Multi-modal Graph Diffusion Model for Depression Detection

doi:10.1007/978-981-95-5634-2_20

2026-01-01

PRE

Huo, Yufu

Xie, Ruitao

Cai, Yunpeng *

DOI：10.1007/978-981-95-5634-2_20

原文链接

原文求助

摘要

En 中文

抑郁症是全球普遍存在的精神障碍，其评估因多个主观因素而具有挑战性，这凸显了基于深度学习的以数据驱动的方法实现更客观有效检测的必要性。本文提出了一种用于抑郁症检测的多模态图扩散模型（MGD4D），该模型将多模态学习与先进的图扩散建模相结合。在其新颖的框架中，脑神经影像数据及辅助信息被表示为结构、功能和信息子图。随后，通过一个包含模态特定编码器、潜在图扩散模型和图分类解码器的端到端流程推断抑郁症概率。该扩散模型的关键创新在于其去噪网络，该网络协同结合了基于Transformer的多头注意力机制和U-Net模块。在实验验证中，公开数据集上的对比实验表明，MGD4D在抑郁症检测中显著优于近期强基线方法，达到当前最佳（SOTA）性能，而消融研究进一步量化了每个核心组件的影响。总体而言，本研究为区分抑郁症患者与健康对照（HCs）提供了一种有前景、客观且智能的方法，具有适应其他精神障碍的巨大潜力。完整代码可在：https://github.com/RosalindFok/MGD4D.git

Keyword:

Depression Detection

Neuroimaging

Latent Graph Diffusion

Multi-modal Learning

期刊

PATTERN RECOGNITION AND COMPUTER VISION, PRCV 2025, PT XIII

IF:

论文数:

被引数:

机构

chinese academy of sciences

学者数:

56.7W

论文数: 44.9W

被引数: 704

引用论文

A Combined Attribute Extraction Method for Detecting Postpartum Depression Using Social Media

Lecture Notes in Computer Science

IF0

2023-01-01

PREAI

Gopalakrishnan,Abinaya; Gururajan,Raj; Venkataraman,Revathi; Zhou,Xujuan; Chan,Ka Ching

Alteration of whole-brain amplitude of low-frequency fluctuation and degree centrality in patients with mild to moderate depression: A resting-state functional magnetic resonance imaging study

FRONTIERS IN PSYCHIATRY

IF3.2

2022-12-07

OAAI

Chen, Fenyang; Wang, Luoyu; Ding, Zhongxiang

Altered effective connectivity in sensorimotor cortices is a signature of severity and clinical course in depression

Proceedings of the National Academy of Sciences

IF0

2021-09-30

OAAI

Dipanjan Ray; Dmitry Bezmaternykh; Mikhail Mel’nikov; Karl J. Friston; Moumita Das

Toward discovery science of human brain function 走向人类大脑功能的发现科学

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA

IF9.1

2010-03-09

2.4K

OAAI

Biswal, Bharat B.; Mennes, Maarten; Zuo, Xi-Nian; Gohel, Suril; Kelly, Clare; Smith, Steve M.; Beckmann, Christian F.; Adelstein, Jonathan S.; Buckner, Randy L.; Colcombe, Stan; Dogonowski, Anne-Marie; Ernst, Monique; Fair, Damien; Hampson, Michelle; Hoptman, Matthew J.; Hyde, James S.; Kiviniemi, Vesa J.; Kotter, Rolf; Li, Shi-Jiang; Lin, Ching-Po; Lowe, Mark J.; Mackay, Clare; Madden, David J.; Madsen, Kristoffer H.; Margulies, Daniel S.; Mayberg, Helen S.; McMahon, Katie; Monk, Christopher S.; Mostofsky, Stewart H.; Nagel, Bonnie J.; Pekar, James J.; Peltier, Scott J.; Petersen, Steven E.; Riedl, Valentin; Rombouts, Serge A. R. B.; Rypma, Bart; Schlaggar, Bradley L.; Schmidt, Sein; Seidler, Rachael D.; Siegle, Greg J.; Sorg, Christian; Teng, Gao-Jun; Veijola, Juha; Villringer, Arno; Walter, Martin; Wang, Lihong; Weng, Xu-Chu; Whitfield-Gabrieli, Susan; Williamson, Peter; Windischberger, Christian; Zang, Yu-Feng; Zhang, Hong-Ying; Castellanos, F. Xavier; Milham, Michael P.

High-Resolution Image Synthesis with Latent Diffusion Models 具有潜在扩散模型的高分辨率图像合成

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

IF0

2022-06-01