Listening to a patients' breathing is certainly a key part of diagnosing their respiratory problems. However, doctors' individual observer bias certainly comes into play, plus it can be difficult to make a diagnosis when there's a lot of background noise. That's why scientists at Hiroshima University have created an electronic system that objectively matches lung sounds to specific maladies.
Developed in partnership with Fukushima Medical University and the Pioneer Corporation, the system consists of an electronic stethoscope and a computer program known as the Respiratory Sounds Visualizer.
In order to create that software, the scientists started by classifying the breathing noises of 878 patients with known respiratory disorders. This data was used to create a mathematical formula that evaluates the length, frequency, and intensity of lung sounds.
As a result, when audio from the stethoscope is analyzed by a computer or mobile device running Respiratory Sounds Visualizer, the software is able to match up patients' lung noises with corresponding disorders from its database. The program can even determine when multiple conditions are present, making themselves heard at the same time.
Plans call for the program to be made available as an app, which even patients could use to monitor ongoing conditions while at home. Similar technology is being developed at North Carolina State University, in the form of a wearable system that analyzes patients' wheezes to determine their condition.
Source: Hiroshima University
