A multimodal approach combining tool-pressure and EEG features for laparoscopic skill classification using machine learning

Aims: Laparoscopic skill assessment traditionally relies on subjective evaluation, which lacks objectivity and consistency. Developing automated multimodal approaches that integrate tool-pressure and neural data can improve the reliability and scalability of skill assessment. Therefore, our objectives were to: (1) integrate a pressure-sensing unit into existing box-trainer simulators and laparoscopic tools to investigate and validate tool pressure features as objective indicators of laparoscopic skill; (2) combine EEG-derived power and phase-locking value (PLV) features with tool pressure data to evaluate the classification performance of different machine learning models. Methods: Tool-pressure, EEG, and ECG data, along with task completion time, error counts, and NASA-TLX workload scores, were collected from 10 surgeons and 13 inexperienced students performing a peg-transfer laparoscopic task. The pressure sensors were integrated into the right and left laparoscopic graspers. EEG features were extracted from the four different frequency bands using both power spectral density (PSD) and phase-locking value (PLV) measures. Three machine learning models (Random Forest Classifier (RFC), Gaussian Process Classifier (GPC), and AdaBoost Classifier (ABC)) were used to classify participants into surgeon and non-experienced groups based on these multimodal features Results : The findings showed that right–left pressure asymmetry was a more reliable indicator of surgical expertise compared to other tool-pressure metrics. Using only the asymmetry feature, RFC achieved up to 78% classification accuracy. The highest performance was obtained when combining theta-band power features with pressure asymmetry during the task, where both RFC and ABC reached 86% accuracy (F1 = 0.83; AUC = 0.92 for RFC and 0.85 for ABC). Theta-band findings support its relevance for surgical skill assessment. Conclusion: Overall, this multimodal approach combining psychomotor and neurophysiological measures enhances the objectivity of surgical skill evaluation and may support real-time feedback systems for laparoscopic training.