Patient Pose Assessment Using a CT-Based Framework for Synthetic Data Generation

2026-08-06

AI Summary

This study addresses the critical challenge that patient positioning during X-ray imaging directly impacts diagnostic quality, yet real-world depth–X-ray paired data with annotations are scarce and constrained by regulatory limitations. To overcome this, the work proposes the first framework that leverages CT scans to synthesize paired depth and X-ray images for pretraining posture assessment models. By integrating CT image processing, synthetic data generation, and deep learning, the approach circumvents the barriers associated with acquiring real annotated data. Pretraining on 3,077 synthesized ankle joint samples significantly enhances model performance in real-world pose estimation, improving accuracy by up to 11 percentage points compared to models trained without such synthetic pretraining.

Abstract

An adequate diagnostic quality of radiographs is essential for reliable diagnoses and treatment planning. The patient's pose during radiography is one of the most important factors determining the diagnostic quality. Since patient positioning is difficult and not standardized, an automated AI-based approach using depth images to automatically assess the patient's pose before the radiograph has been taken would be helpful. Due to regulatory hurdles, however, it is difficult in practice to acquire the required depth images and corresponding radiographs. In this paper, we present a framework that can generate such training data synthetically from Computed Tomography scans. We further show that by pretraining on our generated synthetic dataset consisting of 3077 image pairs of upper ankle joints, the pose assessment of real upper ankle joints can be improved by up to 11 percentage points.

Problem

• patient pose assessment
• radiographic quality
• synthetic data generation
• depth images
• CT scans

Innovation

• synthetic data generation
• CT-based framework
• pose assessment
• depth images
• AI pretraining