Evidential Concept Embedding Models: Towards Reliable Concept Explanations for Skin Disease Diagnosis

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Gao, Yibo, Gao, Zheyao, Gao, Xin, Liu, Yuanye, Wang, Bomin, Zhuang, Xiahai
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909232498999296
author Gao, Yibo
Gao, Zheyao
Gao, Xin
Liu, Yuanye
Wang, Bomin
Zhuang, Xiahai
author_facet Gao, Yibo
Gao, Zheyao
Gao, Xin
Liu, Yuanye
Wang, Bomin
Zhuang, Xiahai
contents Due to the high stakes in medical decision-making, there is a compelling demand for interpretable deep learning methods in medical image analysis. Concept Bottleneck Models (CBM) have emerged as an active interpretable framework incorporating human-interpretable concepts into decision-making. However, their concept predictions may lack reliability when applied to clinical diagnosis, impeding concept explanations' quality. To address this, we propose an evidential Concept Embedding Model (evi-CEM), which employs evidential learning to model the concept uncertainty. Additionally, we offer to leverage the concept uncertainty to rectify concept misalignments that arise when training CBMs using vision-language models without complete concept supervision. With the proposed methods, we can enhance concept explanations' reliability for both supervised and label-efficient settings. Furthermore, we introduce concept uncertainty for effective test-time intervention. Our evaluation demonstrates that evi-CEM achieves superior performance in terms of concept prediction, and the proposed concept rectification effectively mitigates concept misalignments for label-efficient training. Our code is available at https://github.com/obiyoag/evi-CEM.
format Preprint
id arxiv_https___arxiv_org_abs_2406_19130
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Evidential Concept Embedding Models: Towards Reliable Concept Explanations for Skin Disease Diagnosis
Gao, Yibo
Gao, Zheyao
Gao, Xin
Liu, Yuanye
Wang, Bomin
Zhuang, Xiahai
Computer Vision and Pattern Recognition
Due to the high stakes in medical decision-making, there is a compelling demand for interpretable deep learning methods in medical image analysis. Concept Bottleneck Models (CBM) have emerged as an active interpretable framework incorporating human-interpretable concepts into decision-making. However, their concept predictions may lack reliability when applied to clinical diagnosis, impeding concept explanations' quality. To address this, we propose an evidential Concept Embedding Model (evi-CEM), which employs evidential learning to model the concept uncertainty. Additionally, we offer to leverage the concept uncertainty to rectify concept misalignments that arise when training CBMs using vision-language models without complete concept supervision. With the proposed methods, we can enhance concept explanations' reliability for both supervised and label-efficient settings. Furthermore, we introduce concept uncertainty for effective test-time intervention. Our evaluation demonstrates that evi-CEM achieves superior performance in terms of concept prediction, and the proposed concept rectification effectively mitigates concept misalignments for label-efficient training. Our code is available at https://github.com/obiyoag/evi-CEM.
title Evidential Concept Embedding Models: Towards Reliable Concept Explanations for Skin Disease Diagnosis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.19130