Cardiac auscultation is an essential point-of-care method used for the early diagnosis of heart diseases. Automatic analysis of heart sounds for abnormality detection is faced with the challenges of additive noise and sensor-dependent degradation. This paper aims to develop methods to address the cardiac abnormality detection problem when both types of distortions are present in the cardiac auscultation sound. We first mathematically analyze the effect of additive and convolutional noise on short-term filterbank-based features and a Convolutional Neural Network (CNN) layer. Based on the analysis, we propose a combination of linear and logarithmic spectrogram-image features. These 2D features are provided as input to a residual CNN network (ResNet) for heart sound abnormality detection. Experimental validation is performed on an open-access heart sound abnormality detection dataset involving noisy recordings obtained from multiple stethoscope sensors. The proposed method achieves significantly improved results compared to the conventional approaches, with an area under the ROC (receiver operating characteristics) curve (AUC) of 91.36%, F-1 score of 84.09%, and Macc (mean of sensitivity and specificity) of 85.08%. We also show that the proposed method shows the best mean accuracy across different source domains including stethoscope and noise variability, demonstrating its effectiveness in different recording conditions. The proposed combination of linear and logarithmic features along with the ResNet classifier effectively minimizes the impact of background noise and sensor variability for classifying phonocardiogram (PCG) signals. The proposed method paves the way towards developing computer-aided cardiac auscultation systems in noisy environments using low-cost stethoscopes.
Deep Attention-based Representation Learning for Heart Sound Classification
University of Augsburg(奥格斯堡大学);The University of Tokyo(东京大学);Shenzhen University General Hospital(深圳大学总医院);Wenzhou Medical University First Affiliated Hospital(温州医科大学附属第一医院);Imperial College London(帝国理工学院)
Training one model to detect heart and lung sound events from single point auscultations
Institute of Artificial Intelligence, Philipps-University Marburg(马尔堡大学人工智能研究所);Clinic of Oral and Cranio-Maxillofacial Surgery, University Hospital Basel(巴塞尔大学医院口腔颌面外科诊所);Division of Respiratory and Critical Care Medicine, Philipps-University, Marburg(马尔堡大学呼吸与重症医学科);Division of Internal Medicine, University Hospital Basel(巴塞尔大学医院内科);Department of Gastroenterology, Endocrinology, Metabolism and Infectiology, Philipps-University, Marburg(马尔堡大学消化内科、内分泌科、代谢与感染病科);Cardiology Department, University Hospital Gießen and Marburg(吉森和马尔堡大学医院心脏病科);Department of Cardiology, University Heart and Vascular Center Hamburg(汉堡大学心脏和血管中心心脏病科)
Enhanced Heart Sound Classification Using Mel Frequency Cepstral Coefficients and Comparative Analysis of Single vs. Ensemble Classifier Strategies
National Yunlin University of Science and Technology(云林科技大学);Sejong University(世宗大学);Islamic Azad University(伊斯兰阿扎德大学);Prince Sattam bin Abdulaziz University(萨塔姆·本·阿卜杜勒-阿齐兹亲王大学);Taif University(塔伊夫大学);Lebanese French University(黎巴嫩法国大学);UNSW Sydney(新南威尔士大学);Duy Tan University(维新大学)
A Comprehensive Survey on Heart Sound Analysis in the Deep Learning Era
University of Bremen(不来梅大学);Imperial College London(伦敦帝国学院);Griffith University(格里菲斯大学);Beijing Institute of Technology(北京理工大学);University of Augsburg(奥格斯堡大学);Technical University of Munich(慕尼黑工业大学);GLAM, Imperial College London(伦敦帝国学院GLAM机构)
