A Research Prototype for Closed-Loop Generative Design of Customized Foot Orthoses via Semantic-Physics Alignment

2026-07-18

AI Summary

This study addresses the semantic–physical misalignment between clinical prescriptions and three-dimensional foot orthosis design, which lacks an automated and verifiable generation pipeline. To bridge this gap, the authors propose TANS-FO, a closed-loop generative design framework that enables, for the first time, an end-to-end differentiable mapping from clinical text instructions to physical orthotic structures. The approach employs a Text-Aligned Neural Surrogate (TANS) to translate descriptive inputs into lattice density fields and replaces conventional finite element analysis with a graph neural network (GNN) for real-time plantar pressure prediction. Leveraging the PicoFoot-5K database, parametric modeling and lattice generation facilitate rapid customization. Experiments demonstrate excellent agreement between GNN predictions and Abaqus simulations (R² = 0.94), reducing design cycles to minutes. Compared to traditional parametric CAD, the method achieves a 34.7% reduction in peak pressure, a fit error of only 0.42 mm, and significantly improved short-term comfort.

Abstract

Translating unstructured clinical prescriptions into patient-specific foot orthoses (FOs) is hindered by a semantic-physical misalignment: high-level clinical intent is not mapped deterministically onto the 3D geometric parameters of the orthosis, and existing design workflows remain dependent on manual expertise with no instantaneous biomechanical validation. We present TANS-FO, a research prototype-a modular pipeline with closed-loop feedback for computational design automation of customized FOs, not a clinically validated therapeutic device. A Text-Aligned Neural Surrogate (TANS) uses cross-attention to project clinical-text embeddings onto a continuous lattice-density field, while a Graph Neural Network (GNN) surrogate predicts plantar stress in real time as a substitute for Finite Element Analysis (FEA). The framework is anchored on the open-access PicoFoot-5K anthropometric database (5,230 subjects; 30+ anatomical parameters). Under standardized quasi-static loading, the GNN surrogate agrees with an Abaqus reference solver (R^2 = 0.94), and the full pipeline synthesizes manufacturing-ready lattice insoles within minutes. On the Male 18-40 cohort, the proposed system attains a surrogate-predicted peak-pressure reduction of 34.7% over parametric CAD, with a fit error of 0.42 mm. Separately, an exploratory feasibility observation (n = 12; 2-week follow-up; no control group) using VAS pain reporting indicates short-term comfort improvement (VAS 6.4 -> 2.1), but this data is explicitly classified as preliminary observational evidence only-not evidence of clinical efficacy.

Problem

• foot orthoses
• semantic-physical misalignment
• clinical prescription
• generative design
• biomechanical validation

Innovation

• semantic-physics alignment
• closed-loop generative design
• neural surrogate
• graph neural network
• customized foot orthoses