Diagnostic Reliability of ChatGPT in Letournel-Judet Acetabular Fracture Classification

NCT07673991 | Completed

• 1 other identifier: TABED 2-26-1888
• Study Type: observational
• Target: 184
• Locations: 1 country
• Sites: 1

Brief Summary

This study aims to evaluate the diagnostic reliability of the multimodal artificial intelligence model ChatGPT-4o in classifying acetabular fractures using the Letournel-Judet classification system. The study retrospectively analyzed standardized pelvic radiographs (anteroposterior, iliac oblique, and obturator oblique) from 184 patients presenting with pelvic injuries. The diagnostic performance of ChatGPT-4o was compared against the independent assessments of two fourth-year orthopaedic residents and a reference standard established by an experienced trauma surgeon using multiplanar computed tomography (CT) and intraoperative findings. By utilizing a systematic radiographic checklist, the study assesses the AI (artificial intelligence) model's ability to identify key anatomical landmarks and integrate them into a final fracture pattern. This research aims to provide critical data on the current feasibility of using large language models as decision-support tools in complex orthopaedic trauma.

Conditions

Acetabular Fractures; Pelvic Injury

Keywords

Artificial Intelligence; ChatGPT; Letournel-Judet Classification; Acetabular Fracture

Outcome Measures

Primary Outcomes (1)

• Diagnostic Accuracy of ChatGPT-4o in Acetabular Fracture Classification.

  The diagnostic accuracy is defined as the proportion of correctly classified acetabular fractures by ChatGPT-4o compared to the reference standard (determined by 3D CT and intraoperative findings).

  Through study completion

Secondary Outcomes (2)

• Interobserver Agreement (Cohen's Kappa)

  Through study completion

• Systematic Radiographic Checklist Accuracy

  Through study completion

Study Arms (1)

Acetabular Fracture Cases

Retrospective cohort of 184 patients presenting with pelvic injuries who underwent surgical treatment and had standardized Judet radiographs available for analysis

Other: Diagnostic Assessment by ChatGPT-4o

Interventions

Diagnostic Assessment by ChatGPT-4o OTHER

No active intervention was performed; this study retrospectively analyzed radiographic images using a multimodal artificial intelligence model to assess its diagnostic accuracy compared to human clinicians.

Acetabular Fracture Cases

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population consists of 184 patients who presented to Ankara Bilkent City Hospital with pelvic injuries and underwent surgical treatment. The cohort includes individuals with various acetabular fracture patterns requiring the Letournel-Judet classification system for preoperative planning. Excluded populations were patients with non-acetabular pelvic fractures, incomplete radiographic series, poor image quality, or a history of previous pelvic surgery or trauma.

You may qualify if:

• Patients older than 18 years old and presenting with pelvic injuries and undergoing surgical treatment.

You may not qualify if:

• Patients with pelvic injuries but with non-acetabular fractures, those with incomplete imaging, those with poor radiographic quality or those who had previously undergone pelvic surgery or trauma were excluded.

Sponsors & Collaborators

• Ankara City Hospital Bilkent: lead

Study Sites (1)

Ankara Bilkent Şehir Hastanesi

Ankara, Çankaya, 06210, Turkey (Türkiye)

Location

Related Publications (15)

• Lindsey R, Daluiski A, Chopra S, Lachapelle A, Mozer M, Sicular S, Hanel D, Gardner M, Gupta A, Hotchkiss R, Potter H. Deep neural network improves fracture detection by clinicians. Proc Natl Acad Sci U S A. 2018 Nov 6;115(45):11591-11596. doi: 10.1073/pnas.1806905115. Epub 2018 Oct 22.

  PMID: 30348771 BACKGROUND

• Brandser E, Marsh JL. Acetabular fractures: easier classification with a systematic approach. AJR Am J Roentgenol. 1998 Nov;171(5):1217-28. doi: 10.2214/ajr.171.5.9798851. No abstract available.

  PMID: 9798851 BACKGROUND

• Walker AN, Smith JB, Simister SK, Patel O, Choudhary S, Seidu M, Dallas-Orr D, Tse S, Shahzad H, Wise P, Scott M, Saiz AM, Lum ZC. Assessing Inter-rater Reliability of ChatGPT-4 and Orthopaedic Clinicians in Radiographic Fracture Classification. J Orthop Trauma. 2025 Sep 19. doi: 10.1097/BOT.0000000000003079. Online ahead of print.

