Computer-Assisted Stethoscope Hears What the Ear Cannot

Diana Phillips

February 24, 2016

A novel computer-assisted stethoscope offers clinicians real-time help in identifying respiratory sounds that can be difficult to classify on chest auscultation, according to a research letter published online February 16 in the Annals of Internal Medicine. But at least one expert cautions against relying too heavily on electronics when examining a patient.

Differentiating respiratory sounds into the five generally accepted categories specified by the International Lung Sounds Association (normal, rhonchi, wheezes, fine crackles, and coarse crackles) can be challenging because of overlapping frequencies, Shinichiro Ohshimo, MD, PhD, from Hiroshima University in Japan, and colleagues explain. "The frequency of normal respiratory sound is 100 to 1000 Hz, wheeze is 100 to 5000 Hz, rhonchus is 150 Hz, coarse crackle is 350 Hz, and fine crackle is 650 Hz."

To classify the sounds, the computer-assisted stethoscope uses a separation algorithm that physically and mathematically divides them into each of the five groups according to the duration, frequency, and strength of the sound. "It also compares the sounds in each group with sound templates to identify the amount of overlap among each category," the authors write.

The templates are based on lung sounds that were identically classified by three respiratory physicians into each of the five sound categories based on their independent review of recorded lung sounds from 878 individuals.

The visual display consists of a pentagon chart in which each vertex represents one of the respiratory sounds, and lines from the center of the pentagon to each vertex show how much of the respective sound is present.

"The chart displays the measurement by plotting a point on each line, connecting each point to adjacent points with new lines, and coloring the area between the pentagon's center and the new lines in red," the authors write.

The classification and display of the sound information happen in real time, a feature that sets the e-stethoscope apart from others that focus primarily on digitally recording breath sounds, the authors write.

"We believe that the result is a new way of evaluating respiratory sounds that may help clinicians better understand what they hear," the authors write, noting that the recordings can be stored in patients' electronic health records. "We also believe that this stethoscope could be used in medical education; the intensive care unit; the operating room; and other disciplines where invasive procedures are done, such as bronchoscopy and gastroenteroscopy, and respiratory monitoring is believed to be useful."

However, technologies such as this can be a double-edged sword, according to Asghar Rastegar, MD, professor of medicine at Yale Medical School, New Haven, Connecticut. He noted that the device may have diagnostic and educational value, but warned that such tools can lead to complacency.

"With technology, we are able to see and hear things that we couldn't otherwise see or hear, which can help us get insight into what's going on," Dr Rastegar told Medscape Medical News. In contrast, overreliance on automated devices and decision-support technologies may keep clinicians from practicing and perfecting bedside examination skills and looking for diagnostic clues that are critical to understanding the "whole picture."

At its best, he explained, "technology should be an extension of what we are doing, not a replacement."

Dr Ohshimo and one coauthor hold patents for computer-assisted electronic stethoscopes. Dr Rastegar has disclosed no relevant financial relationships.

Ann Intern Med. Published online February 16, 2016. Letter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.