Cyril Mani, Tanya S Paul, Patrick M Archambault, Alexandre Marois
{"title":"探索类任务中边缘计算心律失常检测的机器学习工作流程。","authors":"Cyril Mani, Tanya S Paul, Patrick M Archambault, Alexandre Marois","doi":"10.1038/s41526-024-00409-0","DOIUrl":null,"url":null,"abstract":"<p><p>Deep-space missions require preventative care methods based on predictive models for identifying in-space pathologies. Deploying such models requires flexible edge computing, which Open Neural Network Exchange (ONNX) formats enable by optimizing inference directly on wearable edge devices. This work demonstrates an innovative approach to point-of-care machine learning model pipelines by combining this capacity with an advanced self-optimizing training scheme to classify periods of Normal Sinus Rhythm (NSR), Atrial Fibrillation (AFIB), and Atrial Flutter (AFL). 742 h of electrocardiogram (ECG) recordings were pre-processed into 30-second normalized samples where variable mode decomposition purged muscle artifacts and instrumentation noise. Seventeen heart rate variability and morphological ECG features were extracted by convoluting peak detection with Gaussian distributions and delineating QRS complexes using discrete wavelet transforms. The decision tree classifier's features, parameters, and hyperparameters were self-optimized through stratified triple nested cross-validation ranked on F1-scoring against cardiologist labeling. The selected model achieved a macro F1-score of 0.899 with 0.993 for NSR, 0.938 for AFIB, and 0.767 for AFL. The most important features included median P-wave amplitudes, PRR20, and mean heart rates. The ONNX-translated pipeline took 9.2 s/sample. This combination of our self-optimizing scheme and deployment use case of ONNX demonstrated overall accurate operational tachycardia detection.</p>","PeriodicalId":54263,"journal":{"name":"npj Microgravity","volume":null,"pages":null},"PeriodicalIF":4.4000,"publicationDate":"2024-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11193813/pdf/","citationCount":"0","resultStr":"{\"title\":\"Machine learning workflow for edge computed arrhythmia detection in exploration class missions.\",\"authors\":\"Cyril Mani, Tanya S Paul, Patrick M Archambault, Alexandre Marois\",\"doi\":\"10.1038/s41526-024-00409-0\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Deep-space missions require preventative care methods based on predictive models for identifying in-space pathologies. Deploying such models requires flexible edge computing, which Open Neural Network Exchange (ONNX) formats enable by optimizing inference directly on wearable edge devices. This work demonstrates an innovative approach to point-of-care machine learning model pipelines by combining this capacity with an advanced self-optimizing training scheme to classify periods of Normal Sinus Rhythm (NSR), Atrial Fibrillation (AFIB), and Atrial Flutter (AFL). 742 h of electrocardiogram (ECG) recordings were pre-processed into 30-second normalized samples where variable mode decomposition purged muscle artifacts and instrumentation noise. Seventeen heart rate variability and morphological ECG features were extracted by convoluting peak detection with Gaussian distributions and delineating QRS complexes using discrete wavelet transforms. The decision tree classifier's features, parameters, and hyperparameters were self-optimized through stratified triple nested cross-validation ranked on F1-scoring against cardiologist labeling. The selected model achieved a macro F1-score of 0.899 with 0.993 for NSR, 0.938 for AFIB, and 0.767 for AFL. The most important features included median P-wave amplitudes, PRR20, and mean heart rates. The ONNX-translated pipeline took 9.2 s/sample. This combination of our self-optimizing scheme and deployment use case of ONNX demonstrated overall accurate operational tachycardia detection.</p>\",\"PeriodicalId\":54263,\"journal\":{\"name\":\"npj Microgravity\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":4.4000,\"publicationDate\":\"2024-06-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11193813/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"npj Microgravity\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1038/s41526-024-00409-0\",\"RegionNum\":1,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"MULTIDISCIPLINARY SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"npj Microgravity","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1038/s41526-024-00409-0","RegionNum":1,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MULTIDISCIPLINARY SCIENCES","Score":null,"Total":0}
引用次数: 0
摘要
深空任务需要基于预测模型的预防性护理方法,以识别空间病症。部署此类模型需要灵活的边缘计算,而开放神经网络交换(ONNX)格式可通过直接在可穿戴边缘设备上优化推理来实现。这项工作通过将这种能力与先进的自我优化训练方案相结合,对正常窦性心律(NSR)、心房颤动(AFIB)和心房扑动(AFL)期进行分类,展示了一种创新的护理点机器学习模型管道方法。742 小时的心电图(ECG)记录被预处理成 30 秒的归一化样本,其中可变模式分解清除了肌肉伪影和仪器噪音。通过高斯分布卷积峰值检测和离散小波变换划分 QRS 波群,提取了 17 个心率变异性和形态心电图特征。决策树分类器的特征、参数和超参数通过分层三重嵌套交叉验证进行了自我优化,根据心脏病专家的标记进行 F1 评分排名。所选模型的宏观 F1 得分为 0.899,其中 NSR 为 0.993,AFIB 为 0.938,AFL 为 0.767。最重要的特征包括 P 波振幅中值、PRR20 和平均心率。ONNX翻译管道耗时9.2秒/样本。我们的自我优化方案与 ONNX 部署使用案例相结合,证明了操作性心动过速检测的整体准确性。
Machine learning workflow for edge computed arrhythmia detection in exploration class missions.
Deep-space missions require preventative care methods based on predictive models for identifying in-space pathologies. Deploying such models requires flexible edge computing, which Open Neural Network Exchange (ONNX) formats enable by optimizing inference directly on wearable edge devices. This work demonstrates an innovative approach to point-of-care machine learning model pipelines by combining this capacity with an advanced self-optimizing training scheme to classify periods of Normal Sinus Rhythm (NSR), Atrial Fibrillation (AFIB), and Atrial Flutter (AFL). 742 h of electrocardiogram (ECG) recordings were pre-processed into 30-second normalized samples where variable mode decomposition purged muscle artifacts and instrumentation noise. Seventeen heart rate variability and morphological ECG features were extracted by convoluting peak detection with Gaussian distributions and delineating QRS complexes using discrete wavelet transforms. The decision tree classifier's features, parameters, and hyperparameters were self-optimized through stratified triple nested cross-validation ranked on F1-scoring against cardiologist labeling. The selected model achieved a macro F1-score of 0.899 with 0.993 for NSR, 0.938 for AFIB, and 0.767 for AFL. The most important features included median P-wave amplitudes, PRR20, and mean heart rates. The ONNX-translated pipeline took 9.2 s/sample. This combination of our self-optimizing scheme and deployment use case of ONNX demonstrated overall accurate operational tachycardia detection.
npj MicrogravityPhysics and Astronomy-Physics and Astronomy (miscellaneous)
CiteScore
7.30
自引率
7.80%
发文量
50
审稿时长
9 weeks
期刊介绍:
A new open access, online-only, multidisciplinary research journal, npj Microgravity is dedicated to publishing the most important scientific advances in the life sciences, physical sciences, and engineering fields that are facilitated by spaceflight and analogue platforms.