Search papers, labs, and topics across Lattice.
This paper introduces a hierarchical segmentation framework designed to automate the 3D segmentation of muscles and adipose tissue from multi-source CT data, addressing challenges related to data heterogeneity and high memory requirements. By employing techniques such as Dynamic Spacing, Anisotropic Patching, and Topology-Aware Asymmetric Resampling, the framework achieves high accuracy with per-structure Dice coefficients between 0.924 and 0.982 while maintaining low memory usage. The system processes volumes in an average of 44.5 seconds on a standard CPU workstation, demonstrating its potential for clinical deployment in body composition analysis.
Achieving clinical-grade segmentation accuracy on standard CPU workstations in under 45 seconds could revolutionize body composition analysis in healthcare settings.
Background: Automated 3D segmentation of muscles and adipose tissue from CT is vital for body composition analysis, but multi-source data heterogeneity and high CPU memory demands hinder clinical deployment. Methods: We propose a coarse-to-fine hierarchical framework to segment ten tissue structures. Efficiency is optimized using Dynamic Spacing and Anisotropic Patching, a Group Inference mechanism for low-memory sliding-window processing, and Topology-Aware Asymmetric Resampling for fast post-processing. Results: The framework was trained on 1,558 CT volumes from seven public and two private datasets, and evaluated on an independent test cohort (N=105), per-structure Dice coefficients ranged from 0.924 to 0.982. Eight major structures met the +-10% relative error clinical acceptance limit. On a 12-core CPU workstation, the GPU-free pipeline averaged 44.5 seconds per volume with 4.73 GB peak memory. Conclusion: This framework balances accuracy and efficiency, enabling robust, large-scale body composition analysis on standard CPU workstations.
