Ballistocardiography (BCG) records chest vibrations when the heart beats and ejects blood to peripheral blood vessels. Electrocardiogram (ECG) records the electrical activity change during each cardiac cycle by the ECG machine. Research has shown that main heart malfunctions, such as congestive heart failure and valvular disease, would alter the BCG signal, thereby yielding an excellent potential for passive, noncontact monitoring of the cardiovascular status. (Guidoboni et al., 2020) 

Turtle shell technologies Pvt Ltd (Bengaluru) have developed a device that leverages Ballistocardiography to determine individual heartbeats, breaths and body movements with high accuracy from under the patient’s mattress by capturing the micro-vibrations generated by the patient’s body. 

Nishanth et al. (2021) demonstrated the utility of Dozee’s non-contact novel BCG device in the calculation of systolic time intervals with potential for automated detection. BCG derived manual and automatic Myocardial Performance Index (MPI) correlated with Transthoracic Echocardiogram (TTE ) derived MPI in a variety of Ejection Fraction (EF)   subgroups. The study indicated that manual BCG is an effective method to measure Left Ventricular Ejection Time (LVET), Mitral Valve Closure to Opening Time (MCOT) and MPI. The evaluation of EF remains a challenging task in clinical cardiology and echocardiography. The concept of deriving a measurement like EF from another calculated measure should adhere to good clinical practice and ecologically sound research methods.

 AIM

This study proposes an exploratory model for estimating EF through a novel approach BCG signal. The study will derive a signal processing framework and an adaptive predictive model algorithm for EF calculation vis-a-vis the BCG signal. This will be a quantitative and automated measurement characterising BCG derived MPI to calculate and classify EF to screen for left ventricular heart function.