[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125992-en":3,"doc-seo-125992-105":31,"detail-sidebar-cat-0-en-105":93},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},125992,687207024643,"Oliver","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","Improving Machine Learning Based Sepsis Diagnosis Using Heart Rate Variability","Early and accurate sepsis diagnosis is essential for improving patient outcomes. This study develops a predictive approach using heart rate variability (HRV) features, identifying critical HRV signals via feature engineering methods such as statistical bootstrapping and the Boruta algorithm. XGBoost and Random Forest models are trained with differential hyperparameters, followed by ensemble pooling of high-recall and high-precision probabilities to enhance performance. A neural network achieves an F1 score of 0.805, with precision 0.851 and recall 0.763, and interpretability is analyzed using LIME to clarify decision thresholds.","Improving Machine Learning Based Sepsis Diagnosis Using Heart Rate Variability  \nSai Balaji  \nThe Woodlands College Park High School The Woodlands, TX, United States [sai.s.balaji@gmail.com](sai.s.balaji@gmail.com)  \nChristopher Sun  \nStanford University Stanford, CA, United States [chrisun@stanford.edu](chrisun@stanford.edu)  \nAnaiy Somalwar  \nUniversity of California, Berkeley Berkeley, CA, United States [anaiy@berkeley.edu](anaiy@berkeley.edu)  \narXiv :2408 .02683v1 [ cs .LG] 1 Aug 2024  \nAbstract—The early and accurate diagnosis of sepsis is critical for enhancing patient outcomes. This study aims to use heart rate variability (HRV) features to develop an effective predictive model for sepsis detection. Critical HRV features are identified through feature engineering methods, including statistical bootstrapping and the Boruta algorithm, after which XGBoost and Random Forest classifiers are trained with differential hyperparameter settings. In addition, ensemble models are constructed to pool the prediction probabilities of high-recall and high-precision classifiers and improve model performance. Finally, a neural network model is trained on the HRV features, achieving an F1 score of 0.805, a precision of 0.851, and a recall of 0.763. The best-performing machine learning model is compared to this neural network through an interpretability analysis, where Local Interpretable Model-agnostic Explanations are implemented to determine decision-making criterion based on numerical rangesand thresholds for specific features. This study not only highlights the efficacy of HRV in automated sepsis diagnosis but also increases the transparency of black box outputs, maximizing clinical applicability.  \nIndex Terms—Sepsis, Heart Rate Variability, Feature Engineering, Machine Learning, Model Interpretability  \nI. INTRODUCTION  \nSepsis poses a significant global health issue with a high mortality rate and economic burden [1] [2] . Sepsis is characterized by a dysregulated host response to infection, causing fever and an abnormal white blood cell count [3] . Accurate diagnosis of sepsis is critical for reducing mortality rates and substantial healthcare costs [4] [5] . Though advances like the quick Sequential Organ Failure Assessment (qSOFA) may enhance early sepsis detection, the frequent vital sign monitoring and reassessment required adds significant burdens to nursing workload, in return for already-questionable timelinessand accuracy [6] . Heart rate variability (HRV), which reflects autonomic nervous system activity and is extractable from heart-monitoring devices like ECG, emerges as a promising non-invasive tool for early sepsis detection due to its ability to detect dysregulation before clinical symptoms appear [7],[8], [9], [10] .  \nBedoya et al. (2020) and van Wijk et al. (2023) employ machine learning and continuous ECG monitoring for sepsis prediction, harnessing both invasive and non-invasive patient data to advance early sepsis detection standards [11] [12] . In contrast, our research zeroes in on exploiting only HRV met-  \nrics for non-invasive sepsis diagnosis, offering a streamlined, patient-centric approach.  \nSendak et al. (2020) successfully implement “Sepsis Watch,” a deep learning based sepsis detection tool requiring both invasive and non-invasive clinical data. Integration of this tool into everyday clinical practice demonstrates the practicality of enhancing sepsis management with advanced machine learning techniques [13] . Henry et al. (2015) develop TREWScore, a predictive model for septic shock utilizing a wide array of physiological and laboratory data, outperforming traditional methods in early detection [14] . Building upon these two works, our study aims to demonstrate that similar state-of-the-art models can operate on HRV metrics exclusively, resulting in a simpler feature space and a more explainable and effective method for early sepsis detection ina clinical environment.  \nAdditionally, in comparison t","cbCaic4TwP984LVG","https://ap.wps.com/l/cbCaic4TwP984LVG","pdf",545673,5,1,9,"English","en",105,"# Introduction\n## Clinical challenge of sepsis and need for early detection\n## Promise of heart rate variability (HRV)\n## Related work and study contribution\n# Methods\n## Data availability and dataset description\n## Data splitting, class imbalance handling, and leakage prevention","[{\"question\":\"What is the main goal of the study?\",\"answer\":\"To build an effective sepsis detection predictive model using heart rate variability (HRV) features, aiming for accurate early diagnosis.\"},{\"question\":\"How are important HRV features selected?\",\"answer\":\"Critical HRV features are identified through feature engineering methods including statistical bootstrapping and the Boruta algorithm.\"},{\"question\":\"What model performance and interpretability approach are reported?\",\"answer\":\"The neural network reaches an F1 score of 0.805 (precision 0.851, recall 0.763), and interpretability is provided using LIME to determine decision-making criteria based on feature ranges and thresholds.\"}]","Improving Machine Learning Based Sepsis Diagnosis Using Heart Rate Variability | PDF",1785902437,23,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"improving-machine-learning-based-sepsis-diagnosis-using-heart-rate-variability","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/improving-machine-learning-based-sepsis-diagnosis-using-heart-rate-variability/125992/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-08-24","2026-08-05",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What is the main goal of the study?","Question",{"text":77,"@type":78},"To build an effective sepsis detection predictive model using heart rate variability (HRV) features, aiming for accurate early diagnosis.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How are important HRV features selected?",{"text":82,"@type":78},"Critical HRV features are identified through feature engineering methods including statistical bootstrapping and the Boruta algorithm.",{"name":84,"@type":75,"acceptedAnswer":85},"What model performance and interpretability approach are reported?",{"text":86,"@type":78},"The neural network reaches an F1 score of 0.805 (precision 0.851, recall 0.763), and interpretability is provided using LIME to determine decision-making criteria based on feature ranges and thresholds.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,111,116,121,124,128,131,135],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":22,"doc_module":4,"doc_module_name":47,"category_name":125,"show_sort_weight":126,"slug":127},"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":47,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":20,"slug":138},19,"General","general"]