AbstractClose vital signs monitoring is crucial for the clinical management of patients with dengue. We investigated performance of a non-invasive wearable utilising photoplethysmography (PPG), to provide real-time risk prediction in hospitalised individuals. We performed a prospective observational clinical study in Vietnam between January 2020 and October 2022: 153 patients were included in analyses, providing 1353 h of PPG data.

。我们利用光体积描记术（PPG）研究了非侵入性可穿戴设备的性能，以提供住院患者的实时风险预测。我们在2020年1月至2022年10月期间在越南进行了一项前瞻性观察性临床研究：153名患者被纳入分析，提供了1353小时的PPG数据。

Using a multi-modal transformer approach, 10-min PPG waveform segments and basic clinical data (age, sex, clinical features on admission) were used as features to continuously forecast clinical state 2 h ahead. Prediction of low-risk states (17,939/80,843; 22.1%), defined by NEWS2 and mSOFA < 6, was associated with an area under the precision-recall curve of 0.67 and an area under the receiver operator curve of 0.83.

使用多模式变压器方法，将10分钟PPG波形段和基本临床数据（年龄，性别，入院时的临床特征）用作特征，以连续预测提前2小时的临床状态。由NEWS2和mSOFA 6定义的低风险状态（17939/80843；22.1%）的预测与精确召回曲线下面积为0.67和接收器操作员曲线下面积为0.83相关。

Implementation of such interventions could provide cost-effective triage and clinical care in dengue, offering opportunities for safe ambulatory patient management..

实施此类干预措施可以为登革热提供具有成本效益的分诊和临床护理，为安全的门诊患者管理提供机会。。

IntroductionDengue exerts a significant global healthcare burden. A minority of symptomatic individuals (5-10%) develop severe disease characterised by a vascular leakage syndrome which can then lead to cardiovascular decompensation and death1. Close monitoring of clinical status and vital signs allows for the identification of patients at an increased risk of progression.

引言登革热对全球医疗保健造成重大负担。少数有症状的个体（5-10%）发展为以血管渗漏综合征为特征的严重疾病，然后可能导致心血管失代偿和死亡1。密切监测临床状况和生命体征可以识别进展风险增加的患者。

The interpretation of these clinical features in turn guides delivery of supportive therapies, including fluid resuscitation and organ support measures2. With the growing incidence and expansion of at-risk regions as a result of climate change3, dengue represents a major challenge faced by healthcare systems worldwide.Patients with dengue infection present with a non-specific febrile illness which typically evolve over 3–4 days.

这些临床特征的解释反过来又指导了支持疗法的实施，包括液体复苏和器官支持措施2。随着气候变化导致的高风险地区的发病率和扩张3，登革热是全球医疗保健系统面临的重大挑战。登革热感染患者表现为非特异性发热性疾病，通常在3-4天内发展。

Existing clinical tools, such those published by the WHO, play a crucial role in standardised pre-hospital triage and are commonly used as criteria for hospitalisation. However, such evaluations performed at a single point of encounter are at risk of failing to capture the dynamic nature of this disease4.

现有的临床工具，如世界卫生组织发布的工具，在标准化的院前分流中起着至关重要的作用，通常被用作住院标准。。

Regular, and repeated patient monitoring coupled with contemporaneous assessments allow for better clinical prioritisation, improving patient outcomes as well as healthcare effectiveness5. Within many low- and middle-income countries where dengue is endemic, such in-hospital patient monitoring can prove challenging for a range of reasons, including lack of functional equipment, gaps in training and shortages in healthcare staff6.Non-invasive wearable devices could offer benefits through providing cost-effective individualised monitoring in healthcare7.

定期和重复的患者监测以及同期评估可以更好地确定临床优先级，改善患者预后以及医疗保健效果5。在登革热流行的许多低收入和中等收入国家，由于一系列原因，这种院内患者监测可能具有挑战性，包括缺乏功能设备，培训缺口和医疗保健人员短缺6。非侵入性可穿戴设备可以通过在医疗保健中提供具有成本效益的个性化监测来提供益处7。

Many wearables utilise photoplethysmography (PPG) as the primary sensing modality due to low power requirements and safety8. .

