Key result
An entropygram-based 2D CNN achieves ~99% accuracy diagnosing heart disease from noisy audio signals.
Why the study?
Stethoscope auscultation depends on physician skill and is often performed in areas of high ambient noise, limiting early diagnosis.
Does an automatic diagnosing method using entropygrams and a 2D CNN accurately detect heart disease from audio signals in noisy environments?
Does an automatic diagnosing method using entropygrams and a 2D CNN accurately detect heart disease from audio signals in noisy environments?
An automatic diagnostic method using entropygrams and a 2D CNN achieved 99.43% accuracy in detecting heart disease from audio signals even in highly noisy environments.
The most common diseases worldwide are those that are heart failure related, every year 17.9 million people die from this problem, many of the deaths might have been prevented if they had been diagnosed at early stages. In order to detect some heart diseases, medical doctors auscultate the patient by hearing the heart beats with a stethoscope. The proper diagnosis depends on the skills of the professional, but frequently the doctor has to listen in areas of high ambient noise. We propose an automatic diagnosing method which consists in producing images with the spectrograms of entropy (i.e. entropygrams) from the audio signal of the heart beats of the patients, and using a 2D convolutional neural network for predicting if the entropygram comes from a healthy individual or from a patient with a heart disease. To reproduce a noisy environment we mixed the original audios with white noise achieving a Signal to Noise Ratio up to -10 dB. Entropygrams showed indeed robustness to noise and classification achieved an accuracy of 99.43 % with the noisiest audios.
No takes yet. Share an insight, caveat, or question.
Castro–Coria et al. (2024) studied Heart disease. Automatic diagnosing method using entropygrams and a 2D convolutional neural network was evaluated on Classification accuracy. An automatic diagnosing method using entropygrams and a 2D convolutional neural network achieved 99.43% accuracy in classifying heart disease from audio signals, even with high ambient noise.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: