Deep Unsupervised Representation Learning for Abnormal Heart Sound Classification

Deep Learning in Heart Sound Analysis: From Techniques to Clinical Applications

Importance: Heart sound auscultation is a routinely used physical examination in clinical practice to identify potential cardiac abnormalities. However, accurate interpretation of heart sounds requires specialized training and experience, which limits its generalizability. Deep learning, a subset of machine learning, involves training artificial neural networks to learn from large datasets and perform complex tasks with intricate patterns. Over the past decade, deep learning has been successfully applied to heart sound analysis, achieving remarkable results and accumulating substantial heart sound data for model training. Although several reviews have summarized deep learning algorithms for heart sound analysis, there is a lack of comprehensive summaries regarding the available heart sound data and the clinical applications. Highlights: This review will compile the commonly used heart sound datasets, introduce the fundamentals and state-of-the-art techniques in heart sound analysis and deep learning, and summarize the current applications of deep learning for heart sound analysis, along with their limitations and areas for future improvement. Conclusions: The integration of deep learning into heart sound analysis represents a significant advancement in clinical practice. The growing availability of heart sound datasets and the continuous development of deep learning techniques contribute to the improvement and broader clinical adoption of these models. However, ongoing research is needed to address existing challenges and refine these technologies for broader clinical use.

DDLA: a double deep latent autoencoder for diabetic retinopathy diagnose based on continuous glucose sensors

The current diagnosis of diabetic retinopathy is based on fundus images and clinical experience. However, considering the ineffectiveness and non-portability of medical devices, we aimed to develop a diagnostic model for diabetic retinopathy based on glucose series data from the wearable continuous glucose monitoring system. Therefore, this study developed a novel method, i.e., double deep latent autoencoder, for exploring glycemic variability influence from multi-day glucose data for diabetic retinopathy. Specifically, the model proposed in this research could encode continuous glucose sensor data with non-continuous and variable length via the integration of a data reorganization module and a novel encoding module with fragmented-missing-wise objective function. Additionally, the model implements a double deep autoencoder, which integrated convolutional neural network, long short-term memory, to jointly capturing the inter-day and intra-day glucose latent features from glucose series. The effectiveness of the proposed model is evaluated through a cross-validation method to clinical datasets of 765 type 2 diabetes patients. The proposed method achieves the highest accuracy value (0.89), precision value (0.88), and F1 score (0.73). The results suggest that our model can be used to remotely diagnose and screen for diabetic retinopathy by learning potential features of glucose series data collected by wearable continuous glucose monitoring systems.

Accurate Localization of First and Second Heart Sounds via Template Matching in Forcecardiography Signals

Cardiac auscultation is an essential part of physical examination and plays a key role in the early diagnosis of many cardiovascular diseases. The analysis of phonocardiography (PCG) recordings is generally based on the recognition of the main heart sounds, i.e., S1 and S2, which is not a trivial task. This study proposes a method for an accurate recognition and localization of heart sounds in Forcecardiography (FCG) recordings. FCG is a novel technique able to measure subsonic vibrations and sounds via small force sensors placed onto a subject's thorax, allowing continuous cardio-respiratory monitoring. In this study, a template-matching technique based on normalized cross-correlation was used to automatically recognize heart sounds in FCG signals recorded from six healthy subjects at rest. Distinct templates were manually selected from each FCG recording and used to separately localize S1 and S2 sounds, as well as S1-S2 pairs. A simultaneously recorded electrocardiography (ECG) trace was used for performance evaluation. The results show that the template matching approach proved capable of separately classifying S1 and S2 sounds in more than 96% of all heartbeats. Linear regression, correlation, and Bland-Altman analyses showed that inter-beat intervals were estimated with high accuracy. Indeed, the estimation error was confined within 10 ms, with negligible impact on heart rate estimation. Heart rate variability (HRV) indices were also computed and turned out to be almost comparable with those obtained from ECG. The preliminary yet encouraging results of this study suggest that the template matching approach based on normalized cross-correlation allows very accurate heart sounds localization and inter-beat intervals estimation.

Gun identification from gunshot audios for secure public places using transformer learning

Increased mass shootings and terrorist activities severely impact society mentally and physically. Development of real-time and cost-effective automated weapon detection systems increases a sense of safety in public. Most of the previously proposed methods were vision-based. They visually analyze the presence of a gun in a camera frame. This research focuses on gun-type (rifle, handgun, none) detection based on the audio of its shot. Mel-frequency-based audio features have been used. We compared both convolution-based and fully self-attention-based (transformers) architectures. We found transformer architecture generalizes better on audio features. Experimental results using the proposed transformer methodology on audio clips of gunshots show classification accuracy of 93.87%, with training loss and validation loss of 0.2509 and 0.1991, respectively. Based on experiments, we are convinced that our model can effectively be used as both a standalone system and in association with visual gun-detection systems for better security.

A wearable multi-modal acoustic system for breathing analysis

Chronic obstructive pulmonary disease (COPD) is the third leading cause of death worldwide with over 3 × 106 deaths in 2019. Such an alarming figure becomes frightening when combined with the number of lost lives resulting from COVID-caused respiratory failure. Because COPD exacerbations identified early can commonly be treated at home, early symptom detections may enable a major reduction of COPD patient readmission and associated healthcare costs; this is particularly important during pandemics such as COVID-19 in which healthcare facilities are overwhelmed. The standard adjuncts used to assess lung function (e.g., spirometry, plethysmography, and CT scan) are expensive, time consuming, and cannot be used in remote patient monitoring of an acute exacerbation. In this paper, a wearable multi-modal system for breathing analysis is presented, which can be used in quantifying various airflow obstructions. The wearable multi-modal electroacoustic system employs a body area sensor network with each sensor-node having a multi-modal sensing capability, such as a digital stethoscope, electrocardiogram monitor, thermometer, and goniometer. The signal-to-noise ratio (SNR) of the resulting acoustic spectrum is used as a measure of breathing intensity. The results are shown from data collected from over 35 healthy subjects and 3 COPD subjects, demonstrating a positive correlation of SNR values to the health-scale score.

Machine Listening for Heart Status Monitoring: Introducing and Benchmarking HSS - the Heart Sounds Shenzhen Corpus

Auscultation of the heart is a widely studied technique, which requires precise hearing from practitioners as a means of distinguishing subtle differences in heart-beat rhythm. This technique is popular due to its non-invasive nature, and can be an early diagnosis aid for a range of cardiac conditions. Machine listening approaches can support this process, monitoring continuously and allowing for a representation of both mild and chronic heart conditions. Despite this potential, relevant databases and benchmark studies are scarce. In this paper, we introduce our publicly accessible database, the Heart Sounds Shenzhen Corpus (HSS), which was first released during the recent INTERSPEECH 2018 ComParE Heart Sound sub-challenge. Additionally, we provide a survey of machine learning work in the area of heart sound recognition, as well as a benchmark for HSS utilising standard acoustic features and machine learning models. At best our support vector machine with Log Mel features achieves 49.7% unweighted average recall on a three category task (normal, mild, moderate/severe).