An EIT-based Research On Lung Protective Ventilation

Mechanical ventilation is an important method of auxiliary respiration in ICU,however it may cause serious lung injury when the ventilator parameters are not properly set.As a a non-invasive fast imaging technique,EIT has been used to guide the setting of ventilator parameters in some studies of lung protective ventilation.However,when the ventilator itself is not albe to meet the best ventilation mode,the EIT-based guidance will be greatly restricted.Based on the reasons mentioned above,this paper designed a high performance EIT system CarPulMon1 after a comprehensive and in-depth investigation of the research status of EIT,among the the electrode system,channel switching time,constant current source correction,bad electrode detection have been improved.And then,lung ventilation characteristics and heart activity / perfusion impedance characteristics were studied of five five adult male,which lay a technical foundation for further study of protective ventilation.CarPulMon1 uses a 16 electrode system,adjacent excitation mode,GREIT image reconstruction algorithm,has a frame rate of 6~78fps and a 128×128 reconstructed image.The amplitude of the constant current source is 0.96 mApp and the frequency range is 50kHZ~250kHZ.The corrected system still has an 80 dB SNR at the 1MHZ excitation frequency.At all the test frequency of 50kHZ~250kHZ,the relative standard deviation of the image is less than 0.03%.Among them,a dual-band electrode belt was designed.In more than four hours of testing,the dual-band electrode belt was shown to be able to always keep in touch well.Single factor analysis method was used to determine the main factors affecting the recovery time of channel switching,the results show that the recovery time is mainly affected by the HPF of voltmeter,while the electrode belt and the imaging subjects have little effect on recovery time.In final system,signal recover time is about 23 us,3dB attenuation band of HPF is 100 HZ.This design is able to ensure a high SNR(NumAvg=8)and a fast scanning speed(20fps)when the excitation frequency is not less than 50 kHZ.In-vivo imaging experiments were completed with five healthy adult males as imaging subjects.Among them,PCA method was used to separate the lung and cardiac activities/perfusion.The relationship of ventilation characteristics and electrode layer,body position and spontaneous breathing frequency were in-depth studied,and a preliminary study of the frequency responseof cardiac activities/perfusion impedance has also been completed.The results show that: i).The Layer2 electrode layer(about 6cm above Xiphoid Bone,or between the the second rib and third above Xiphoid Bone)has the worst lung ventilation uniformity,and is able to maximally reflect the degree of degree of lung injury.ii).Supine position ventilation has better uniformity than sitting and prone position.iii).Pulmonary ventilation curve is able to be roughly divided into five stages,including three stages of inspiration(slow-fast-slow)and two stages of respiration(fast-slow).iii).There was no significant increase in the volume of lung ventilation when the respiratory period is more than 4S.iv).PCA-based method is able to extracte weak cardiac activities/perfusion impedance signal from measured signal.v).In the range of 50kHZ~250kHZ,cardiac activities/perfusion has the biggest impedance response amplitude when the excitation frequency is 50 kHZ.