التفاصيل البيبلوغرافية
العنوان: |
Compositionally Equivariant Representation Learning. |
المؤلفون: |
Liu X, Sanchez P, Thermos S, O'Neil AQ, Tsaftaris SA |
المصدر: |
IEEE transactions on medical imaging [IEEE Trans Med Imaging] 2024 Jun; Vol. 43 (6), pp. 2169-2179. Date of Electronic Publication: 2024 Jun 03. |
نوع المنشور: |
Journal Article |
اللغة: |
English |
بيانات الدورية: |
Publisher: Institute of Electrical and Electronics Engineers Country of Publication: United States NLM ID: 8310780 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1558-254X (Electronic) Linking ISSN: 02780062 NLM ISO Abbreviation: IEEE Trans Med Imaging Subsets: MEDLINE |
أسماء مطبوعة: |
Original Publication: New York, NY : Institute of Electrical and Electronics Engineers, c1982- |
مواضيع طبية MeSH: |
Brain*/diagnostic imaging , Deep Learning* , Magnetic Resonance Imaging*/methods , Image Processing, Computer-Assisted*/methods , Algorithms*, Humans ; Tomography, X-Ray Computed/methods |
مستخلص: |
Deep learning models often need sufficient supervision (i.e., labelled data) in order to be trained effectively. By contrast, humans can swiftly learn to identify important anatomy in medical images like MRI and CT scans, with minimal guidance. This recognition capability easily generalises to new images from different medical facilities and to new tasks in different settings. This rapid and generalisable learning ability is largely due to the compositional structure of image patterns in the human brain, which are not well represented in current medical models. In this paper, we study the utilisation of compositionality in learning more interpretable and generalisable representations for medical image segmentation. Overall, we propose that the underlying generative factors that are used to generate the medical images satisfy compositional equivariance property, where each factor is compositional (e.g., corresponds to human anatomy) and also equivariant to the task. Hence, a good representation that approximates well the ground truth factor has to be compositionally equivariant. By modelling the compositional representations with learnable von-Mises-Fisher (vMF) kernels, we explore how different design and learning biases can be used to enforce the representations to be more compositionally equivariant under un-, weakly-, and semi-supervised settings. Extensive results show that our methods achieve the best performance over several strong baselines on the task of semi-supervised domain-generalised medical image segmentation. Code will be made publicly available upon acceptance at https://github.com/vios-s. |
تواريخ الأحداث: |
Date Created: 20240126 Date Completed: 20240603 Latest Revision: 20240604 |
رمز التحديث: |
20240604 |
DOI: |
10.1109/TMI.2024.3358955 |
PMID: |
38277249 |
قاعدة البيانات: |
MEDLINE |