FP7.1 A pilot study of AI-driven 3D reconstruction from a single radiograph for acetabular version assessment

Clicks: 9
ID: 331268
2026
Free PDF
Article Quality & Performance Metrics
Overall Quality
Not rated
Combines reader engagement with the AI quality analysis. This article has not been analysed, so there is no overall score — reader engagement is measured and shown alongside.
AI Quality Assessment
Not analyzed
Readership in this journal
Steady

Ranked #434 of 444 articles by views in Journal of hip preservation surgery

Most read Least read

Bar heights use a square-root scale. Only the 120 most-read articles are drawn; the journal has 444 in total.

Mint this article as an NFT
Not yet minted

Create a permanent, verifiable on-chain record of this article on the Scimatic Network. The NFT is held in your Journament account, and you can withdraw it to your own wallet at any time.

5 SUSD one-off · no wallet required
Sign in to mint
Abstract
Abstract ICLB07 - The Digital Surgeon: Practical Tools and Navigation Basics for Your Everyday Practice, Session Room 12 (BMCC Level 3), October 8, 2026, 14:00 - 16:00 Acetabular version (AV) is a clinically relevant parameter describing femoral head coverage in the axial plane, which cannot be directly observed on standard anteroposterior (AP) radiographs. Accurate estimation typically requires CT imaging. This study investigates whether AV can be reliably derived from a single AP hip radiograph using AI-based 3D reconstruction. A retrospective dataset of 94 asymptomatic subjects was used, including 75 training cases and 19 evaluation cases with corresponding CT ground truth. CT scans had an isotropic resolution of 1 mm. The cohort consisted of 54 males and 42 females with a mean age of 67 years. For each subject, a single AP radiograph of the hip joint centered on the femoral head was used as input. This was obtained for both the left and right hip. A deep implicit neural network framework was trained to reconstruct a patient-specific 3D volumetric representation from the 2D radiograph, producing both a CT-like volume and corresponding 3D bone models. Acetabular version was computed at on the reconstructed bone models and compared to measurements derived from the original CT-based bone models. AV was measured consistently at the level of the femoral head center in both modalities. The proposed method demonstrated strong agreement between predicted bone models and the CT measurements. The mean absolute error for acetabular version was 2.05° with a standard deviation of 1.33°. The 95% confidence interval ranged from 1.58° to 2.52°. Two outliers were observed with deviations up to 9°, likely related to imperfect reconstruction. This pilot study shows that AI-driven 3D reconstruction from a single AP radiograph can estimate acetabular version with clinically promising accuracy. The results suggest that 3D anatomical inference from limited 2D imaging may enable extraction of otherwise inaccessible biomechanical parameters. Future work will focus on expanding this to other parameters, improving robustness the address the outlier cases, and validating performance on larger and more pathological cohorts.
Reference Key
openalex_W7220429119 Use this key to autocite in the manuscript while using SciMatic Manuscript Manager or Thesis Manager
Authors Roel Huysentruyt, Ide Van den Borre, Emmanuel A. Audenaert
Are you one of these authors? Claim your profile
Journal Journal of hip preservation surgery
Year 2026
DOI
10.1093/jhps/hnag043.049
URL
https://doi.org/10.1093/jhps/hnag043.049
Keywords Keywords not found

Citations

No citations found. To add a citation, contact the admin at info@scimatic.org

Comments

No comments yet. Be the first to comment on this article.