Research output per year
Research output per year
Meike Nauta*, Johannes H. Hegeman, Jeroen Geerdink, Jörg Schlötterer, Maurice van Keulen, Christin Seifert
Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review
Part-prototype models are explainable-by-design image classifiers, and a promising alternative to black box AI. This paper explores the applicability and potential of interpretable machine learning, in particular PIP-Net, for automated diagnosis support on real-world medical imaging data. PIP-Net learns human-understandable prototypical image parts and we evaluate its accuracy and interpretability for fracture detection and skin cancer diagnosis. We find that PIP-Net’s decision making process is in line with medical classification standards, while only provided with image-level class labels. Because of PIP-Net’s unsupervised pretraining of prototypes, data quality problems such as undesired text in an X-ray or labelling errors can be easily identified. Additionally, we are the first to show that humans can manually correct the reasoning of PIP-Net by directly disabling undesired prototypes. We conclude that part-prototype models are promising for medical applications due to their interpretability and potential for advanced model debugging.
Original language | English |
---|---|
Title of host publication | Artificial Intelligence. ECAI 2023 International Workshops - XAI^3, TACTIFUL, XI-ML, SEDAMI, RAAIT, AI4S, HYDRA, AI4AI, 2023, Proceedings |
Editors | Sławomir Nowaczyk, Przemysław Biecek, Neo Christopher Chung, Mauro Vallati, Paweł Skruch, Joanna Jaworek-Korjakowska, Simon Parkinson, Alexandros Nikitas, Martin Atzmüller, Tomáš Kliegr, Ute Schmid, Szymon Bobek, Nada Lavrac, Marieke Peeters, Roland van Dierendonck, Saskia Robben, Eunika Mercier-Laurent, Gülgün Kayakutlu, Mieczyslaw Lech Owoc, Karl Mason, Abdul Wahid, Pierangela Bruno, Francesco Calimeri, Francesco Cauteruccio, Giorgio Terracina, Diedrich Wolter, Jochen L. Leidner, Michael Kohlhase, Vania Dimitrova |
Publisher | Springer |
Pages | 198-215 |
Number of pages | 18 |
ISBN (Print) | 9783031503955 |
DOIs | |
Publication status | Published - 21 Jan 2024 |
Event | 26th European Conference on Artificial Intelligence, ECAI 2023 - Kraków, Poland Duration: 30 Sept 2023 → 4 Oct 2023 Conference number: 26 |
Name | Communications in Computer and Information Science |
---|---|
Volume | 1947 |
ISSN (Print) | 1865-0929 |
ISSN (Electronic) | 1865-0937 |
Conference | 26th European Conference on Artificial Intelligence, ECAI 2023 |
---|---|
Abbreviated title | ECAI |
Country/Territory | Poland |
City | Kraków |
Period | 30/09/23 → 4/10/23 |
Research output: Working paper › Preprint › Academic