Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Self-Supervised Representation Learning for Dynamic Functional Connectivity With Subjectwise and Temporal Contrasts

Zhu, Jianfei ; An, Lijun LU orcid ; Zhu, Haiqi ; Jiang, Feng ; Liu, Shaohui ; Wei, Baichun and Yi, Chunzhi (2026) In IEEE Transactions on Neural Systems and Rehabilitation Engineering 34. p.2870-2882
Abstract

Dynamic functional connectivity (dFC) captures temporal dynamics in functional magnetic resonance imaging (fMRI) to better characterize brain activity, supporting multiple downstream analyses. However, effective modeling strategies for extracting representative signals from dFC remain underexplored. We proposed a novel Subject-wise and Temporal Contrastive Transformer (STCT) framework that advances dynamic connectivity modeling through a dual-constraint contrastive learning strategy. Our STCT framework uniquely integrates two objectives: 1) subject-wise contrast to preserve inter-individual specificity, and 2) temporal contrast to capture dynamic dependencies, which is a critical dimension overlooked in prior methods. Comprehensive... (More)

Dynamic functional connectivity (dFC) captures temporal dynamics in functional magnetic resonance imaging (fMRI) to better characterize brain activity, supporting multiple downstream analyses. However, effective modeling strategies for extracting representative signals from dFC remain underexplored. We proposed a novel Subject-wise and Temporal Contrastive Transformer (STCT) framework that advances dynamic connectivity modeling through a dual-constraint contrastive learning strategy. Our STCT framework uniquely integrates two objectives: 1) subject-wise contrast to preserve inter-individual specificity, and 2) temporal contrast to capture dynamic dependencies, which is a critical dimension overlooked in prior methods. Comprehensive evaluations demonstrated superior performance of STCT over both supervised and self-supervised approaches based only on subject-wise contrast in various predictive domains spanning demographics, cognition, and mental disorder diagnosis. Interpretability analysis further revealed lateralized connectivity patterns in Autism Spectrum Disorder (ASD) classification, suggesting the potential of STCT to highlight clinically relevant connectivity patterns.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
contrastive learning, Dynamic functional connectivity, resting-state fMRI
in
IEEE Transactions on Neural Systems and Rehabilitation Engineering
volume
34
pages
13 pages
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
external identifiers
  • pmid:42213551
  • scopus:105040988662
ISSN
1534-4320
DOI
10.1109/TNSRE.2026.3698236
language
English
LU publication?
yes
id
1803b611-80ff-4116-8a29-2c3c938fc5ae
date added to LUP
2026-09-15 14:44:26
date last changed
2026-09-16 03:38:20
@article{1803b611-80ff-4116-8a29-2c3c938fc5ae,
  abstract     = {{<p>Dynamic functional connectivity (dFC) captures temporal dynamics in functional magnetic resonance imaging (fMRI) to better characterize brain activity, supporting multiple downstream analyses. However, effective modeling strategies for extracting representative signals from dFC remain underexplored. We proposed a novel Subject-wise and Temporal Contrastive Transformer (STCT) framework that advances dynamic connectivity modeling through a dual-constraint contrastive learning strategy. Our STCT framework uniquely integrates two objectives: 1) subject-wise contrast to preserve inter-individual specificity, and 2) temporal contrast to capture dynamic dependencies, which is a critical dimension overlooked in prior methods. Comprehensive evaluations demonstrated superior performance of STCT over both supervised and self-supervised approaches based only on subject-wise contrast in various predictive domains spanning demographics, cognition, and mental disorder diagnosis. Interpretability analysis further revealed lateralized connectivity patterns in Autism Spectrum Disorder (ASD) classification, suggesting the potential of STCT to highlight clinically relevant connectivity patterns.</p>}},
  author       = {{Zhu, Jianfei and An, Lijun and Zhu, Haiqi and Jiang, Feng and Liu, Shaohui and Wei, Baichun and Yi, Chunzhi}},
  issn         = {{1534-4320}},
  keywords     = {{contrastive learning; Dynamic functional connectivity; resting-state fMRI}},
  language     = {{eng}},
  pages        = {{2870--2882}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  series       = {{IEEE Transactions on Neural Systems and Rehabilitation Engineering}},
  title        = {{Self-Supervised Representation Learning for Dynamic Functional Connectivity With Subjectwise and Temporal Contrasts}},
  url          = {{http://dx.doi.org/10.1109/TNSRE.2026.3698236}},
  doi          = {{10.1109/TNSRE.2026.3698236}},
  volume       = {{34}},
  year         = {{2026}},
}