Self-Supervised Graph Transformer with Contrastive Learning for Brain Connectivity Analysis Towards Improving Autism Detection

Wed, 01 Jan 2025 00:00:00 +0000

Applied Computer Vision and Medical Image Analysis

Mon, 01 Jan 2024 00:00:00 +0000

Overview

This research thrust applies deep learning and computer vision techniques to impactful application domains, with a current emphasis on medical image analysis and privacy-preserving sensing.

Active Projects

Brain Connectivity Analysis for Autism Detection: Developing self-supervised graph transformers with contrastive learning to analyze brain connectivity patterns from fMRI data, with the goal of improving autism spectrum disorder detection (published at ISBI 2025).

Medical Image Segmentation: Collaborating on advanced segmentation architectures for biomedical imaging applications, including work targeting ECCV venues.

Privacy-Preserving Posture Monitoring (UWB-PostureGuard): A UWB-based RF sensing system for continuous ergonomic sitting posture monitoring that avoids the privacy concerns of camera-based approaches (published at HICSS 2026, NSF I-Corps participation).

Sediment Concentration Estimation: Applying machine learning models to estimate suspended sediment concentration from laboratory data, bridging computer vision/ML with environmental monitoring.

Medical Imaging | Sen He

Self-Supervised Graph Transformer with Contrastive Learning for Brain Connectivity Analysis Towards Improving Autism Detection

Applied Computer Vision and Medical Image Analysis

Overview

Active Projects