Deep Learning Framework for Continuous Depth of Anesthesia Forecasting

Condition(s) studied

Study summary

The integration of Artificial Intelligence (AI) in anesthesiology offers the potential to shift patient monitoring from reactive to predictive. Deep learning architectures, specifically Long Short-Term Memory (LSTM) networks, excel at processing complex, time-series data to forecast future clinical states.

While standard PK/PD models (such as the state of the art Eleveld model for Propofol and Remifentanil) estimate target-site drug concentrations (Ce), they do not account for real-time, patient-specific dynamic responses. This study aims to deploy an AI framework designed to predict future physiological states.

Eligibility

Primary outcome measure(s)

• Calibration error of the predictive uncertainty cone — Continuous - Perioperative
  Calibration error of the predictive uncertainty cone - Calibration error of the predictive uncertainty cone is the discrepancy between a model's stated confidence level (e.g., predicting that 95% of future values will fall within a specific range) and the actual frequency with which the true values actually land inside that predicted boundary.
• Mean Absolute Error (MAE) — Continuous - perioperative
  Mean Absolute Error (MAE)
• Trend accuracy — Continuous - perioperative
  Trend accuracy measures a predictive model's ability to correctly forecast the future direction and rate of change of a variable (such as whether a patient's anesthesia depth is actively lightening or deepening), independent of the absolute numerical error at any single point in time.

Trial sites (1)

More Universitair Ziekenhuis Brussel trials in Belgium

Other trials for the same condition