Self-learning fuzzy logic control of neuromuscular block

Area of Science:

• Anesthesiology
• Biomedical Engineering
• Control Systems

Background:

• Maintaining precise neuromuscular block depth is crucial for patient safety during surgery.
• Automated drug delivery systems aim to improve anesthetic control and reduce human error.
• Fuzzy logic controllers offer potential for adaptive drug administration in clinical settings.

Purpose of the Study:

• To evaluate the performance of a self-learning fuzzy logic controller for atracurium administration.
• To achieve a target neuromuscular block depth of 10% T1 twitch height.
• To compare the controller's efficacy against previous adaptive and fixed-rule fuzzy controllers.

Main Methods:

• A self-learning fuzzy logic controller was employed in 20 ASA I/II patients undergoing prolonged surgery (>90 min).
• Neuromuscular block depth was monitored using a Datex Relaxograph, targeting 10% T1 twitch height.
• The controller autonomously adjusted atracurium infusion rates via a Graseby 3400 pump, starting with an empty rule-base.

Main Results:

• The system achieved stable neuromuscular block control with a mean T1 error of -0.52% (SD 0.55%).
• Atracurium infusion rates ranged from 0.25-0.44 mg kg-1 h-1, demonstrating adaptability.
• Infusion rate variation was lower than with a previously studied fixed-rules fuzzy controller.

Conclusions:

• The self-learning fuzzy logic controller provides effective and stable control of neuromuscular block depth.
• This approach is a viable, less computationally demanding alternative to complex adaptive control strategies.
• The controller demonstrates improved infusion rate consistency, enhancing its clinical applicability.