BioCompNet: a deep learning workflow enabling automated body composition analysis toward precision management of cardiometabolic disorders
Peer-Reviewed Publication
Updates every hour. Last Updated: 23-Dec-2025 21:11 ET (24-Dec-2025 02:11 GMT/UTC)
A research paper by scientists at Shanghai Jiao Tong University School of Medicine presented BioCompNet—an end-to-end deep learning workflow that integrates dual-parametric magnetic resonance imaging (MRI) sequences (water/fat) with a hierarchical U-Net architecture to enable fully automated quantification of 15 biomechanically critical BC components..
The research paper, published on Aug. 20, 2025 in the journal Cyborg and Bionic Systems.Soil moisture is a key factor driving the Earth's water and carbon cycles, but large-scale monitoring has long been hindered by sparse ground observations and model uncertainties.
Across Asia's vast drylands, a new study reveals a critical imbalance between degradation and recovery.
Scientists from Harbin Institute of Technology, together with collaborators from The University of Hong Kong and Southern University of Science and Technology, has developed a decoupled titanium dioxide–viologen hybrid material that integrates energy storage and color-changing display functions within a single system. This breakthrough enables high-capacity, high-efficiency electrochromic energy storage and marks an important step toward the next generation of smart, interactive energy devices.