Our Technology

AUDICOR® simultaneously evaluates the electrical and acoustical activities of the heart to detect, accurately interpret, and clearly document the presence of the third heart sound (S3) by using the company’s patented Acoustic Cardiography technology. Unlike other cardiac diagnostic technologies that focus on either the electrical or the mechanical capabilities of the heart, AUDICOR® analyzes the entire function of the heart concurrently and provides clinicians a complete analysis of the heart’s hemodynamic capabilities. AUDICOR® is currently demonstrating clinical utility by facilitating the assessment of left ventricular dysfunction, filling patterns and pressures, low ejection fractions, and monitoring changes in cardiac dysynchrony as a result of therapy.

Acoustic Cardiography

Acoustic cardiography simultaneously integrates heart sounds and single-channel electrocardiography input to generate multiple parameters that correlate to established hemodynamic measures.




Click here to view a Case Study illustrating acoustic cardiography applied to clinical practice.

Detecting the Sound Data

Acoustic cardiography technology uses two dual-function sensors capable of simultaneously capturing the heart’s electrical and sound characteristics and two electrocardiographic electrodes.  The technology requires only a single channel of electrocardiography for timing purposes and several acoustic cardiography measurements.

The AUDICOR® heart sounds algorithm receives three synchronous inputs: two single-channel sound signals collected by proprietary Audicor® sensors in the standard V3 and V4 positions and a single-channel of electrocardiography for timing purposes from two standard electrodes. The stream of data analyzed is ten seconds long. After analysis of both sound channels is complete, detection results are compared, and the channel with the clearest evidence of an S3 and/or S4 is selected for presentation on the report. Statements are displayed only for the selected channel.

Processing the Sound Data
The sound data for each channel is then processed by removing offsets, prescaling, and filtering the data into narrow frequency bands to optimize the detection of each S1 through S4 heart sound. Noise floors are identified in each filtered frequency band and a beat-to-beat correlation performed. If the noise levels are too high or the beat-to-beat correlation is too low, then the analysis is aborted and the user informed of excessive noise or poor signal quality.

Using ECG for timing purposes, the S1 and S2 detection time windows are identified for each beat. Utilizing a threshold adaptively computed from a moving window root mean square for each frequency band, the location of each S1 and S2 is determined within the computed detection window.

The S3 detection time windows are located using information within the single-channel of ECG and the computed position of the S2 offset. The energy content is determined within the S3 detection time window. Using a set of proprietary rules based on frequency and amplitude measurements, possible S3s are detected within the S3 windows and rejected if there is noise at the same time extending into higher frequency bands. The sum of non-rejected S3s above pre-set energy thresholds is compared to the number of possible beats and a determination is made.

The S4 detection time windows are located based on PQ intervals and Q-wave onset positions. Further processing on S4s is similar to that described above for S3s.

Filtering Noise and Artifact
Ambient noise and artifact are handled in a variety of ways. Aggressive but non-distorting bandpass filtering keeps unwanted signal energy out of the frequency bands of interest for S3 and S4 (in the under 100 Hz range). The algorithm searches for noise within individual detection time windows (being broader in energy content than that expected for an S3 or S4), and if such broader energy is present, the beat is rejected for further analysis.

If the heart rate is increased to where the detection time window of the S3 merges with the detection time window of the S4, it becomes impossible to know with certainty if there is solely an S4, an S3, or if there are both. In this situation, the algorithm indicates that there is a summation likely.

The results of the heart sound analysis (S3 detected, S4 detected, or possible summation) are printed on the AUDICOR report. Finally, a representative beat is chosen for display on the AUDICOR report as one with low noise characteristics and the absence of ectopy.

Roos M, Toggweler S, Zuber M, Jamshidi P, Erne P. Acoustic cardiographic parameters and their relationship to invasive hemodynamic measurements in patients with LV systolic dysfunction. Congestive Heart Failure. 2006;12(4 suppl 1):19-24.

Top