Acoustic diagnosis of pulmonary hypertension: automated speech- recognition-inspired classification algorithm outperforms physicians

Abstract

We hypothesized that an automated speech- recognition-inspired classification algorithm could differentiate between the heart sounds in subjects with and without pulmonary hypertension (PH) and outperform physicians. Heart sounds, electrocardiograms, and mean pulmonary artery pressures (mPAp) were recorded simultaneously. Heart sound recordings were digitized to train and test speech-recognition-inspired classification algorithms. We used mel-frequency cepstral coefficients to extract features from the heart sounds. Gaussian-mixture models classified the features as PH (mPAp ≥ 25 mmHg) or normal (mPAp < 25 mmHg). Physicians blinded to patient data listened to the same heart sound recordings and attempted a diagnosis. We studied 164 subjects: 86 with mPAp ≥ 25 mmHg (mPAp 41 ± 12 mmHg) and 78 with mPAp < 25 mmHg (mPAp 17 ± 5 mmHg) (p < 0.005). The correct diagnostic rate of the automated speech-recognition-inspired algorithm was 74% compared to 56% by physicians (p = 0.005). The false positive rate for the algorithm was 34% versus 50% (p = 0.04) for clinicians. The false negative rate for the algorithm was 23% and 68% (p = 0.0002) for physicians. We developed an automated speech-recognition-inspired classification algorithm for the acoustic diagnosis of PH that outperforms physicians that could be used to screen for PH and encourage earlier specialist referral.

Deep learning to predict elevated pulmonary artery pressure in patients with suspected pulmonary hypertension using standard chest X ray

“…However, the AUC was limited as 0.60-0.62. Moreover, there were several methods including laboratory data, electrocardiogram, and physical examinations for detection of PH [24][25][26][27] . In previous studies, the number of invasive data was limited, and the accuracy was also limited as AUCs up to 0.65 for these methods.…”

The virtual physical exam in the 21st century

“…29 Digital stethoscopes can record, analyse and send heart sounds to the clinician to diagnose heart murmurs or pulmonary hypertension. 30,31 Point-of-care ultrasounds are now connected to smartphones and are more accessible than ever with decreasing costs.…”

Digital stethoscope: technology update

“…For instance, Kaddoura et al used automated machine learning and language-recognition-inspired-speech algorithm to ascertain if the digitally acquired heart sounds were linked to pulmonary hypertension (PH). 29 The algorithm used was able to closely examine the heart sound and collect information such as amplitude, intensity, shape, and frequency. The heart sounds from PH patients were compared to non-PH patients, and they discovered the algorithm accurately diagnosed PH 74% of the time.…”

“…The heart sounds from PH patients were compared to non-PH patients, and they discovered the algorithm accurately diagnosed PH 74% of the time. 29 These recent discoveries have opened the door for further research to further optimize the current technology, to ultimately empower the physician to better assist the patient.…”