由于低功耗和安全性，许多可穿戴设备使用光体积描记法（PPG）作为主要的传感方式8。

Age 8 or older admitted to hospital, and

8岁或以上入院，以及

Clinical diagnosis of dengue, and

登革热的临床诊断，以及

Hospitalisation less than 28 h at time of enrolment.

入院时住院时间少于28小时。

Exclusion criteria are:

排除标准为：

Lack of informed consent /assent from patient (or from relatives in children), or

缺乏患者（或儿童亲属）的知情同意/同意，或

Clinical reasons at discretion of treating staff.

临床原因由治疗人员自行决定。

Settings for patient recruitment include the Emergency Department, Paediatric and Adult intensive care units and general wards to increase the diversity of presentations. Dengue diagnosis was defined as a clinically compatible febrile syndrome with a positive dengue NS1 assay. These definitions are clinically pragmatic and consistent with guidelines published by the national Ministry of Health in Vietnam.Enrolment and study processesInformed consent was obtained by attending clinicians in accordance with Good Clinical Practice.

招募患者的环境包括急诊科，儿科和成人重症监护病房以及普通病房，以增加演示的多样性。登革热诊断被定义为登革热NS1检测阳性的临床相容性发热综合征。这些定义在临床上是实用的，并且与越南国家卫生部发布的指南一致。登记和研究过程根据良好临床实践，由临床医生获得知情同意。

For children aged 12 or older, assent was obtained in addition to consent from the parent or guardian. Clinical information including demographics, presenting symptoms, examination findings and vital signs were recorded on enrolment. Specifically, blood pressure, pulse rate, respiratory rate, tympanic temperature and oxygen saturation were captured on study case report forms.

对于12岁或12岁以上的儿童，除了获得父母或监护人的同意外，还获得了同意。登记时记录临床信息，包括人口统计学，症状，检查结果和生命体征。具体而言，研究病例报告表中记录了血压，脉搏率，呼吸频率，鼓室温度和血氧饱和度。

Additional administered treatment and vital signs during the 24-h study period performed as part of standard clinical care were recorded; with patients typically undergoing observations every 1–6 h depending on clinical need.PPG data recording was done with a battery-operated wrist wearable containing a transmissive finger probe (SmartCare Analytics, Oxford, UK).

记录了作为标准临床护理的一部分在24小时研究期间进行的额外治疗和生命体征；根据临床需要，患者通常每1-6小时进行一次观察。PPG数据记录是使用电池供电的腕部可穿戴设备完成的，该腕部可穿戴设备包含一个透射式手指探针（SmartCare Analytics，英国牛津）。

This device continuously captured timestamped PPG signals from the patient using red and infra-red wavelengths at 100 Hz, and stored data on the internal memory which was downloaded after end of the recording period. Participants wore the device on their finger for up to 24 h after enrolment. A description of the device and its appearance is presented in supplementary fig.

该设备使用100 Hz的红色和红外波长连续捕获来自患者的带时间戳的PPG信号，并将数据存储在记录期结束后下载的内部存储器上。。该装置及其外观的描述见supplementary fig。

1.Sample sizeThe target sample size for the study was 250, based on pragmatic factors including feasibility of clinical recruitment during one de.

1、样本量该研究的目标样本量为250，基于实用因素，包括在一次de期间临床招募的可行性。

i.

一。

The NEWS2 score22 is based on vital signs measurements and aggregates derangements in six accessible vital signs parameters: respiratory rate, pulse rate, oxygen saturation, systolic blood pressure, level of consciousness and body temperature. There exists a close relationship between NEWS2 score and all-cause mortality across different conditions23..

NEWS2评分22基于生命体征测量，并汇总了六个可获得的生命体征参数的紊乱：呼吸频率，脉搏频率，血氧饱和度，收缩压，意识水平和体温。NEWS2评分与不同条件下的全因死亡率之间存在密切关系23。。

ii.

