Electrical impedance tomography and ventilator data integration for ventilation strategy optimization in mechanically ventilated.

Pierluigi Ravese is a Senior Clinical Engineer at St. James’s Hospital, specializing in ventilation and anesthesia technology for critical care. He holds a Bachelor’s Degree in Clinical Engineering from Sapienza University of Rome (2011) and a Master’s Degree in Bioengineering from Roma Tre University (2013). Since joining the Department of Medical Physics & Bioengineering at St. James’s Hospital in 2015, he has been actively involved in critical care support.

Since 2021, Pierluigi has focused on various aspects of optimizing ventilation strategies using EIT. He has presented his research on the processing of EIT images at the European Congress of Medical Physics (ECMP) 2022, and in 2024, he will present at both BEAI and ECMP on the integration of ventilator and EIT data. In 2016, he presented at BEAI and the European Medical and Biological Engineering Conference (EMBEC) 2017 on a critical care patient rehabilitation device.

Electrical impedance tomography and ventilator data integration for ventilation strategy optimization in mechanically ventilated patients

BACKGROUND: Integrating electrical impedance tomography (EIT) and ventilation data across different medical devices manufacturers presents significant challenges for functional lung monitoring aimed at optimising ventilation strategies in mechanically ventilated patients. Built-in real-time bedside analysis tools are often too inflexible to support local clinical protocols, handle motion and clinical artefacts, and custom data visualisation, diminishing their clinical utility

METHODS: An offline Python analysis framework was developed to handle data from a ventilator and EIT monitors from different manufacturers. EIT frame series and impedance waveforms (sampled at 20 Hz), along with ventilator data (sampled at 100 Hz) — comprising flow, pressure, and volume waveforms, as well as ventilator-derived measurements — were synchronised and merged using a tailored approach to align with the hospital's ventilation strategy optimization protocols.

RESULTS: Customizable breath grouping via EIT/ventilator waveform segmentation on the configurable dashboard offers improved data visualization and ventilation strategy optimization protocols analysis such as PEEP (Positive End-Expiratory Pressure) titration and breath analysis. Integrating user-defined measurements from raw data improves understanding of aeration patterns and relationships between ventilator waveforms, impedance waveforms, and EIT imaging and their impact on patient ventilation management. Furthermore, the framework enhances measurement reliability for clinical decision-making by allowing artefact filtering and statistical analysis on aggregated data.

CONCLUSIONS: The proposed method effectively overcomes the integration challenges between EIT monitors and ventilators from different manufacturers. The framework, provides a comprehensive data analysis platform for ventilation strategy optimization, enhancing EIT’s potential in a clinical setting by offering a high level of data analysis and visualisation customization.