Improving the anesthetic process by a fuzzy rule based medical decision system

Area of Science:

• Anesthesiology
• Biomedical Engineering
• Artificial Intelligence

Background:

• General anesthesia requires precise drug titration for patient safety and comfort.
• Manual drug administration can be complex, demanding significant clinician attention.
• Automating drug delivery can optimize patient care and free up clinical resources.

Purpose of the Study:

• To design and implement a fuzzy logic tool for automated drug delivery during general anesthesia.
• To adjust drug dosage based on real-time patient needs using clinician expertise.
• To reduce the routine workload for anesthesiologists, allowing focus on critical patient variables.

Main Methods:

• A two-level computer decision system utilizing a fuzzy logic controller was developed.
• The controller used Bispectral Index (BIS) to monitor hypnotic state, with BIS error and rate as inputs.
• Propofol was administered via an infusion pump in a closed-loop system targeting a BIS of 50.

Main Results:

• An observational study compared the fuzzy logic closed-loop (FCL) system with manual drug administration in 81 ambulatory surgery patients.
• The FCL group (42 patients) demonstrated superior accuracy in maintaining the target BIS compared to the manual group (39 patients).
• The system successfully reached the target BIS of 50 in both groups.

Conclusions:

• The fuzzy logic closed-loop system, based on clinician experience, performed satisfactorily.
• The automated system outperformed manual propofol administration in accuracy during the maintenance stage of anesthesia.
• This technology offers a promising approach to enhance anesthesia management and patient safety.