二。

The dengue-specific modified Sequential Organ Failure Assessment score (mSOFA) was calculated. The score aims to account for limitations in existing approaches to assessing disease severity after hospitalisation. The score has been validated for in a large cohort of dengue patients for our setting and is predictive of clinically-relevant outcomes including requirement for organ support, length of stay and mortality24..

计算登革热特异性改良序贯器官衰竭评估评分（mSOFA）。该评分旨在解释住院后评估疾病严重程度的现有方法的局限性。该评分已在我们的一大批登革热患者中得到验证，并可预测临床相关结果，包括器官支持要求，住院时间和死亡率24。。

A binary outcome was derived based on the physiological derangement in order to identify the patients as low risk of severe disease and do not require immediate intervention. A low clinical risk score was defined as:

基于生理紊乱得出了二元结果，以将患者识别为严重疾病的低风险，不需要立即干预。低临床风险评分定义为：

i.

一。

NEWS2 score less than 6, and

NEWS2得分低于6，并且

ii.

二。

mSOFA score less than 6 and

iii.

三。

the absence of shock or significant bleeding on clinical assessment

临床评估无休克或明显出血

We chose a NEWS2 score less than 6 based on the threshold recommendations from the Royal College of Physicians for urgent clinical intervention22, and the mSOFA score less than 6 to include patients in the lower 25th percentile of risk from the original derivation study24.Data partitioningIn order to better evaluate the generalisation of the model for this task, we partitioned our dataset into the usual train, validation and test datasets.

我们根据皇家医师学院针对紧急临床干预的阈值建议22选择了NEWS2评分小于6分，mSOFA评分小于6分，以包括原始推导研究中风险较低的25%的患者24。数据划分为了更好地评估该任务模型的普遍性，我们将数据集划分为常规的训练，验证和测试数据集。

At the first step, we first randomly partition the data at a 7:1 ratio to create a hold-out test set with a unique set of patients while maintaining the distribution of classes. Similarly, the remaining data is then split through a stratified 5-fold cross validation to create training and validation sets where the patients are unique to each set while maintaining the distribution of samples between the datasets.

在第一步中，我们首先以7:1的比例随机划分数据，以创建一个具有唯一患者集的保持测试集，同时保持类的分布。类似地，剩余的数据然后通过分层的5倍交叉验证进行分割，以创建训练和验证集，其中患者对于每组是唯一的，同时保持数据集之间的样本分布。

The split is implemented using the Stratified K-Fold from the Python scikit-learn package. For each fold, the optimised model was then evaluated against the independent hold out test set to report summary performances. The dataset was randomly split into a training (n = 132) and independent hold out test set (n = 21) with the latter used exclusively for model evaluation.Model features and approachesPPG signals were the primary feature used in the study.

分割是使用Python scikit学习包中的分层K折叠实现的。对于每个折叠，然后针对独立的保持测试集评估优化模型，以报告总结性能。数据集被随机分成训练集（n=132）和独立的保持测试集（n=21），后者专门用于模型评估。模型特征和接近SPG信号是研究中使用的主要特征。

These were collected continuously during the first 24 h of patient admission to hospital and were processed through a rolling window approach. For each individual the feature set consisted of 10-min segments of PPG from start, which were incremented by 1 min until end of the recording period. PPG segments were discarded if the amount of data with corresponding labels was less than the minimum sequence length used by the model (10 min of PPG).

这些是在患者入院的前24小时内连续收集的，并通过滚动窗口方法进行处理。对于每个人，特征集由PPG的10分钟片段组成，从开始到记录期结束增加1分钟。如果具有相应标签的数据量小于模型使用的最小序列长度（PPG的10分钟），则丢弃PPG片段。

For each 10-min PPG segment, a bina.

对于每个10分钟的PPG片段，一个bina。

Data availability

数据可用性

Access to the original PPG and clinical data can be applied through the Scientific and Ethical Committee of the Hospital for Tropical Diseases, Ho Chi Minh City.

