Temporal Changes of Fiducial Seismocardiogram Points Due to Different Sensor Placements on the Chest Rabie Fadil 1* , Parshuram Aarotale 1* , Bradley Hoffmann 1* , Farzad Khosrow-Khavar 2 , Zhen Gang Xiao 2 , Alireza Akhbardeh 3 , Kouhyar Tavakolian 1 1 University of North Dakota School of Electrical Engineering & Computer Science 2 Heart Force Medical Inc, Vancouver, Canada 3 Department of Radiology, Johns Hopkins University, Baltimore, MD, USA * Authors contributed equally to this work Abstract This research focused on morphology changes due to sensor placement of the seismocardiogram (SCG). Data was recorded from seven male subjects, in supine position, accelerometers were placed in both vertical (V) and horizontal (HL) positions leading from the xiphoid process. The subjects were asked to perform a Valsalva maneuver and signals were analyzed before and after. Features of aortic valve closing (AC), aortic valve opening (AO) and the mitral valve closing (MC), were annotated. The features were performed for five cardiac cycles for every participant. The average of absolute percentage changes over the seven subjects was calculated for both V and HL cases. The correlation coefficients between all signals were also calculated to quantify the linear relationships. Moving ¼ of the length of the sternum changes MC point an average of 3.2%, AO annotation 2.6% but AC did not change. At the suprasternal notch annotations change significantly. HL array placement near apex showed negative correlation and inversion of peaks. It was found that the signal morphology does depend on the placement upon the sternum, however, placement differences might not induce significant changes in the timing fiducial points of SCG. 1. Introduction Cardiovascular diseases (CVD) are the leading causes of death worldwide [1]. In the United States, one person dies every 37 seconds from CVD [2], resulting in an estimated cost of $351.2 billion in 2014-2015 [3]. Early detection and diagnosis of CVD depends a variety of clinical modalities, including electrocardiogram (ECG), echocardiography, cardiac catheterization, computerized tomography, and magnetic resonance imaging scans [4]. ECG plays an important role in the initial diagnosis and monitoring of CVD. ECG measures the cardiac electrical activity that can diagnose diseases such as myocardial ischemia or arrhythmias. However, it does not provide information about the mechanical activity of the heart, which can aid in diagnosis of cardiac contractility dysfunction. Echocardiography is one of the most accepted methods for diagnosing heart diseases by providing a picture of the heart in real time with high fidelity [5, 6]. However, the echocardiography exam has high operational cost and require specialists. SCG has emerged as a promising technique for the assessment of cardio-mechanical function. SCG measures local vibrations due to the heartbeat against the chest [7, 8]. Such information might be used in complement with other methods (such as ECG, and echocardiography) for clinical diagnoses of heart diseases. Past investigations have looked at the efficiency and applicability of SCG in monitoring cardiac events. It was shown that it is possible to detect cardiac events observed in echocardiography using SCG [9, 10]. During SCG recording, these events include mitral valve closure (MC), aortic valve opening (AO), isovolumic moment (IM) during the systole, and aortic valve closure (AC) during diastole [11] (Figure 1). Detection of these features depends on SCG morphology, which is influenced by many factors such as: body posture, sensors placement, and inter-subject variability [12]. Thus, the goal of this work is to investigate changes in the morphology of SCG when the sensors are placed along the sternum or to the left of the chest. Figure 1: ECG and SCG waveforms showing cardiac events. ECG SCG P Q R S T IM AC AO MC Computing in Cardiology 2020; Vol 47 Page 1 ISSN: 2325-887X DOI: 10.22489/CinC.2020.394