  PMID: 40970709 BACKGROUND

• Olczak J, Fahlberg N, Maki A, Razavian AS, Jilert A, Stark A, Skoldenberg O, Gordon M. Artificial intelligence for analyzing orthopedic trauma radiographs. Acta Orthop. 2017 Dec;88(6):581-586. doi: 10.1080/17453674.2017.1344459. Epub 2017 Jul 6.

  PMID: 28681679 BACKGROUND

• Cha Y, Kim JT, Park CH, Kim JW, Lee SY, Yoo JI. Artificial intelligence and machine learning on diagnosis and classification of hip fracture: systematic review. J Orthop Surg Res. 2022 Dec 1;17(1):520. doi: 10.1186/s13018-022-03408-7.

  PMID: 36456982 BACKGROUND

• Kuo RYL, Harrison C, Curran TA, Jones B, Freethy A, Cussons D, Stewart M, Collins GS, Furniss D. Artificial Intelligence in Fracture Detection: A Systematic Review and Meta-Analysis. Radiology. 2022 Jul;304(1):50-62. doi: 10.1148/radiol.211785. Epub 2022 Mar 29.

  PMID: 35348381 BACKGROUND

• Yucens M, Aydemir AN, Demirkan AF. ASSESSMENT OF INTEROBSERVER RELIABILITY FOR THE LETOURNEL AND JUDET CLASSIFICATION. Acta Ortop Bras. 2024 Mar 22;32(1):e267640. doi: 10.1590/1413-785220243201e267640. eCollection 2024.

  PMID: 38532863 BACKGROUND

• Mack LA, Harley JD, Winquist RA. CT of acetabular fractures: analysis of fracture patterns. AJR Am J Roentgenol. 1982 Mar;138(3):407-12. doi: 10.2214/ajr.138.3.407.

  PMID: 6977989 BACKGROUND

• Erickson BJ, Korfiatis P, Akkus Z, Kline TL. Machine Learning for Medical Imaging. Radiographics. 2017 Mar-Apr;37(2):505-515. doi: 10.1148/rg.2017160130. Epub 2017 Feb 17.

  PMID: 28212054 BACKGROUND

• O'Toole RV, Cox G, Shanmuganathan K, Castillo RC, Turen CH, Sciadini MF, Nascone JW. Evaluation of computed tomography for determining the diagnosis of acetabular fractures. J Orthop Trauma. 2010 May;24(5):284-90. doi: 10.1097/BOT.0b013e3181c83bc0.

  PMID: 20418733 BACKGROUND

• Petrisor BA, Bhandari M, Orr RD, Mandel S, Kwok DC, Schemitsch EH. Improving reliability in the classification of fractures of the acetabulum. Arch Orthop Trauma Surg. 2003 Jun;123(5):228-33. doi: 10.1007/s00402-003-0507-y. Epub 2003 Apr 26.

  PMID: 12720013 BACKGROUND

• Beaule PE, Dorey FJ, Matta JM. Letournel classification for acetabular fractures. Assessment of interobserver and intraobserver reliability. J Bone Joint Surg Am. 2003 Sep;85(9):1704-9.

  PMID: 12954828 BACKGROUND

• Letournel E. Acetabulum fractures: classification and management. Clin Orthop Relat Res. 1980 Sep;(151):81-106.

  PMID: 7418327 BACKGROUND

• JUDET R, JUDET J, LETOURNEL E. FRACTURES OF THE ACETABULUM: CLASSIFICATION AND SURGICAL APPROACHES FOR OPEN REDUCTION. PRELIMINARY REPORT. J Bone Joint Surg Am. 1964 Dec;46:1615-46. No abstract available.

  PMID: 14239854 BACKGROUND

• Ziran N, Soles GLS, Matta JM. Outcomes after surgical treatment of acetabular fractures: a review. Patient Saf Surg. 2019 Mar 16;13:16. doi: 10.1186/s13037-019-0196-2. eCollection 2019.

  PMID: 30923570 BACKGROUND

Study Design

• Study Type: observational
• Observational Model: CASE ONLY
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Principal Investigator, Orthopaedic Surgery Resident

Study Record Dates

• First Submitted: June 22, 2026
• First Posted: June 29, 2026
• Study Start: February 1, 2026
• Primary Completion: April 1, 2026
• Study Completion: May 12, 2026
• Last Updated: June 29, 2026

Record last verified: 2026-06

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP
• Time Frame: Data will be available starting from 6 months after the publication date and will remain available for 5 years.
• Access Criteria: Researchers interested in the data must submit a formal request to the corresponding author, including a research proposal and a data-sharing agreement. Access will be granted to those whose proposals are aligned with the original study objectives and have institutional ethical approval.

Locations

TU (1)