NCT06937138

Artificial Intelligence-Assisted vs Surgeon-Planned Trajectories in Freehand Femoral Neck Screw Fixation

Comparison of AI-Assisted vs Surgeon-Planned Trajectories in Freehand Femoral Neck Screw Fixation: A Multicenter Randomized Controlled Trial

• Status: Recruiting
• Phase: N/A
• Study type: Interventional
• Enrollment: 334 (estimated)
• Sponsor: Union Hospital, Tongji Medical College, Huazhong University of Science and Technology · Academic / Other
• Sex: All
• Age: 18 Years
• Healthy volunteers: Not accepted

Summary

The goal of this clinical trial is to compare two different methods of screw path planning-AI-assisted versus surgeon-directed-in freehand percutaneous femoral neck fracture fixation surgery. The study will include adult patients diagnosed with femoral neck fractures who are eligible for cannulated screw fixation under fluoroscopic guidance.The main questions it aims to answer are: Does AI-assisted screw path planning improve the radiographic accuracy of screw placement (screw deviation, tip position, and inter-screw parallelism)? Does AI-assisted planning reduce operative time, number of intraoperative fluoroscopy exposures, intraoperative blood loss (mL) and surgeon workload compared with surgeon-directed planning? Does AI-assisted planning reduce postoperative complications and improve functional outcomes compared to surgeon-directed planning? Researchers will compare the AI-assisted planning group to the surgeon-directed planning group to determine whether AI guidance contributes to enhanced surgical precision, reduced intraoperative burden, and improved recovery outcomes. Participants will: Undergo freehand percutaneous internal fixation of femoral neck fractures with either AI-assisted or surgeon-directed screw path planning, Receive standardized perioperative care and follow-up at defined intervals, Be evaluated through clinical assessments, imaging studies, and documentation of intraoperative and postoperative metrics over a 12-month follow-up period.

Detailed description

Femoral neck fractures, occurring between the femoral head and the base of the femoral neck, are among the most common hip injuries, particularly in the elderly population. While surgical fixation with closed reduction and cannulated screws is a widely accepted standard, challenges such as suboptimal screw placement, prolonged fluoroscopy exposure, and increased risk of complications like nonunion or avascular necrosis persist-largely influenced by surgeon experience and intraoperative variability. To address these limitations,this trial investigates the effectiveness and safety of artificial intelligence (AI)-assisted versus surgeon-directed screw path planning in freehand percutaneous internal fixation of femoral neck fractures.

Conditions

• Femoral Neck Fractures

Interventions

Type	Name	Description
PROCEDURE	AI-Assisted Screw Path Planning	The trajectory for screw placement during femoral neck fracture fixation will be guided by an AI algorithm based on intraoperative X-ray imaging. The system will automatically suggest the screw entry point and trajectory, which are displayed for the surgeon to follow during freehand guidewire insertion under fluoroscopy. The surgeon will proceed with the operation after confirming the feasibility of the AI-generated plan. In principle, surgeons are advised not to modify the AI-recommended trajectory unless necessary, to preserve the independent evaluative value of the AI-assisted plan. If significant disagreement arises between the surgeon's judgment and the AI-recommended trajectory, a third-party orthopedic specialist-blinded to group allocation-will conduct an independent postoperative assessment of the screw placement's appropriateness and accuracy.
PROCEDURE	Surgeon-Directed Screw Path Planning	The screw trajectory will be entirely determined manually by the operating surgeon, based on personal experience and interpretation of intraoperative fluoroscopy, without reliance on any AI recommendation module.

Timeline

• Start date: 2025-11-14
• Primary completion: 2027-04-15
• Completion: 2027-06-15
• First posted: 2025-04-22
• Last updated: 2025-11-17

Locations

4 sites across 1 country: China