Forward Modelling of the Human Thorax for Electrical Impedance Tomography Measurements

Electrical Impedance Tomography (EIT) is a medical imaging technique that reflects the inner conductivity distribution of biological tissue based on boundary voltages measured at the electrodes around the thorax while injecting an alternating current with a low amplitude. In this work, a 16-electrode thorax mesh is implemented for the pyEIT library. This mesh builds the basis for the reconstruction of the conductivity distributions from real human thorax voltage measurements. Post-processing techniques, such as the K-means method and the region based segmentation, have also been implemented. The resulting re-constructed images reflect the conductivity changes during the breathing cycle and by post-processing artifacts are eliminated leading to a clear image of the lungs and the heart.

I. Introduction

Medical imaging allows doctors to find diseases in their early stages and without surgery, which is very favorable for patients, by catching diseases early. Thanks to the rapid evolution of instrumental science, many imaging technologies have appeared and have been used in clinical applications, namely: Computed Tomography (CT), Ultrasonic Tomography, Magnitude Resonance Imaging (MRI), Nuclear medicine imaging including Positron-Emission Tomography (PET). However, all these techniques still present undeniable drawbacks, such as the harmful body ionizing radiation exposition in CT (X-rays radiations) and PET (Gamma rays radiations). Mean-while, Electrical Impedance Tomography (EIT) is becoming an auxiliary tool for medical diagnosis. It is commonly used for human thorax imaging. It has many advantages over other medical imaging techniques such that it is non-invasive, non-ionizing, easy to implement, provides real-time imaging and low cost. EIT thorax imaging setup is as follows: firstly, a certain number of electrodes, generally 8, 16 or 32, are placed around the thorax using a belt. An alternating current signal with a certain amplitude and frequency is generated by the signal source and injected into the body. The boundary voltages that appear on the surface of the electrodes are then measured. The inner thorax electrical impedance distribution is determined from the measured voltage value, using an EIT software and image reconstruction algorithms.