| Creators: |
Müller, Dominik and Meyer, Philip and Rentschler, Lukas and Manz, Robin and Hieber, Daniel and Bäcker, Jonas and Cramer, Samantha and Wengenmayr, Christoph and Märkl, Bruno and Huss, Ralf and Kramer, Frank and Soto-Rey, Inaki and Raffler, Johannes |
| Abstract (ENG): |
Prostate cancer is a dominant health concern calling for advanced diagnostic tools. Utilizing digital pathology and artificial intelligence, this study explores the potential of 11 deep neural network architectures for automated Gleason grading in prostate carcinoma focusing on comparing traditional and recent architectures. A standardized image classification pipeline, based on the AUCMEDI framework, facilitated robust evaluation using an in-house dataset consisting of 34,264 annotated tissue tiles. The results indicated varying sensitivity across architectures, with ConvNeXt demonstrating the strongest performance. Notably, newer architectures achieved superior performance, even though with challenges in
differentiating closely related Gleason grades. The ConvNeXt model was capable of learning a balance between complexity and generalizability. Overall, this study lays the groundwork for enhanced Gleason grading systems, potentially improving diagnostic efficiency for prostate cancer. |
| Citation: |
Müller, Dominik and Meyer, Philip and Rentschler, Lukas and Manz, Robin and Hieber, Daniel and Bäcker, Jonas and Cramer, Samantha and Wengenmayr, Christoph and Märkl, Bruno and Huss, Ralf and Kramer, Frank and Soto-Rey, Inaki and Raffler, Johannes
(2024)
Assessing the Performance of Deep Learning for Automated Gleason Grading in Prostate Cancer.
In: (Proceedings of the) Medical Informatics Europe Conference (MIE), September, 25-29, 2024, Athens, Greece, pp. 1110-1114.
(Studies in Health Technology and Informatics; 316).
ISBN 9781643685335
|
