Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap!
Abstract Good practices in artificial intelligence (AI) model validation are key for achieving trustworthy AI. Within the cancer imaging domain, attracting the attention of clinical and technical AI enthusiasts, this work discusses current gaps in AI validation strategies, examining existing practic...
Saved in:
Main Authors: | , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2025-01-01
|
Series: | European Radiology Experimental |
Subjects: | |
Online Access: | https://doi.org/10.1186/s41747-024-00543-0 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
_version_ | 1832594949403574272 |
---|---|
author | Ioanna Chouvarda Sara Colantonio Ana S. C. Verde Ana Jimenez-Pastor Leonor Cerdá-Alberich Yannick Metz Lithin Zacharias Shereen Nabhani-Gebara Maciej Bobowicz Gianna Tsakou Karim Lekadir Manolis Tsiknakis Luis Martí-Bonmati Nikolaos Papanikolaou |
author_facet | Ioanna Chouvarda Sara Colantonio Ana S. C. Verde Ana Jimenez-Pastor Leonor Cerdá-Alberich Yannick Metz Lithin Zacharias Shereen Nabhani-Gebara Maciej Bobowicz Gianna Tsakou Karim Lekadir Manolis Tsiknakis Luis Martí-Bonmati Nikolaos Papanikolaou |
author_sort | Ioanna Chouvarda |
collection | DOAJ |
description | Abstract Good practices in artificial intelligence (AI) model validation are key for achieving trustworthy AI. Within the cancer imaging domain, attracting the attention of clinical and technical AI enthusiasts, this work discusses current gaps in AI validation strategies, examining existing practices that are common or variable across technical groups (TGs) and clinical groups (CGs). The work is based on a set of structured questions encompassing several AI validation topics, addressed to professionals working in AI for medical imaging. A total of 49 responses were obtained and analysed to identify trends and patterns. While TGs valued transparency and traceability the most, CGs pointed out the importance of explainability. Among the topics where TGs may benefit from further exposure are stability and robustness checks, and mitigation of fairness issues. On the other hand, CGs seemed more reluctant towards synthetic data for validation and would benefit from exposure to cross-validation techniques, or segmentation metrics. Topics emerging from the open questions were utility, capability, adoption and trustworthiness. These findings on current trends in AI validation strategies may guide the creation of guidelines necessary for training the next generation of professionals working with AI in healthcare and contribute to bridging any technical-clinical gap in AI validation. Relevance statement This study recognised current gaps in understanding and applying AI validation strategies in cancer imaging and helped promote trust and adoption for interdisciplinary teams of technical and clinical researchers. Key Points Clinical and technical researchers emphasise interpretability, external validation with diverse data, and bias awareness in AI validation for cancer imaging. In cancer imaging AI research, clinical researchers prioritise explainability, while technical researchers focus on transparency and traceability, and see potential in synthetic datasets. Researchers advocate for greater homogenisation of AI validation practices in cancer imaging. Graphical Abstract |
format | Article |
id | doaj-art-041704d8640a4525af12caf8c56f4cd8 |
institution | Kabale University |
issn | 2509-9280 |
language | English |
publishDate | 2025-01-01 |
publisher | SpringerOpen |
record_format | Article |
series | European Radiology Experimental |
spelling | doaj-art-041704d8640a4525af12caf8c56f4cd82025-01-19T12:09:30ZengSpringerOpenEuropean Radiology Experimental2509-92802025-01-019111510.1186/s41747-024-00543-0Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap!Ioanna Chouvarda0Sara Colantonio1Ana S. C. Verde2Ana Jimenez-Pastor3Leonor Cerdá-Alberich4Yannick Metz5Lithin Zacharias6Shereen Nabhani-Gebara7Maciej Bobowicz8Gianna Tsakou9Karim Lekadir10Manolis Tsiknakis11Luis Martí-Bonmati12Nikolaos Papanikolaou13School of Medicine, Aristotle University of ThessalonikiInstitute of Information Science and Technologies of the National Research Council of ItalyComputational Clinical Imaging Group (CCIG), Champalimaud Research, Champalimaud FoundationQuibim SLBiomedical Imaging Research Group (GIBI230), La Fe Health Research InstituteData Analysis and Visualization, University of KonstanzDepartment of Pharmacy, Kingston University LondonFaculty of Health, Science, Social Care & Education, Kingston University London2nd Department of Radiology, Medical University of GdanskResearch and Development Lab, Gruppo Maggioli Greek BranchDepartament de Matemàtiques i Informàtica, Artificial Intelligence in Medicine