Acceptability, deviation, and efficiency of automated artificial intelligence-based implant planning methods: A...

Area of Science:

• Dental implantology
• Artificial intelligence in medicine
• Surgical planning

Background:

• Single-tooth replacement is a common dental procedure.
• Accurate implant planning is crucial for successful outcomes.
• Traditional human-based planning can be time-consuming.

Purpose of the Study:

• To compare the acceptability, deviation, and efficiency of two AI-driven implant planning methods against a human-based approach.
• To evaluate automated methods for single-tooth replacement planning.

Main Methods:

• Retrospective analysis of 32 patients' cone-beam computed tomography (CBCT) and intraoral scans.
• Three planning methods: human-based (HP), automated CBCT-image-based (AA), and automated CBCT-segmentation-based (RL).
• Assessment of implant position acceptability, positional deviations (linear and angular), and total planning time.

Main Results:

• The RL method showed similar surgical and prosthetic acceptability to HP, with significantly reduced planning time.
• The AA method was the most efficient but demonstrated significantly lower acceptability and greater positional deviation compared to RL.
• No significant differences in implant-mandibular canal distance were found among the three methods.

Conclusions:

• The automated CBCT-segmentation-based (RL) method offers a viable alternative to human-based planning, balancing accuracy and efficiency.
• The automated CBCT-image-based (AA) method, while fastest, requires further refinement to improve positional accuracy and clinical acceptability.
• AI-driven tools show promise in optimizing dental implant planning workflows.