We present first results showing that it is possible to automatically distinguish coughing sounds produced by patients with TB and those with other lung ailments . Our experiments are based on a dataset of cough recording in a real-world clinic setting from 16 patients confirmed to besuffering from TB and 33 patients that are suffering from respiratory conditions, confirmed as other than TB . We conclude that automaticclassification of cough audio sounds is promising as a viable means of low-costeasily-deployable front-line screening for TB, which will greatly benefitdeveloping countries with a heavy TB burden . This systemachieves a sensitivity of 93% at a specificity of 95% and thus exceeds the 90\%sensitivity at 70% specificity specification considered by the WHO as minimal requirements for community-based TB triage test . The best system achieves anarea under the ROC curve (AUC) of 0.94 using 23 features selected from a set of78 high-resolution mel-frequency cepstral coefficients (MFCCs). This system achieves a set .

Author(s) : Madhurananda Pahar, Marisa Klopper, Byron Reeve, Grant Theron, Rob Warren, Thomas Niesler

Links : PDF - Abstract

Code :
Coursera

Keywords : tb - patients - cough - screening - based -

Leave a Reply

Your email address will not be published. Required fields are marked *