive experiments using benchmark data with varying degrees of imbalance and various naturally imbalanced health data demonstrate the effectiveness and superiority of our method. Our work pushes the envelope of uncertainty quantification from theoretical studies to realistic healthcare application scenarios. By enhancing uncertainty estimation for class-imbalanced data, we contribute to the development of more reliable and practical deep learning-enabled health diagnostic systems. 1

show abstract

Section: Class Imbalancementioning

confidence: 99%