Prospective Comparison Between Manual and Computer-Assisted (FOX) Cochlear Implant Fitting in Newly Implanted...

Area of Science:

• Audiology and hearing science
• Biomedical engineering
• Artificial intelligence in healthcare

Background:

• Traditional manual fitting of cochlear implants (CIs) is time-consuming and may not always yield optimal auditory outcomes.
• The development of AI-driven tools aims to enhance the efficiency and effectiveness of CI programming.
• The Fitting to Outcome eXpert (FOX) system represents an AI-based approach to CI fitting.

Purpose of the Study:

• To compare the efficacy of manual CI fitting versus AI-based fitting (FOX system) in newly implanted patients.
• To evaluate auditory outcomes, patient-reported comfort, and resource utilization between the two fitting methods.
• To assess long-term patient preference and performance with AI-optimized versus manually programmed CI maps.

Main Methods:

• A 1-year prospective, longitudinal, randomized controlled trial with a crossover design was conducted.
• Twenty-four first-time CI patients were randomized to either manual or FOX fitting, then crossed over after 1 year.
• Auditory outcomes (speech intelligibility, pure-tone thresholds, etc.), patient questionnaires, and CI parameters were compared.

Main Results:

• The AI-FOX group showed significantly better speech intelligibility in quiet and noise at 3 and 12 months postactivation (p < 0.05).
• In the manual group, 100% chose to keep the FOX map, with 82% showing improved performance.
• In the FOX group, 63% preferred the manual map for comfort, despite better objective outcomes with the FOX map.

Conclusions:

• The AI-FOX approach is equivalent or superior to manual fitting in hearing performance, comfort, and resource efficiency.
• The AI-FOX system demonstrates continuous learning capabilities, suggesting significant future advancements in CI fitting.
• AI-driven fitting holds promise for improving CI patient outcomes and optimizing clinical workflows.