Localized Muscular Fatigue in Robotic-Assisted Laparoscopic Surgery: Predictive Modeling Study.

Area of Science:

• Surgical Technology
• Biomedical Engineering
• Ergonomics

Background:

• Minimally invasive surgery (MIS) presents ergonomic challenges for surgeons despite its widespread adoption.
• Robotic-assisted surgery (RAS) is increasingly standard, but surgeon fatigue remains a concern.
• Wearable technology and AI offer potential solutions for analyzing and enhancing surgical ergonomics.

Purpose of the Study:

• To develop and validate a predictive model for localized muscle fatigue.
• To analyze muscle fatigue during conventional laparoscopic surgery (LAP) and RAS.
• To utilize electromyography (EMG) data for fatigue prediction.

Main Methods:

• Four surgical tasks (dissection, labyrinth, peg transfer, suturing) were performed in LAP and RAS settings.
• Wireless EMG sensors recorded muscle activity, with data analyzed for localized muscle fatigue.
• Multiple linear regression (MLR) and multilayer perceptron (MLP) models were trained and validated using surgeon expertise and surgical type data.

Main Results:

• RAS demonstrated reduced muscle fatigue in novice surgeons compared to LAP, but increased fatigue in expert surgeons.
• The predictive model accurately estimated localized muscle fatigue values with satisfactory R² and RMSE.
• The MLR model outperformed the MLP model in predictive accuracy.

Conclusions:

• A novel predictive model for localized muscle fatigue in MIS was successfully developed and validated.
• The model leverages wearable technology and artificial intelligence (MLR and MLP).
• Findings suggest potential for improved surgeon well-being and performance in MIS.