Not_yet_recruitingNCT07536230

Validation of a Deep Learning Framework for Continuous Forecasting of Pharmacodynamic Responses and Physiological Trajectories During General Anesthesia

Sponsor: Universitair Ziekenhuis Brussel

Trial Details

Condition	BIS, BIS-EEG, Artifical Intelligence, Intraoperative, Machine Learning, Anesthesia, Anesthesia Awareness, Predictive Model
Status	Not_yet_recruiting
Participants	115
Est. Completion	1 Aug 2026
Last Updated	17 Apr 2026

Primary Outcome

Calibration error of the predictive uncertainty cone; Mean Absolute Error (MAE); Trend accuracy

Eligibility Criteria

Inclusion Criteria

Inclusion Criteria: * Patients scheduled for elective surgery requiring general anesthesia. * Procedures requiring continuous depth of anesthesia monitoring (BIS). Exclusion Criteria: \- Procedures where the primary anesthetic plan does not involve continuous electronic data capture.

Trial Sites

Countries

Belgium

View Registry Entry

Related Analysis

Phase IV Validation Study Opens: AI-Powered Anesthesia Monitoring Framework — Real-Time Prediction of Patient Responses During Surgery

Belgian researchers are validating an AI framework using LSTM networks to predict patient responses during anesthesia in real-time. The study aims to shift anesthesia monitoring from reactive to predictive, potentially preventing awareness events and optimizing drug dosing.

Read full article →

This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service