可以通过胡志明市热带病医院科学与伦理委员会申请获得原始PPG和临床数据。

Code availability

代码可用性

The code and example notebooks for the data analysis and the optimised model are available at a GitHub repository [hosted at https://github.com/jsmdaniels/VITAL].

数据分析和优化模型的代码和示例笔记本可在GitHub存储库中找到[托管于https://github.com/jsmdaniels/VITAL]。

ReferencesWilder-Smith, A., Ooi, E.-E., Horstick, O., & Wills, B. Dengue. Lancet 393, 350–363 (2019).Article

参考文献维尔德·史密斯，A.，Ooi，E.-E，霍斯蒂克，O.，和威尔斯，B.登革热。《柳叶刀》393350–363（2019）。文章

PubMed

PubMed

Google Scholar

谷歌学者

Yacoub, S. & Wills, B. Predicting outcome from dengue. BMC Med. 12, 147 (2014).Article

Yacoub，S。和Wills，B。预测登革热的结果。BMC Med.12147（2014）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Choisy, M. et al. Climate change and health in Southeast Asia—defining research priorities and the role of the Wellcome Trust Africa Asia Programmes. Wellcome Open Res. 6, 278 (2021).Article

Choisy，M.等人，《东南亚的气候变化与健康》，确定了惠康信托基金亚非项目的研究重点和作用。Wellcome Open Res.6278（2021）。文章

PubMed

PubMed

Google Scholar

谷歌学者

World Health Organization. Dengue Guidelines for Diagnosis, Treatment, Prevention and Control: New Edition. (World Health Organization, 2009).Baker, T. et al. Critical care of tropical disease in low income countries: report from the task Force on tropical diseases by the World Federation of societies of intensive and critical care medicine.

世界卫生组织。登革热诊断、治疗、预防和控制指南：新版。。Baker，T.等人，《低收入国家热带病的重症监护：世界重症监护医学会联合会热带病工作组的报告》。

J. Crit. Care 42, 351–354 (2017).Article .

《动物护理杂志》第42期，第351-354页（2017年）。第条。

PubMed

PubMed

Google Scholar

谷歌学者

Turner, H. C. et al. Achieving affordable critical care in low-income and middle-income countries. BMJ Glob. Health 4, e001675 (2019).Article

Turner，H.C.等人，《在低收入和中等收入国家实现负担得起的重症监护》。BMJ全球。健康4，e001675（2019）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Ming, D. K. et al. Continuous physiological monitoring using wearable technology to inform individual management of infectious diseases, public health and outbreak responses. Int. J. Infect. Dis. 96, 648–654 (2020).Article

Ming，D.K.等人。使用可穿戴技术进行连续生理监测，为传染病、公共卫生和疫情应对的个人管理提供信息。Int.J.感染。Dis。96648-654（2020）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Castaneda, D., Esparza, A., Ghamari, M., Soltanpur, C. & Nazeran, H. A review on wearable photoplethysmography sensors and their potential future applications in health care. Int J. Biosens. Bioelectron. 4, 195–202 (2018).PubMed

Castaneda，D.，Esparza，A.，Ghamari，M.，Soltanpur，C。＆Nazeran，H。可穿戴式光体积描记传感器及其在医疗保健中的潜在未来应用的综述。国际J.Biosens。生物电子。4195-202（2018）。PubMed出版社

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Burn, S. L., Chilton, P. J., Gawande, A. A. & Lilford, R. J. Peri-operative pulse oximetry in low-income countries: a cost–effectiveness analysis. Bull. World Health Organ 92, 858–867 (2014).Article

Burn，S.L.，Chilton，P.J.，Gawande，A.A。＆Lilford，R.J。低收入国家的围手术期脉搏血氧仪：成本效益分析。公牛。世界卫生组织92858-867（2014）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Alboksmaty, A. et al. Effectiveness and safety of pulse oximetry in remote patient monitoring of patients with COVID-19: a systematic review. Lancet Digit. Health 4, e279–e289 (2022).Article

