Salvato in:
| Autori principali: | , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2404.18279 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913415366180864 |
|---|---|
| author | Hong, Zesheng Yue, Yubiao Chen, Yubin Cong, Lele Lin, Huanjie Luo, Yuanmei Wang, Mini Han Wang, Weidong Xu, Jialong Yang, Xiaoqi Chen, Hechang Li, Zhenzhang Xie, Sihong |
| author_facet | Hong, Zesheng Yue, Yubiao Chen, Yubin Cong, Lele Lin, Huanjie Luo, Yuanmei Wang, Mini Han Wang, Weidong Xu, Jialong Yang, Xiaoqi Chen, Hechang Li, Zhenzhang Xie, Sihong |
| contents | Computer-aided diagnostics has benefited from the development of deep learning-based computer vision techniques in these years. Traditional supervised deep learning methods assume that the test sample is drawn from the identical distribution as the training data. However, it is possible to encounter out-of-distribution samples in real-world clinical scenarios, which may cause silent failure in deep learning-based medical image analysis tasks. Recently, research has explored various out-of-distribution (OOD) detection situations and techniques to enable a trustworthy medical AI system. In this survey, we systematically review the recent advances in OOD detection in medical image analysis. We first explore several factors that may cause a distributional shift when using a deep-learning-based model in clinic scenarios, with three different types of distributional shift well defined on top of these factors. Then a framework is suggested to categorize and feature existing solutions, while the previous studies are reviewed based on the methodology taxonomy. Our discussion also includes evaluation protocols and metrics, as well as the challenge and a research direction lack of exploration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_18279 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Out-of-distribution Detection in Medical Image Analysis: A survey Hong, Zesheng Yue, Yubiao Chen, Yubin Cong, Lele Lin, Huanjie Luo, Yuanmei Wang, Mini Han Wang, Weidong Xu, Jialong Yang, Xiaoqi Chen, Hechang Li, Zhenzhang Xie, Sihong Computer Vision and Pattern Recognition Computer-aided diagnostics has benefited from the development of deep learning-based computer vision techniques in these years. Traditional supervised deep learning methods assume that the test sample is drawn from the identical distribution as the training data. However, it is possible to encounter out-of-distribution samples in real-world clinical scenarios, which may cause silent failure in deep learning-based medical image analysis tasks. Recently, research has explored various out-of-distribution (OOD) detection situations and techniques to enable a trustworthy medical AI system. In this survey, we systematically review the recent advances in OOD detection in medical image analysis. We first explore several factors that may cause a distributional shift when using a deep-learning-based model in clinic scenarios, with three different types of distributional shift well defined on top of these factors. Then a framework is suggested to categorize and feature existing solutions, while the previous studies are reviewed based on the methodology taxonomy. Our discussion also includes evaluation protocols and metrics, as well as the challenge and a research direction lack of exploration. |
| title | Out-of-distribution Detection in Medical Image Analysis: A survey |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.18279 |