Recording of Heart Signals From the Chest Wall

The study goal is to investigate the effect of dialysis/medicinal treatment on cardiac function and heart sounds by recording heart signals from the chest wall.

The mechanical functionality of the cardiovascular system is governed by a complex interplay between pressure gradients, determined by the contraction force of the myocardial cells, the dynamics of blood flow and the compliance of cardiac chambers and blood vessels. These mechanical processes produce vibrations and acoustic signals that can be recorded over the chest wall. Vibro-acoustic heart signals, including heart sounds (phonocardiogram), apical pulse (apexcardiogram) and arterial pulse (e.g. carotid pulse) carry valuable clinical information, but their use has been mostly limited to qualitative assessment by manual methods [1] (Figure 1).

The primary research hypothesis of this work is that clinical information regarding the mechanical functionality of the cardiovascular system can be automatically extracted from the vibro-acoustic heart signals by combining medical algorithms with digital signal processing techniques and computational learning algorithms.

The utilization of vibro-acoustic signals in clinical diagnosis and monitoring, by means of computerized devices, has been overlooked for many years due to the introduction of more sophisticated imaging techniques such as echocardiography, cardiac CT and cardiac MRI. However, these valuable techniques require complex and expensive equipment, as well as expert operators and interpreters. In particular, these imaging techniques can not be used continuously or outside of the hospital environment. Recent advancements in sensor technology, wireless communication and miniaturization of high-performance computing devices enable to re-approach the analysis of mechanical heart signals using a broad interdisciplinary view.

The research methodology for achieving the goal of the trial will be as follows:

1. Vibro-acoustic heart signals including phonocardiogram, apexcardiogram and carotid pulse will be recorded from subjects undergoing dialysis/medicinal Treatment.
2. The correlation between the progress of the dialysis/medicinal treatment process and the changes in the temporal and morphological characteristics of the vibro-acoustic signals will be investigated.
3. Signal processing algorithms will be used to automatically analyze the vibro-acoustic signals.

The recorded signals will be saved digitally to the hard-disk of the recording system, along with the measured reference parameters. Signal processing methods [2][3] will be used to segment the signals into distinct components and extract temporal and morphological features. Statistical linear regression will be used to identify significant correlations between features of the vibro-acoustic signals and the reference parameters. Computational learning algorithms will be used to explore non-linear relations and to evaluate the potential of estimating hemodynamic indexes from the vibro-acoustic signals.

This study is intended to evaluate novel methods for non-invasive estimation of cardiac indexes that reflect the mechanical functionality of the heart. Modern digital signal processing techniques and efficient computational learning algorithms can be combined to attain automatic real-time processing of vibro-acoustic signals for continuous monitoring of cardiac functionality and early detection of cardiac pathologies.

The study will be conducted on a maximum of 200 adult subjects (age above 18), from 2 groups:

Group 1: Dialysis Patients Group 2: Patients after Myocardial Infarction,Patients suffering from CHF, patients in ICCU and patients hospitalized in Cardiology Unit.

• 1
  Dialysis Group
• 2
  Cardiac Malfunction Group

Inclusion Criteria:

• Subject or subject's guardian is able to comprehend and give an Informed consent for participation in the study.
• Subject has gone through a full physical examination.

Exclusion Criteria:

• Subject is under 18.
• Subject has artificial heart valves.
• Subject suffers from obesity (BMI ≥40).
• Subject suffers from any kind of skin disease.
• Subject is clinically unstable (by physician assessment).
• If subject is a female: subject is pregnant.
• Subject objection to the study.
• Concurrent participation in other clinical study.
• Physician objection.