Predicting Scores of Medical Imaging Segmentation Methods with Meta-learning

Tom van Sonsbeek, Veronika Cheplygina

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)


Deep learning has led to state-of-the-art results for many medical imaging tasks, such as segmentation of different anatomical structures. With the increased numbers of deep learning publications and openly available code, the approach to choosing a model for a new task becomes more complicated, while time and (computational) resources are limited. A possible solution to choosing a model efficiently is meta-learning, a learning method in which prior performance of a model is used to predict the performance for new tasks. We investigate meta-learning for segmentation across ten datasets of different organs and modalities. We propose four ways to represent each dataset by meta-features: one based on statistical features of the images and three are based on deep learning features. We use support vector regression and deep neural networks to learn the relationship between the meta-features and prior model performance. On three external test datasets these methods give Dice scores within 0.10 of the true performance. These results demonstrate the potential of meta-learning in medical imaging.

Original languageEnglish
Title of host publicationInterpretable and Annotation-Efficient Learning for Medical Image Computing - 3rd International Workshop, iMIMIC 2020, 2nd International Workshop, MIL3iD 2020, and 5th International Workshop, LABELS 2020, Held in Conjunction with MICCAI 2020, Proceedings
EditorsJaime Cardoso, Wilson Silva, Ricardo Cruz, Hien Van Nguyen, Badri Roysam, Nicholas Heller, Pedro Henriques Abreu, Jose Pereira Amorim, Ivana Isgum, Vishal Patel, Kevin Zhou, Steve Jiang, Ngan Le, Khoa Luu, Raphael Sznitman, Veronika Cheplygina, Samaneh Abbasi, Diana Mateus, Emanuele Trucco
Number of pages12
ISBN (Print)9783030611651
Publication statusPublished - 2020
EventLABELS 2020 - Lima, Peru
Duration: 4 Oct 20208 Oct 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12446 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferenceLABELS 2020
Otherheld in conjunction with the 23rd International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2020


  • Feature extraction
  • Meta-learning
  • Segmentation


Dive into the research topics of 'Predicting Scores of Medical Imaging Segmentation Methods with Meta-learning'. Together they form a unique fingerprint.

Cite this