听诊器
心音图
远程医疗
心音
医学
话筒
远程病人监护
噪音(视频)
卷积神经网络
计算机科学
异常
人工神经网络
波形
人工智能
实时计算
深度学习
语音识别
心前检查
可扩展性
信号(编程语言)
计算机硬件
心脏监护
钥匙(锁)
工程类
QRS波群
作者
Aida Jose,Aiswarya Pramod,Dominic Wilson,Godwin Raju,Sunitha S Pillai
出处
期刊:
[Institute of Electrical and Electronics Engineers]
日期:2026-07-01
卷期号:: 1-6
标识
DOI:10.1109/ictest69239.2026.11722718
摘要
Cardiovascular diseases (CVDs) are among the leading causes of death worldwide, a problem that hits hardest in low-resource and rural settings where specialized diagnostic equipment and cardiologists are scarce. Conventional acoustic stethoscopes depend on the physician's hearing and clinical experience, which introduces subjectivity and reduces detection accuracy for subtle cardiac abnormalities. This paper presents an AI-enabled smart digital stethoscope designed for cardiovascular abnormality detection and remote cardiac monitoring. The system uses an electronic stethoscope sensor for heart sound acquisition, followed by analog signal conditioning via a low-noise preamplifier, fourth-order band-pass filtering, and digital noise suppression to produce high-quality phonocardiogram (PCG) signals. These signals are converted into Mel-spectrogram representations and fed into Convolutional Neural Network (CNN)-based deep learning models for automated classification of normal and abnormal cardiac conditions. The hardware runs on a Raspberry Pi embedded platform integrated with a secure IoT-enabled telemedicine interface supporting remote waveform visualization, cloud-based storage, and physician consultation. Experimental evaluation yielded an overall classification accuracy of 92.9% with an Area Under the Curve (AUC) of 0.995, confirming the approach's effectiveness for early cardiovascular abnormality detection. The result is a low-cost, portable, and scalable cardiac monitoring solution suited to telemedicine applications, rural healthcare centers, and continuous patient monitoring.
科研通智能强力驱动
Strongly Powered by AbleSci AI