Cardiac eigen imaging: a novel method to isolate cardiac activity in thoracic electrical impedance tomography

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Cardiovascular Research

Background:

• Cardiovascular disease (CVD) burden necessitates advanced monitoring solutions.
• Current non-invasive cardiac hemodynamic monitoring lacks accuracy, precision, and continuity.
• Electrical impedance tomography (EIT) offers a potential low-cost, real-time imaging approach.

Purpose of the Study:

• To compare a standard approach, regular cardiac gated averaging (RCGA), with a novel method, cardiac eigen-imaging (CEI), for isolating cardiac activity in thoracic EIT.
• To assess the efficacy of CEI for non-invasive cardiac hemodynamic monitoring.
• To evaluate EIT's potential for continuous cardiovascular healthcare.

Main Methods:

• EIT and heart-rate (HR) data were collected from 20 heart-failure patients.
• Features from RCGA and CEI images were correlated with stroke volume (SV) obtained via echocardiography.
• Image reconstruction parameters were optimized using leave-one-out cross-validation.

Main Results:

• CEI achieved a higher Pearson correlation coefficient (0.80) with SV compared to RCGA (0.72).
• CEI identified 33 high-correlating pixels, significantly more than RCGA's 8.
• High-correlating pixels for CEI concentrated in the right ventricle (RV).

Conclusions:

• CEI demonstrates superior performance over RCGA for cardiac monitoring with EIT.
• CEI is independent of heart-rate data quality, unlike RCGA.
• CEI represents a significant advancement for EIT-based cardiac monitoring in diverse healthcare settings.