Comparing AI (GPT-4) and surgeons in diagnosing and treating painful or failed hip replacements
Observational Study on the Accuracy and Completeness of General Artificial Intelligence in the Diagnosis and Therapeutic Recommendations for Failed or Painful Total Hip Arthroplasty
This project tests whether GPT-4 gives the same diagnoses and treatment suggestions as orthopedic surgeons for adults with painful or failed total hip replacements.
Quick facts
| Study type | Observational |
|---|---|
| Enrollment | 20 (estimated) |
| Ages | 18 Years to 80 Years |
| Sex | All |
| Sponsor | Istituto Ortopedico Rizzoli Academic / other |
| Locations | 1 site (Bologna) |
| Trial ID | NCT07012577 on ClinicalTrials.gov |
What this trial studies
This observational analysis uses 20 anonymized cases of failed or painful total hip arthroplasty drawn from a single institutional database. Each case includes full preoperative history, imaging, surgical reports and documented outcomes, and is reviewed by GPT-4 and by three orthopedic clinicians at different training levels. Reviewers' diagnostic labels and treatment recommendations are categorized for accuracy and completeness against the documented reference diagnosis and treatment. The design compares concordance and gaps between the AI and human reviewers to characterize where GPT-4 matches, misses, or adds to surgeon decision-making.
Who should consider this trial
Good fit: Ideal cases are adults aged 18–80 with a documented painful or failed total hip arthroplasty and complete preoperative clinical history, imaging, surgical reports, and outcome documentation.
Not a fit: Patients with well-functioning hip implants, incomplete records, poor-quality images, or complex multifactorial failures are unlikely to benefit from the comparisons made in this project.
Why it matters
Potential benefit: If successful, this could provide faster, more consistent second opinions that help guide correct diagnosis and treatment planning for patients with failed hip replacements.
How similar studies have performed: Some AI tools have shown promise in orthopedic imaging and decision support, but using GPT-4 for comprehensive diagnosis and treatment recommendations in failed hip arthroplasty is relatively novel and not yet proven.
Eligibility criteria
Show full inclusion / exclusion criteria
Inclusion Criteria: * Adults (≥18 and ≤80 years old). * Documented painful or failed total hip arthroplasty requiring clinical/radiological evaluation (2004-2024). * Complete pre-operative clinical history, imaging (X-ray/tomography), and surgical reports. * Clear diagnosis of failure mode (e.g., aseptic loosening, infection, fracture, wear). * Treatment and outcomes fully documented in the institutional database. * "Exemplary" cases with minimal diagnostic ambiguity (per Engh/MusculoSkleletal Infection Society criteria, etc.). Exclusion Criteria: * total hip arthroplasty with no documented failure/pain (well-functioning implants). * Incomplete clinical/radiological records (e.g., missing pre-operative imaging or surgical notes). * Complex/multifactorial failures (e.g., concurrent infection + loosening + fracture). * Radiographs/images non-interpretable (poor quality, missing views). * Cases with conflicting diagnoses/treatments in original records.
Where this trial is running
Bologna
- SC Ortopedia e Traumatologia e Chirurgia Protesica e dei Reimpianti di Anca e Ginocchio, IRCCS Istituto Ortopedico Rizzoli — Bologna, Italy (Recruiting)
Study contacts
- Principal investigator: Francesco Castagnini, MD — IRCCS Istituto Ortopedico Rizzoli
- Study coordinator: Francesco Castagnini, MD
- Email: francescocastagnini@hotmail.it
- Phone: +390516366418
How to participate
- Review the eligibility criteria above with your treating physician.
- Visit the official trial page on ClinicalTrials.gov for the most current contact information and recruitment status.
- Contact the listed study coordinator or principal investigator to request pre-screening. Pre-screening is free and never obligates you to enroll.