IEEE journal of biomedical and health informaticsImagingThree-DimensionalHumansMicroscopyNeural Networks
Multi-Aperture Fusion of Transformer-Convolutional Networks With Curvature-Aware Loss Function Enhances 3D Segmentation of Clinical and Microscopy Images.
Siyavash Shabani, Sahar A Mohammed, Muhammad Sohaib, Bahram Parvin
Published: 202510.1109/JBHI.2025.3625474
Abstract
Automated 3D Radiological and Microscopy image analysis enables quantitative therapeutics assessment. However, accurate segmentation and smaller footprints are needed. To address these issues, the current study introduces the Multi-Aperture Transform…
Preview only. Read the full abstract at the source