Patient-anatomy-referenced multiscale modelling of median nerve conduction in carpal tunnel syndrome

Carpal Tunnel Syndrome (CTS) is a common compressive peripheral neuropathy causing focal conduction slowing and sensory dysfunction. Existing computational models are either mechanical or electrophysiological in isolation, precluding mechanistic understanding of nerve dysfunction across severity grades. We introduce a multiscale computational system built on a patient-anatomy-referenced median nerve pathway, rigidly registered and anatomically deformed against this patient’s own segmented bone and skin geometry, with the compression zone defined at 0–29.4 mm from the detected wrist point. A reduced-order conduction-velocity model, a FitzHugh–Nagumo (FHN) reaction–diffusion model and an unfitted McIntyre–Richardson–Grill (MRG) double-cable model were analysed on this pathway. The velocity and FHN models were calibrated to Kimura’s Normal and Mild data (zone velocities 57–59 and about 41 m/s) and predicted about 24 and 7 m/s for Moderate and Severe CTS. The MRG model predicted a normal velocity of 56.1 m/s without fitting and complete conduction block in Severe CTS, but under-predicted Mild and Moderate slowing (52.8 and 43.6 m/s). Convergence, sensitivity analysis and ten-run registration repeatability are reported, with registration limitations (16.0% of tube-surface points outside the skin) disclosed rather than assumed away. The framework is a research platform for peripheral neuropathy investigation, not a validated diagnostic tool.