In the Picture: Medical Imaging Datasets, Artifacts, and their Living Review
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916968592834560 |
|---|---|
| author | Jiménez-Sánchez, Amelia Avlona, Natalia-Rozalia de Boer, Sarah Campello, Víctor M. Feragen, Aasa Ferrante, Enzo Ganz, Melanie Gichoya, Judy Wawira González, Camila Groefsema, Steff Hering, Alessa Hulman, Adam Joskowicz, Leo Juodelyte, Dovile Kandemir, Melih Kooi, Thijs Lérida, Jorge del Pozo Li, Livie Yumeng Pacheco, Andre Rädsch, Tim Reyes, Mauricio Sourget, Théo van Ginneken, Bram Wen, David Weng, Nina Xu, Jack Junchi Zając, Hubert Dariusz Zuluaga, Maria A. Cheplygina, Veronika |
| author_facet | Jiménez-Sánchez, Amelia Avlona, Natalia-Rozalia de Boer, Sarah Campello, Víctor M. Feragen, Aasa Ferrante, Enzo Ganz, Melanie Gichoya, Judy Wawira González, Camila Groefsema, Steff Hering, Alessa Hulman, Adam Joskowicz, Leo Juodelyte, Dovile Kandemir, Melih Kooi, Thijs Lérida, Jorge del Pozo Li, Livie Yumeng Pacheco, Andre Rädsch, Tim Reyes, Mauricio Sourget, Théo van Ginneken, Bram Wen, David Weng, Nina Xu, Jack Junchi Zając, Hubert Dariusz Zuluaga, Maria A. Cheplygina, Veronika |
| contents | Datasets play a critical role in medical imaging research, yet issues such as label quality, shortcuts, and metadata are often overlooked. This lack of attention may harm the generalizability of algorithms and, consequently, negatively impact patient outcomes. While existing medical imaging literature reviews mostly focus on machine learning (ML) methods, with only a few focusing on datasets for specific applications, these reviews remain static -- they are published once and not updated thereafter. This fails to account for emerging evidence, such as biases, shortcuts, and additional annotations that other researchers may contribute after the dataset is published. We refer to these newly discovered findings of datasets as research artifacts. To address this gap, we propose a living review that continuously tracks public datasets and their associated research artifacts across multiple medical imaging applications. Our approach includes a framework for the living review to monitor data documentation artifacts, and an SQL database to visualize the citation relationships between research artifact and dataset. Lastly, we discuss key considerations for creating medical imaging datasets, review best practices for data annotation, discuss the significance of shortcuts and demographic diversity, and emphasize the importance of managing datasets throughout their entire lifecycle. Our demo is publicly available at http://inthepicture.itu.dk/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_10727 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | In the Picture: Medical Imaging Datasets, Artifacts, and their Living Review Jiménez-Sánchez, Amelia Avlona, Natalia-Rozalia de Boer, Sarah Campello, Víctor M. Feragen, Aasa Ferrante, Enzo Ganz, Melanie Gichoya, Judy Wawira González, Camila Groefsema, Steff Hering, Alessa Hulman, Adam Joskowicz, Leo Juodelyte, Dovile Kandemir, Melih Kooi, Thijs Lérida, Jorge del Pozo Li, Livie Yumeng Pacheco, Andre Rädsch, Tim Reyes, Mauricio Sourget, Théo van Ginneken, Bram Wen, David Weng, Nina Xu, Jack Junchi Zając, Hubert Dariusz Zuluaga, Maria A. Cheplygina, Veronika Computer Vision and Pattern Recognition Artificial Intelligence Computers and Society Digital Libraries Image and Video Processing Datasets play a critical role in medical imaging research, yet issues such as label quality, shortcuts, and metadata are often overlooked. This lack of attention may harm the generalizability of algorithms and, consequently, negatively impact patient outcomes. While existing medical imaging literature reviews mostly focus on machine learning (ML) methods, with only a few focusing on datasets for specific applications, these reviews remain static -- they are published once and not updated thereafter. This fails to account for emerging evidence, such as biases, shortcuts, and additional annotations that other researchers may contribute after the dataset is published. We refer to these newly discovered findings of datasets as research artifacts. To address this gap, we propose a living review that continuously tracks public datasets and their associated research artifacts across multiple medical imaging applications. Our approach includes a framework for the living review to monitor data documentation artifacts, and an SQL database to visualize the citation relationships between research artifact and dataset. Lastly, we discuss key considerations for creating medical imaging datasets, review best practices for data annotation, discuss the significance of shortcuts and demographic diversity, and emphasize the importance of managing datasets throughout their entire lifecycle. Our demo is publicly available at http://inthepicture.itu.dk/. |
| title | In the Picture: Medical Imaging Datasets, Artifacts, and their Living Review |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Computers and Society Digital Libraries Image and Video Processing |
| url | https://arxiv.org/abs/2501.10727 |