Alboksmaty，A。等人。脉搏血氧仪在COVID-19患者远程患者监测中的有效性和安全性：系统评价。柳叶刀数字。健康4，e279–e289（2022）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Trieu, H. T. et al. The compensatory reserve index predicts recurrent shock in patients with severe dengue. BMC Med. 20, 109 (2022).Article

Trieu，H.T.等人。代偿储备指数预测严重登革热患者反复休克。BMC Med.20109（2022）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Charlton, P. H. et al. Wearable photoplethysmography for cardiovascular monitoring. Proc. IEEE 110, 355–381 (2022).Article

Charlton，P.H。等人。用于心血管监测的可穿戴式光体积描记术。Proc。IEEE 110355–381（2022）。文章

Google Scholar

谷歌学者

Ming, D. K. et al. Applied machine learning for the risk-stratification and clinical decision support of hospitalised patients with dengue in Vietnam. PLOS Digit. Health 1, e0000005 (2022).Article

Ming，D.K.等人将机器学习应用于越南登革热住院患者的风险分层和临床决策支持。PLOS数字。健康1，e0000005（2022）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Ming, D. et al. Connectivity of rapid-testing diagnostics and surveillance of infectious diseases. Bull. World Health Organ. 97, 242–244 (2019).Article

Ming，D.等人。传染病快速检测诊断和监测的连通性。公牛。世界卫生组织。97242-244（2019）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Schultz, M. J. et al. Current challenges in the management of sepsis in ICUs in resource-poor settings and suggestions for the future. Intensiv. Care Med. 43, 612–624 (2017).Article

Schultz，M.J.等人，《资源贫乏环境下ICU脓毒症管理的当前挑战和未来建议》。强度。护理医学43612-624（2017）。文章

Google Scholar

谷歌学者

Worsley-Tonks, K. E. L. et al. Strengthening global health security by improving disease surveillance in remote rural areas of low-income and middle-income countries. Lancet Glob. Health 10, e579–e584 (2022).Article

。柳叶刀球。健康10，e579–e584（2022）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Radin, J. M., Wineinger, N. E., Topol, E. J. & Steinhubl, S. R. Harnessing wearable device data to improve state-level real-time surveillance of influenza-like illness in the USA: a population-based study. Lancet Digit. Health 2, e85–e93 (2020).Article

Radin，J.M.，Wineinger，N.E.，Topol，E.J。＆Steinhubl，S.R。利用可穿戴设备数据改进美国流感样疾病的国家级实时监测：一项基于人群的研究。柳叶刀数字。。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Cass, A. E. G. & Sharma, S. Microneedle enzyme sensor arrays for continuous in vivo monitoring. in Methods in Enzymology Vol. 589, 413–427 (Elsevier, 2017).Karolcik, S., Ming, D. K., Yacoub, S., Holmes, A. H. & Georgiou, P. A multi-site, multi-wavelength PPG platform for continuous non-invasive health monitoring in hospital settings.

Cass，A。E。G。＆Sharma，S。用于连续体内监测的微针酶传感器阵列。在《酶学方法》第589413-427卷（Elsevier，2017）中。Karolcik，S.，Ming，D.K.，Yacoub，S.，Holmes，A.H。＆Georgiou，P。一个多站点，多波长PPG平台，用于在医院环境中进行连续无创健康监测。

IEEE Trans. Biomed. Circuits Syst. 17, 349–361 (2023).Article .

IEEE Trans。生物医学。电路系统。17349-361（2023）。文章。

PubMed

PubMed

Google Scholar

谷歌学者

Chanh, H. Q. et al. Applying artificial intelligence and digital health technologies, Viet Nam. Bull. World Health Organ 101, 487–492 (2023).Article

Chanh，H.Q.等人，《应用人工智能和数字健康技术》，越南。公牛。世界卫生组织101487-492（2023）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

von Elm, E. et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J. Clin. Epidemiol. 61, 344–349 (2008).Article

von Elm，E.等人，《加强流行病学观察性研究报告》（STROBE）声明：观察性研究报告指南。J、 临床。流行病。61344-349（2008）。文章

Google Scholar

谷歌学者

Working Party of the Royal College of Physicians. National Early Warning Score (NEWS) 2 Standardising the Assessment of Acute-Illness Severity in the NHS https://www.rcp.ac.uk/media/ctulqqbn/news2-executive-summary_0.pdf (2017).Scott, L. J. et al. Prognostic value of National Early Warning Scores (NEWS2) and component physiology in hospitalised patients with COVID-19: a multicentre study.

