A Customizable Tool for Anesthesia Signal Selection and Standardization

Automation of general anesthesia presents a complex challenge that requires high quality datasets and advanced data-driven techniques. This thesis focuses on the preprocessing and analysis of VitalDB, an open-source, high-resolution anesthesia database. The primary objective is to develop a user-friendly tool that allows clinicians and researchers to extract customized sub-datasets based on specific criteria, such as drug administration, selected physiological variables, and patient demographics, while automatically addressing missing data, sampling inconsistencies, and formatting issues. The tool integrates statistical reporting, graphical visualizations, and preprocessing functionalities to streamline data preparation for machine learning applications in anesthesiology. By aligning signals, normalizing measurement units, and unifying sampling rates, the system produces clean, analysis ready datasets. This project lays the foundation for more accurate predictive modeling and supports the development of closed-loop anesthesia control systems, ultimately enhancing patient safety and surgical outcomes.