_version_ 1866916883667615744
author Christodoulou, Evangelia
Reinke, Annika
Andrè, Pascaline
Godau, Patrick
Kalinowski, Piotr
Houhou, Rola
Erkan, Selen
Sudre, Carole H.
Burgos, Ninon
Boutaj, Sofiène
Loizillon, Sophie
Solal, Maëlys
Cheplygina, Veronika
Heitz, Charles
Kozubek, Michal
Antonelli, Michela
Rieke, Nicola
Gilson, Antoine
Mayer, Leon D.
Tizabi, Minu D.
Cardoso, M. Jorge
Simpson, Amber
Kopp-Schneider, Annette
Varoquaux, Gaël
Colliot, Olivier
Maier-Hein, Lena
author_facet Christodoulou, Evangelia
Reinke, Annika
Andrè, Pascaline
Godau, Patrick
Kalinowski, Piotr
Houhou, Rola
Erkan, Selen
Sudre, Carole H.
Burgos, Ninon
Boutaj, Sofiène
Loizillon, Sophie
Solal, Maëlys
Cheplygina, Veronika
Heitz, Charles
Kozubek, Michal
Antonelli, Michela
Rieke, Nicola
Gilson, Antoine
Mayer, Leon D.
Tizabi, Minu D.
Cardoso, M. Jorge
Simpson, Amber
Kopp-Schneider, Annette
Varoquaux, Gaël
Colliot, Olivier
Maier-Hein, Lena
contents Performance comparisons are fundamental in medical imaging Artificial Intelligence (AI) research, often driving claims of superiority based on relative improvements in common performance metrics. However, such claims frequently rely solely on empirical mean performance. In this paper, we investigate whether newly proposed methods genuinely outperform the state of the art by analyzing a representative cohort of medical imaging papers. We quantify the probability of false claims based on a Bayesian approach that leverages reported results alongside empirically estimated model congruence to estimate whether the relative ranking of methods is likely to have occurred by chance. According to our results, the majority (>80%) of papers claims outperformance when introducing a new method. Our analysis further revealed a high probability (>5%) of false outperformance claims in 86% of classification papers and 53% of segmentation papers. These findings highlight a critical flaw in current benchmarking practices: claims of outperformance in medical imaging AI are frequently unsubstantiated, posing a risk of misdirecting future research efforts.
format Preprint
id arxiv_https___arxiv_org_abs_2505_04720
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle False Promises in Medical Imaging AI? Assessing Validity of Outperformance Claims
Christodoulou, Evangelia
Reinke, Annika
Andrè, Pascaline
Godau, Patrick
Kalinowski, Piotr
Houhou, Rola
Erkan, Selen
Sudre, Carole H.
Burgos, Ninon
Boutaj, Sofiène
Loizillon, Sophie
Solal, Maëlys
Cheplygina, Veronika
Heitz, Charles
Kozubek, Michal
Antonelli, Michela
Rieke, Nicola
Gilson, Antoine
Mayer, Leon D.
Tizabi, Minu D.
Cardoso, M. Jorge
Simpson, Amber
Kopp-Schneider, Annette
Varoquaux, Gaël
Colliot, Olivier
Maier-Hein, Lena
Computer Vision and Pattern Recognition
Performance comparisons are fundamental in medical imaging Artificial Intelligence (AI) research, often driving claims of superiority based on relative improvements in common performance metrics. However, such claims frequently rely solely on empirical mean performance. In this paper, we investigate whether newly proposed methods genuinely outperform the state of the art by analyzing a representative cohort of medical imaging papers. We quantify the probability of false claims based on a Bayesian approach that leverages reported results alongside empirically estimated model congruence to estimate whether the relative ranking of methods is likely to have occurred by chance. According to our results, the majority (>80%) of papers claims outperformance when introducing a new method. Our analysis further revealed a high probability (>5%) of false outperformance claims in 86% of classification papers and 53% of segmentation papers. These findings highlight a critical flaw in current benchmarking practices: claims of outperformance in medical imaging AI are frequently unsubstantiated, posing a risk of misdirecting future research efforts.
title False Promises in Medical Imaging AI? Assessing Validity of Outperformance Claims
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.04720