皇家医师学院工作组。国家早期预警评分（NEWS）2标准化了NHS急性疾病严重程度的评估https://www.rcp.ac.uk/media/ctulqqbn/news2-executive-summary_0.pdf（2017年）。Scott，L.J.等人，《国家早期预警评分（NEWS2）和成分生理学在住院新型冠状病毒肺炎患者中的预后价值：一项多中心研究》。

Emerg. Med. J. 39, 589–594 (2022).Article .

《急诊医学杂志》39589-594（2022）。第[UNK]条。

PubMed

PubMed

Google Scholar

谷歌学者

McBride, A. et al. A modified Sequential Organ Failure Assessment score for dengue: development, evaluation and proposal for use in clinical trials. BMC Infect. Dis. 22, 722 (2022).Article

McBride，A.等人，《登革热改良序贯器官衰竭评估评分：临床试验的发展、评估和建议》。。Dis。22722（2022）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Le, V.-K. D. et al. vital_sqi: a Python package for physiological signal quality control. Front. Physiol. 13, 1020458 (2022).Article

Le，V.-K.D.等人。vital\u sqi：用于生理信号质量控制的Python软件包。正面。生理学。。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Lim, B., Arik, S. O., Loeff, N. & Pfister, T. Temporal fusion transformers for interpretable multi-horizon time series forecasting. Int J. Forecast. 37, 1748–1764 (2021).Article

Lim，B.，Arik，S.O.，Loeff，N。＆Pfister，T。用于可解释的多地平线时间序列预测的时间融合变压器。Int J.预测。371748-1764（2021）。文章

Google Scholar

谷歌学者

Alzubaidi, L. et al. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. J. Big Data 8, 53 (2021).Article

Alzubaidi，L.等人，《深度学习回顾：概念，CNN架构，挑战，应用，未来方向》。J、 大数据8,53（2021）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Caruana, R. Multitask Learning. Mach. Learn. 28, 41–75 (1997).Article

Caruana，R。多任务学习。马赫。学习。28，41-75（1997）。文章

Google Scholar

谷歌学者

Download referencesAcknowledgementsWe are grateful to the patients at the Hospital for Tropical Diseases, Ho Chi Minh City, Vietnam and the clinical staff delivering care during the time of the study. We acknowledge the Research Computing Services at Imperial College London for support during the project.

下载参考文献致谢我们感谢越南胡志明市热带病医院的患者以及在研究期间提供护理的临床工作人员。我们感谢伦敦帝国理工学院的研究计算服务在项目期间提供的支持。

The research was funded by the Wellcome Trust (215010/Z/18/Z) and partially funded by the Department of Health and Social Care for Antimicrobial Optimisation, at Imperial College, London. The funders had no role in the design, data collection, analysis or writing of the manuscript.Author informationAuthor notesA list of members and their affiliations appears in the Supplementary Information.Authors and AffiliationsCentre for Antimicrobial Optimisation, Imperial College London, London, UKDamien Keng Ming, Bernard Hernandez, Alison Helen Holmes & Pantelis GeorgiouCentre for Bio-Inspired Technology, Imperial College London, London, UKJohn Daniels, Stefan Karolcik, Vasileios Manginas & Pantelis GeorgiouOxford University Clinical Research Unit, Ho Chi Minh City, VietnamHo Quang Chanh, Quang Huy Nguyen & Sophie YacoubHospital for Tropical Diseases, Ho Chi Minh City, VietnamVan Hao Nguyen, Tu Qui Phan, Thi Hue Tai Luong, Huynh Trung Trieu & Vinh Tho PhanDepartment of Global Health and Infectious Diseases, University of Liverpool, Liverpool, UKAlison Helen HolmesCentre for Tropical Medicine and Global Health, Nuffield Department of Medicine, University of Oxford, Oxford, UKSophie YacoubAuthorsDamien Keng MingView author publicationsYou can also search for this author in.