Elastic Net Regularization and Gabor Dictionary for Classification of Heart Sound Signals using Deep Learning
Department of Signal Theory and Communications, Universidad Carlos III de Madrid(卡洛斯三世大学信号理论与通信系);Department of Electrical Engineering, Faculty of Engineering, Aswan University(阿斯旺大学电气工程系)
Recognizing Abnormal Heart Sounds Using Deep Learning
Philips Research North America(飞利浦北美研究院);PARC, A Xerox Company(施乐帕洛阿尔托研究中心)
本文介绍的工作将深度学习应用于自动心脏听诊的任务,即识别心音中的异常。我们描述了一种自动心音分类算法,该算法将时频热图表示与深度卷积神经网络(CNN)的使用相结合。考虑到错误分类的成本敏感性,我们的CNN架构使用修改后的损失函数进行训练,该函数直接优化了灵敏度和特异性之间的权衡。我们在2016年PhysioNet Computing in Cardiology挑战赛中评估了我们的算法,其目标是从单个,短的,可能有噪音的录音中准确分类正常和异常心音。我们进入挑战的最终特异性为0.95,灵敏度为0.73,总分为0.84。我们在所有挑战条目中获得了最大的特异性得分,并且仅使用单个CNN,与使用集成方法的排名第一的完成者相比,我们的算法在总体得分上仅相差0.02。
Time-Frequency Distributions of Heart Sound Signals: A Comparative Study using Convolutional Neural Networks
King’s College London(伦敦国王学院);Kuwait College of Science and Technology(科威特科技学院);Université du Luxembourg(卢森堡大学);Carthage University(迦太基大学);Université de Kindu(金杜大学)
Deep Recurrent Learning for Heart Sounds Segmentation based on Instantaneous Frequency Features
Facultad de Ingeniería, Universidad de Buenos Aires(布宜诺斯艾利斯大学工程学院);Instituto Argentino de Matemática ”Alberto P. Calderón”, CONICET(国家科学技术研究委员会“阿尔贝托·P·卡尔德隆”阿根廷数学研究所)
Temporal-Framing Adaptive Network for Heart Sound Segmentation without Prior Knowledge of State Duration
Southeast University(东南大学);Emory University School of Medicine(埃默里大学医学院);Georgia Institute of Technology(佐治亚理工学院)
目的:提出了一种基于时间成帧自适应网络(TFAN)的心音分割算法,该算法包括状态转移丢失和动态推断,用于解码最可能的状态序列。研究方法:与以前的最先进的方法相比,基于TFAN的方法不需要任何关于心音状态持续时间的知识,因此可能推广到非窦性心律。基于TFAN的方法在2016年PhysioNet/Computer in Cardiology Challenge的训练集A中随机选择的50个记录上进行训练,并在其他12个独立的训练和测试数据库(2099个记录和52180次心跳)上进行测试。将用于细分的数据库分为三个难度递增的等级(第一、第二和第三等级),用于性能报告。结果如下:基于TFAN的方法在所有12个数据库中获得了优异的F1分数,除了“测试B”,平均为96.7%,而最先进的方法为94.6%。此外,基于TFAN的方法在LEVEL-I、-II和-III数据上的总体F1评分分别为99.2%、94.4%和91.4%,而当前最先进的方法分别为98.4%、88.54%和79.80%。结论:因此,基于TFAN的方法提供了实质性的改进,特别是对于更困难的情况,以及在公共训练数据中未表示的数据集。意义:所提出的方法是高度灵活的,可能适用于其他非平稳时间序列。需要进一步的工作来理解这种方法将在多大程度上提供改进的诊断性能,尽管假设更好的分割将导致改进的诊断是合乎逻辑的。
H-Infinity Filter Enhanced CNN-LSTM for Arrhythmia Detection from Heart Sound Recordings
Department of Computer Science and Engineering(计算机科学与工程系);National Institute of Technology Karnataka, Surathkal, Karnataka, India 575025(卡纳塔克邦苏拉塔卡尔国家理工学院)
本文提出CNN-H-Infinity-LSTM架构,通过引入受控制理论H-Infinity滤波器启发的可训练参数,增强模型鲁棒性与泛化能力,在PhysioNet CinC Challenge 2016数据集上实现99.42%准确率,有效解决心音记录心律失常检测问题。
Automatic Feature Extraction for Heartbeat Anomaly Detection
Romanian Institute of Science and Technology(罗马尼亚科学技术研究院);Babes-Bolyai University(巴比什-波雅依大学);Max Planck Institute for Mathematics in the Sciences(马克斯·普朗克数学与科学研究所)
Scaling to Multimodal and Multichannel Heart Sound Classification with Synthetic and Augmented Biosignals
School of Electrical Engineering, Computing, and Mathematical Sciences (EECMS), Faculty of Science and Engineering, Curtin University, Bentley, WA 6102, Australia(电气工程、计算与数学科学学院(EECMS),科学与工程学院, Curtin 大学,Bentley,WA 6102,澳大利亚)
Diffusion-Based Heart Sound Generation: Evaluation with Physiological Signal Metrics, Classifiers, and Expert Listening
Department of Information Science & Engineering, KTH Royal Institute of Technology(信息科学与工程系,皇家理工学院);Rutherford Appleton Laboratory(拉瑟福德·苹果顿实验室);Peng Cheng Laboratory(鹏城实验室);Department of Engineering, King’s College London(工程系,伦敦国王学院)