Distil or Cluster?: Data-efficient Learning for Ultrasound in Practice
Ochmann J, Müller J, Erick F, Kainz B (2026)
Publication Type: Conference contribution
Publication year: 2026
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Pages Range: 170-176
Conference Proceedings Title: Informatik aktuell
Event location: Lübeck
ISBN: 9783658510992
DOI: 10.1007/978-3-658-51100-5_36
Abstract
Training deep learning models on ultrasound videos is like drinking from a firehose, most frames are redundant or noisy, yet drive high computational cost. We ask: can we learn just as well from less? We benchmark coreset construction, clustering, dataset distillation, and random sampling on large-scale echocardiography datasets. Clustering-based coresets, especially those using CNN embeddings,Wasserstein distance, and two-pass DBSCAN, match or surpass full-data training while reducing cost by up to 15×. They also rival dataset distillation in accuracy but are up to 30× faster. Surprisingly, random subsets sometimes outperform engineered coresets, with macro-F1 peaking at just 5% of the data. These results show that less can be more, offering a scalable path to efficient, fair training in medical video analysis.
Authors with CRIS profile
Johanna Müller
Professur für Image Data Exploration and Analysis
Franciskus Erick
Professur für Image Data Exploration and Analysis
Bernhard Kainz
Professur für Image Data Exploration and Analysis
How to cite
APA:
Ochmann, J., Müller, J., Erick, F., & Kainz, B. (2026). Distil or Cluster?: Data-efficient Learning for Ultrasound in Practice. In Heinz Handels, Katharina Breininger, Thomas Deserno, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 170-176). Lübeck, DE: Springer Science and Business Media Deutschland GmbH.
MLA:
Ochmann, Jennifer, et al. "Distil or Cluster?: Data-efficient Learning for Ultrasound in Practice." Proceedings of the Bildverarbeitung für die Medizin Workshop, BVM 2026, Lübeck Ed. Heinz Handels, Katharina Breininger, Thomas Deserno, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff, Springer Science and Business Media Deutschland GmbH, 2026. 170-176.
BibTeX: Download