这项研究由惠康信托基金（215010/Z/18/Z）资助，部分由伦敦帝国理工学院卫生与社会保健抗菌优化部资助。资助者在手稿的设计，数据收集，分析或撰写中没有任何作用。作者信息作者注释补充信息中列出了成员及其所属机构的列表。抗菌优化的作者和附属机构，伦敦帝国理工学院，伦敦，UKDamien Keng Ming，Bernard Hernandez，Alison Helen Holmes＆Pantelis GeorgiouCentre for Bio-Inspired Technology，伦敦帝国理工学院，UKJohn Daniels，Stefan Karolcik，Vasileios Manginas＆Pantelis GeorgiouOxford University Clinical Research Unit，胡志明市，VietnamHo Quang Chanh，Quang Huy Nguyen＆Sophie YacoubHospital for热带病，胡志明市，VietnamVan Hao Nguyen，Tu Quiphan，Thi Tai Luong Huynh Trung Trieu＆Vinh Tho PhanDepartment of Global Health and Infectious Diseases，利物浦大学，UKAlison Helen HolmesCentre for Tropical Medicine and Global Health，Nuffield Department of Medicine，Oxford，Oxford，UKSophie Yacoubathorsdamien Keng MingView作者出版物您也可以在中搜索这位作者。

PubMed Google ScholarJohn DanielsView author publicationsYou can also search for this author in

PubMed谷歌学者John DanielsView作者出版物您也可以在

PubMed Google ScholarHo Quang ChanhView author publicationsYou can also search for this author in

PubMed谷歌学者Ho Quang ChanhView作者出版物您也可以在

PubMed Google ScholarStefan KarolcikView author publicationsYou can also search for this author in

PubMed Google ScholarBernard HernandezView author publicationsYou can also search for this author in

PubMed Google ScholarVasileios ManginasView author publicationsYou can also search for this author in

PubMed Google ScholarVasileios ManginasView作者出版物您也可以在

PubMed Google ScholarVan Hao NguyenView author publicationsYou can also search for this author in

PubMed Google ScholarVan Hao NguyenView作者出版物您也可以在

PubMed Google ScholarQuang Huy NguyenView author publicationsYou can also search for this author in

PubMed Google ScholarQuang Huy NguyenView作者出版物您也可以在

PubMed Google ScholarTu Qui PhanView author publicationsYou can also search for this author in

PubMed Google ScholarTu Qui PhanView作者出版物您也可以在

PubMed Google ScholarThi Hue Tai LuongView author publicationsYou can also search for this author in

PubMed Google ScholarThi Hue Tai LuongView作者出版物您也可以在

PubMed Google ScholarHuynh Trung TrieuView author publicationsYou can also search for this author in

PubMed Google ScholarHuynh Trung TrieuView作者出版物您也可以在

PubMed Google ScholarAlison Helen HolmesView author publicationsYou can also search for this author in

PubMed Google ScholarAlison Helen HolmesView作者出版物您也可以在

PubMed Google ScholarVinh Tho PhanView author publicationsYou can also search for this author in

PubMed Google ScholarVinh Tho PhanView作者出版物您也可以在

PubMed Google ScholarPantelis GeorgiouView author publicationsYou can also search for this author in

PubMed Google ScholarPantelis GeorgiouView作者出版物您也可以在

PubMed Google ScholarSophie YacoubView author publicationsYou can also search for this author in

PubMed Google ScholarSophie YacoubView作者出版物您也可以在

PubMed Google ScholarConsortiaOn behalf of the VITAL consortiumContributionsD.K.M. wrote the first draft of the manuscript. D.K.M., H.Q.C., and J.D. performed the analyses and had access to original data. B.H., S.K., and V.M. developed auxiliary methodology to support model development.

