English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In recent years, many innovative solutions for recording and viewing soundsfrom a stethoscope have become available. However, to fully utilize suchdevices, there is a need for an automated approach for detecting abnormal lungsounds, which is better than the existing methods that typically have beendeveloped and evaluated using a small and non-diverse dataset.  We propose a machine learning based approach for detecting crackles in lungsounds recorded using a stethoscope in a large health survey. Our method istrained and evaluated using 209 files with crackles classified by expertlisteners. Our analysis pipeline is based on features extracted from smallwindows in audio files. We evaluated several feature extraction methods andclassifiers. We evaluated the pipeline using a training set of 175 cracklewindows and 208 normal windows. We did 100 cycles of cross validation where weshuffled training sets between cycles. For all the division between trainingand evaluation was 70%-30%.  We found and evaluated a 5-dimenstional vector with four features from thetime domain and one from the spectrum domain. We evaluated several classifiersand found SVM with a Radial Basis Function Kernel to perform best. Our approachhad a precision of 86% and recall of 84% for classifying a crackle in a window,which is more accurate than found in studies of health personnel. Thelow-dimensional feature vector makes the SVM very fast. The model can betrained on a regular computer in 1.44 seconds, and 319 crackles can beclassified in 1.08 seconds.  Our approach detects and visualizes individual crackles in recorded audiofiles. It is accurate, fast, and has low resource requirements. It can be usedto train health personnel or as part of a smartphone application for Bluetoothstethoscopes.

Feature Extraction for Machine Learning Based Crackle Detection in Lung Sounds from a Health Survey