Lab (BCN-AIM), Universitat de BarcelonaComputational BioMedicine Laboratory (CBML), Foundation for Research and Technology-Hellas (FORTH)Biomedical Imaging Research Group (GIBI230), La Fe Health Research InstituteComputational Clinical Imaging Group (CCIG), Champalimaud Research, Champalimaud FoundationAbstract Good practices in artificial intelligence (AI) model validation are key for achieving trustworthy AI. Within the cancer imaging domain, attracting the attention of clinical and technical AI enthusiasts, this work discusses current gaps in AI validation strategies, examining existing practices that are common or variable across technical groups (TGs) and clinical groups (CGs). The work is based on a set of structured questions encompassing several AI validation topics, addressed to professionals working in AI for medical imaging. A total of 49 responses were obtained and analysed to identify trends and patterns. While TGs valued transparency and traceability the most, CGs pointed out the importance of explainability. Among the topics where TGs may benefit from further exposure are stability and robustness checks, and mitigation of fairness issues. On the other hand, CGs seemed more reluctant towards synthetic data for validation and would benefit from exposure to cross-validation techniques, or segmentation metrics. Topics emerging from the open questions were utility, capability, adoption and trustworthiness. These findings on current trends in AI validation strategies may guide the creation of guidelines necessary for training the next generation of professionals working with AI in healthcare and contribute to bridging any technical-clinical gap in AI validation. Relevance statement This study recognised current gaps in understanding and applying AI validation strategies in cancer imaging and helped promote trust and adoption for interdisciplinary teams of technical and clinical researchers. Key Points Clinical and technical researchers emphasise interpretability, external validation with diverse data, and bias awareness in AI validation for cancer imaging. In cancer imaging AI research, clinical researchers prioritise explainability, while technical researchers focus on transparency and traceability, and see potential in synthetic datasets. Researchers advocate for greater homogenisation of AI validation practices in cancer imaging. Graphical Abstracthttps://doi.org/10.1186/s41747-024-00543-0Artificial intelligenceDiagnostic imagingNeoplasmsResearch designSurveys and questionnaires |
spellingShingle | Ioanna Chouvarda Sara Colantonio Ana S. C. Verde Ana Jimenez-Pastor Leonor Cerdá-Alberich Yannick Metz Lithin Zacharias Shereen Nabhani-Gebara Maciej Bobowicz Gianna Tsakou Karim Lekadir Manolis Tsiknakis Luis Martí-Bonmati Nikolaos Papanikolaou Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap! European Radiology Experimental Artificial intelligence Diagnostic imaging Neoplasms Research design Surveys and questionnaires |
title | Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap! |
title_full | Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap! |
title_fullStr | Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap! |
title_full_unstemmed | Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap! |
title_short | Differences in technical and clinical perspectives on AI validation in cancer imaging: mind the gap! |
title_sort | differences in technical and clinical perspectives on ai validation in cancer imaging mind the gap |
topic | Artificial intelligence Diagnostic imaging Neoplasms Research design Surveys and questionnaires |
url | https://doi.org/10.1186/s41747-024-00543-0 |
work_keys_str_mv | AT ioannachouvarda differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT saracolantonio differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT anascverde differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT anajimenezpastor differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT leonorcerdaalberich differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT yannickmetz differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT lithinzacharias differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT shereennabhanigebara differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT maciejbobowicz differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT giannatsakou differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT karimlekadir differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT manolistsiknakis differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT luismartibonmati differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap AT nikolaospapanikolaou differencesintechnicalandclinicalperspectivesonaivalidationincancerimagingmindthegap |