PubMed谷歌学术联盟（Google ScholarConsortiaOn）代表重要财团（VITAL ConsortiumContributions）d。K、 M.写了手稿的初稿。D、 K.M.，H.Q.C.和J.D.进行了分析，并获得了原始数据。B、 H.，S.K。和V.M.开发了辅助方法来支持模型开发。

C.Q.H., V.H.N., Q.H.N., T.Q.P., T.H.T.L., and C.T.P. were involved in the clinical studies, patient recruitment and follow up. A.H.H., P.G. provided expertise throughout the study. All authors (D.K.M., J.D., H.Q.C., S.K., B.H., V.M., V.H.N., Q.H.N., T.Q.P., T.H.T.L., H.T.T., A.H.H., V.T.P., P.G., S.Y.) have read and approved the final draft of the manuscript, and had final responsibility in submission for publication.Corresponding authorCorrespondence to.

C、 Q.H.，V.H.N.，Q.H.N.，T.Q.P.，T.H.T.L.和C.T.P.参与了临床研究，患者招募和随访。A、 H.H.，P.G.在整个研究过程中提供了专业知识。。对应作者对应。

Damien Keng Ming.Ethics declarations

Damien Keng Ming。道德宣言

Competing interests

相互竞争的利益

The authors declare no competing interests.

作者声明没有利益冲突。

Ethical approval

道德认可

The study protocol was approved by the Scientific and Ethical Committee of the Hospital for Tropical Diseases, Ho Chi Minh City (reference CS/BND/20/03) and by the Oxford Tropical Research Ethics Committee (reference 522-20). All study procedures were carried out in accordance with the Declaration of Helsinki..

该研究方案经胡志明市热带病医院科学与伦理委员会（参考文献CS/BND/20/03）和牛津热带研究伦理委员会（参考文献522-20）批准。所有研究程序均按照赫尔辛基宣言进行。。

Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Supplementary informationSupplementary materialRights and permissions

Additional informationPublisher的注释Springer Nature在已发布的地图和机构隶属关系中的管辖权主张方面保持中立。补充信息补充材料权利和权限

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.

开放获取本文是根据知识共享署名4.0国际许可证授权的，该许可证允许以任何媒体或格式使用，共享，改编，分发和复制，只要您对原始作者和来源给予适当的信任，提供知识共享许可证的链接，并指出是否进行了更改。

The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

本文中的图像或其他第三方材料包含在文章的知识共享许可中，除非在材料的信用额度中另有说明。。

To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/..

要查看此许可证的副本，请访问http://creativecommons.org/licenses/by/4.0/..

Reprints and permissionsAbout this articleCite this articleMing, D.K., Daniels, J., Chanh, H.Q. et al. Predicting deterioration in dengue using a low cost wearable for continuous clinical monitoring.

转载和许可本文引用本文Ming，D.K.，Daniels，J.，Chanh，H.Q。等人使用低成本可穿戴设备进行连续临床监测，预测登革热恶化。

npj Digit. Med. 7, 306 (2024). https://doi.org/10.1038/s41746-024-01304-4Download citationReceived: 13 May 2024Accepted: 15 October 2024Published: 02 November 2024DOI: https://doi.org/10.1038/s41746-024-01304-4Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard.

npj数字。医学杂志7306（2024）。https://doi.org/10.1038/s41746-024-01304-4Download引文接收日期：2024年5月13日接受日期：2024年10月15日发布日期：2024年11月2日OI：https://doi.org/10.1038/s41746-024-01304-4Share本文与您共享以下链接的任何人都可以阅读此内容：获取可共享链接对不起，本文目前没有可共享的链接。复制到剪贴板。

Provided by the Springer Nature SharedIt content-sharing initiative

由Springer Nature SharedIt内容共享计划提供

Subjects

主题

Predictive markersRisk factorsViral infection

预测标记风险因素病毒感染