13 March 2025 in The Journal of Pathology: Clinical Research
This work aimed to evaluate both the usefulness and user acceptance of five gradient‐based explainable artificial intelligence (XAI) methods in the use case of a prostate carcinoma clinical decision support system environment. In addition, we aimed to determine whether XAI helps to increase the acceptance of artificial intelligence (AI) and recommend a particular method for this use case. The evaluation was conducted on a tool developed in‐house with different visualization approaches to the AI‐generated Gleason grade and the corresponding XAI explanations on top of the original slide. The study was a heuristic evaluation of five XAI methods. The participants were 15 pathologists from the University Hospital of Augsburg with a wide range of experience in Gleason grading and AI. The evaluation consisted of a user information form, short questionnaires on each XAI method, a ranking of the methods, and a general questionnaire to evaluate the performance and usefulness of the AI. There were significant differences between the ratings of the methods, with Grad‐CAM++ performing best. Both AI decision support and XAI explanations were seen as helpful by the majority of participants. In conclusion, our pilot study suggests that the evaluated XAI methods can indeed improve the usefulness and acceptance of AI. The results obtained are a good indicator, but further studies involving larger sample sizes are warranted to draw more definitive conclusions.
17 March 2026 in Histopathology
Aim: Whole slide imaging (WSI) scanners are fundamental to digital pathology, providing high‐resolution digitized histology slides. While scanner models vary in throughput, usability and image quality, the impact of these differences on pathologists' diagnostic performance remains insufficiently understood. This study evaluates the perceived image quality and diagnostic sufficiency of four commonly used high‐throughput WSI scanners. Methods and results: We selected four bright‐field WSI scanners: Leica's Aperio GT 450 DX, Hamamatsu's NanoZoomer S360, 3DHISTECH's Pannoramic 1000 DX and Philips' IntelliSite Ultra Fast Scanner (first generation). A set of 601 slides was digitized on each scanner, and 96 representative slides were selected for evaluation. Fourteen pathologists from the University Hospital of Augsburg's Institute of Pathology assessed sharpness, brightness, contrast, colour realism and artefact frequency. To evaluate diagnostic sufficiency, pathologists were asked whether each scanned image was adequate for making a precise diagnosis when considering brightness/contrast/colour realism, sharpness and artefact frequency. Significant differences in perceived quality were observed across scanners, yet diagnostic sufficiency was rated more positively overall. Here, the Aperio GT 450 DX and NanoZoomer S360 scanners were ranked equally, each achieving the highest median rating for all image quality aspects. The Pannoramic 1000 DX and IntelliSite UFS scanners closely followed with deficits in colour realism and sharpness, respectively. Depending on the scanner model, 4.2% to 22.2% of images were deemed insufficient for diagnosis, with performance varying by tissue type and staining method. Conclusion: Scanner differences in diagnostic sufficiency ratings were less pronounced than differences in perceived image quality. These findings highlight the importance of selecting appropriate scanning solutions based on workload and tissue characteristics.