The Association Between Heatmap Position and the Diagnostic Accuracy of Artificial Intelligence for Colorectal Polyp Diagnosis
Higher percentages of heatmap covering the polyp were associated with correct diagnoses of four AI algorithms. This indicates that it is clinically relevant to strive for AI predictions with heatmaps covering as much colorectal polyp tissue as possible. Knowing how to interpret heatmaps could increase trust in AI and, with that, benefit the implementation of AI in clinical practice.
Authors
- Ayla Thijssen (ORCID: https://orcid.org/0000-0002-4488-4333)
- Ankie Reumkens (ORCID: https://orcid.org/0000-0002-3326-3793)
- Erik J. Schoon (ORCID: https://orcid.org/0000-0003-4515-1235)
- Nikoo Dehghani (ORCID: https://orcid.org/0009-0008-7225-3118)
- Peter H. N. de With (ORCID: https://orcid.org/0000-0002-7639-7716)
- Eveline Rondagh
- Ruud W.M. Schrauwen (ORCID: https://orcid.org/0000-0002-4680-3699)
- Quirine E. W. van der Zander (ORCID: https://orcid.org/0000-0002-8640-5521)
- Khalida Soufidi
- Björn Winkens (ORCID: https://orcid.org/0000-0002-6747-6228)
- Eric T.P. Keulen (ORCID: https://orcid.org/0000-0001-6666-8773)
- Fons van der Sommen (ORCID: https://orcid.org/0000-0002-3593-2356)
- Paul H. A. Bours
- Mark H. P. van Avesaat
Institutions
- Radboud University Nijmegen (NL)
- Catharina Ziekenhuis (NL)
- University Medical Center (US)
- Maastricht University (NL)
- Zuyderland Medisch Centrum (NL)
- Ziekenhuis Bernhoven (NL)
- Eindhoven University of Technology (NL)
Publication Details
- Journal
- Cancers
- Published
- 2025-05-10
- DOI
- https://doi.org/10.3390/cancers17101620
- Citations
- 1
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 1.36
Funders
- Maastricht Universitair Medisch Centrum
- KWF Kankerbestrijding