1. Recommendations on test datasets for evaluating AI solutions in pathology
- Author
-
Homeyer, André, Geißler, Christian, Schwen, Lars Ole, Zakrzewski, Falk, Evans, Theodore, Strohmenger, Klaus, Westphal, Max, Bülow, Roman David, Kargl, Michaela, Karjauv, Aray, Munné-Bertran, Isidre, Retzlaff, Carl Orge, Romero-López, Adrià, Sołtysiński, Tomasz, Plass, Markus, Carvalho, Rita, Steinbach, Peter, Lan, Yu-Chia, Bouteldja, Nassim, Haber, David, Rojas-Carulla, Mateo, Sadr, Alireza Vafaei, Kraft, Matthias, Krüger, Daniel, Fick, Rutger, Lang, Tobias, Boor, Peter, Müller, Heimo, Hufnagl, Peter, and Zerbe, Norman
- Subjects
FOS: Computer and information sciences ,Computer Science - Machine Learning ,Artificial Intelligence (cs.AI) ,Computer Science - Artificial Intelligence ,Computer Vision and Pattern Recognition (cs.CV) ,Image and Video Processing (eess.IV) ,Computer Science - Computer Vision and Pattern Recognition ,FOS: Electrical engineering, electronic engineering, information engineering ,FOS: Physical sciences ,Medical Physics (physics.med-ph) ,Electrical Engineering and Systems Science - Image and Video Processing ,Physics - Medical Physics ,Machine Learning (cs.LG) - Abstract
Artificial intelligence (AI) solutions that automatically extract information from digital histology images have shown great promise for improving pathological diagnosis. Prior to routine use, it is important to evaluate their predictive performance and obtain regulatory approval. This assessment requires appropriate test datasets. However, compiling such datasets is challenging and specific recommendations are missing. A committee of various stakeholders, including commercial AI developers, pathologists, and researchers, discussed key aspects and conducted extensive literature reviews on test datasets in pathology. Here, we summarize the results and derive general recommendations for the collection of test datasets. We address several questions: Which and how many images are needed? How to deal with low-prevalence subsets? How can potential bias be detected? How should datasets be reported? What are the regulatory requirements in different countries? The recommendations are intended to help AI developers demonstrate the utility of their products and to help regulatory agencies and end users verify reported performance measures. Further research is needed to formulate criteria for sufficiently representative test datasets so that AI solutions can operate with less user intervention and better support diagnostic workflows in the future.
- Published
- 2022
- Full Text
- View/download PDF