Robust Repeatable Respiratory Monitoring with EIT

Lead Research Organisation: University of Manchester
Department Name: Mathematics


Mechanical ventilation in the intensive care unit (ICU) is often a life-saving intervention for patients suffering from acute respiratory distress syndrome (ARDS) and acute lung injury (ALI). However, in recent years, it has become clear that mechanical ventilation can exacerbate lung damage and may even be the primary factor in lung injury e.g. ventilator-induced lung injury (VILI) and ventilator associated pneumonia (VAP). Consequently, lung protective ventilation (LPV) strategies have been developed, including positive end expiratory pressure (PEEP) and high frequency oscillatory ventilation (HFOV). Despite the use of LPV, the mortality associated with ARDS and ALI is still great, with reported rates between 33 and 55%.

Thoracic electrical impedance tomography (EIT) has emerged as a promising new imaging tool for bedside use. It is a non-invasive technique which provides the internal conductivity distribution based on electrical measurements on the skin. Existing research and commercial EIT systems have typically arranged measurement electrodes in rings around the thorax and produced two-dimensional image slices based on over-simplified two-dimensional generic shapes e.g. circular or elliptical cross-sections. Unfortunately, such an approach creates significant image artefacts and poor levels of repeatability. In order to eliminate this problem, EIT measurements and image reconstruction must be treated as a fully three-dimensional problem, taking into account the electrode positions and the body shape of the patient. One of the aims of this project is to provide reliable and robust three-dimensional EIT image reconstruction based on 3D informed body shape acquired from other modalities such as magnetic resonance imaging (MRI) and x-ray computed tomography.

Currently, there exists no consensus amongst clinicians on how to optimize PEEP ventilator settings. Improved knowledge about the distribution of ventilation will enable clinicians to set more appropriate ventilation parameters in order to reduce the potential occurrence of VILI. In particular, the value of bedside EIT monitoring for determining the effect of PEEP in comparison with traditional indirect methods, such as arterial oxygenation or global tidal volumes, is still largely unknown. Additionally, despite the clear link between lung mechanics and regional ventilation, ventilator settings are normally based on blood gases and pressure-volume curves rather than measurements of lung mechanics. A significant aim of this project is to provide clinicians with robust image-based lung mechanics models which will further aid the validation of EIT in the ICU.
Previous clinical EIT studies have generally utilized EIT instruments operating at relatively slow frame rates, inflexible drive and measurement strategies along with limited number of electrodes. In this project, we will develop a state-of-the-art fast 3D EIT system using advanced measurement techniques which have not yet had a significant impact in the clinical environment. In fact, there has never been a substantial investigation of EIT in the ICU which has drawn together the unique combination of high quality EIT measurements, three-dimensional image reconstruction and the validation of lung mechanics modelling.

We propose to establish a world-leading capability in the measurement and imaging by EIT in the ICU. This project builds upon the substantial expertise in EIT at the University of Manchester (UoM) and the UK's leading clinical group in applied thoracic EIT at Kings College London and Guy's and St Thomas' NHS Trust. We will also build upon the world-leading capability at UoM in modelling and dynamic MRI acquisition including oxygen-enhanced and dynamic contrast-enhanced MRI. The proposed project registers strongly on the EPSRC research theme of healthcare technologies.

Planned Impact

1) The main expected impact of this project is a reduction in ventilator related lung injury in mechanically ventilated patients. This will be achieved through improved real-time, spatial monitoring of lung function enabling the ventilator controls to be changed to maximize recruitment of alvioli and mimimize damage due to over distension.
2) Improved EIT monitoring of lungs is also expected to improve understanding of the mechanism of lung injury in mechanically ventilated patients leading to improved protocols for lung protective ventilation
3) Our experience with industrial collaboration and technology transfer means that this work can move rapidly from a laboratory experimental system to a production system that can be purchased by other medical EIT research groups. This will provide a mechanism for wider testing and development of the technique, and together with extensive conference and journal publication, to clinical acceptance of the technique.
4) Focusing on one specific application of EIT with a strong clinical pull serves as a driver for accelerated development of both hardware and reconstruction algorithms. We expect the instrumentation developments to have an impact more widely in medical (and possibly industrial and geophysical) applications of EIT.
5) The developments in forward and inverse solution algorithms and software is transferable to other medical and non- medical applications of EIT, as well as other inverse problems such as diffuse optical tomography. Our distribution of software as open source, our activities providing training to users via masterclasses, and our links with to a wide range of application areas in inverse problems mean that the impact of these developments will reach a wide range of disciplines including other medical inverse problems but as diverse as process monitoring, geophysics (eg pollution monitoring, carbon sequestration monitoring, salt water ingress), archaeology, land mine detection, and plant root monitoring.


10 25 50
Description A new system has been developed for monitoring lungs using electrical impedance and it is undergoing testing. Progress has been made on reconstruction algorithms including automatic segmentation and meshing based on lung images. The feasibility of using EIT data for automatic control of ventilation has been demonstrated in simulations.
Exploitation Route We are moving forward to testing the systems on volunteers and patients
Sectors Healthcare

Description The MIRAN segmentation software we developed with NE Scientific has been released and is being used. Software from our GREIT project was incorporated in a commercial EIT respiratory monitoring system which has been deployed in intensive care units in Germany
First Year Of Impact 2019
Sector Healthcare
Title R3M EIT system for respiratory monitoring 
Description This is an EIT system developed for clinical use and optimized for respiratory monitoring. It is currently in the process of MHRA approval so that it can be tested in Guy's and St Thomas's Hospital ICU 
Type Diagnostic Tool - Imaging
Current Stage Of Development Refinement. Clinical
Year Development Stage Completed 2019
Development Status Under active development/distribution
Impact Improved EIT algorithms including theory behind respiratory control using an EIT system. We also developed a software r system for inputting the body shape and segmenting organs from a DIcom image, which is then used to generate a FE mesh for EIT with correct body shape and electrode positions 
Description The EIDORS library provides a set of open source tools for EIT reconstruction. EIDORS Version 3.10 release on31 December 2019 incorporated work on the EPSRC R3M grant 
Type Of Technology Software 
Year Produced 2018 
Open Source License? Yes  
Impact The GRET component is now used in a commercial EIT system, of which at least 100 units have been sold to intensive care units in hospitals in Germany