Deep Learning Based High-Resolution Reconstruction of Trabecular Bone Microstructures from Low-Resolution CT Scans using GANCIRCLE

摘要

Osteoporosis is a common age-related disease characterized by reduced bone density and increased fracture-risk.Microstructural quality of trabecular bone (Tb), commonly found at axial skeletal sites and at the end of long bones, is animportant determinant of bone-strength and fracture-risk. High-resolution emerging CT scanners enable in vivomeasurement of Tb microstructures at peripheral sites. However, resolution-dependence of microstructural measures andwide resolution-discrepancies among various CT scanners together with rapid upgrades in technology warrant dataharmonization in CT-based cross-sectional and longitudinal bone studies. This paper presents a deep learning-basedmethod for high-resolution reconstruction of Tb microstructures from low-resolution CT scans using GAN-CIRCLE. Anetwork was developed and evaluated using post-registered ankle CT scans of nineteen volunteers on both low- and highresolutionCT scanners. 9,000 matching pairs of low- and high-resolution patches of size 64×64 were randomly harvestedfrom ten volunteers for training and validation. Another 5,000 matching pairs of patches from nine other volunteers wereused for evaluation. Quantitative comparison shows that predicted high-resolution scans have significantly improvedstructural similarity index (p < 0.01) with true high-resolution scans as compared to the same metric for low-resolutiondata. Different Tb microstructural measures such as thickness, spacing, and network area density are also computed fromlow- and predicted high-resolution images, and compared with the values derived from true high-resolution scans.Thickness and network area measures from predicted images showed higher agreement with true high-resolution CT (CCC= [0.95, 0.91]) derived values than the same measures from low-resolution images (CCC = [0.72, 0.88]).