Development of a Mobile Health Personalized Physiologic Analytics Tool for Pediatric Patients with Sepsis

为脓毒症儿科患者开发移动健康个性化生理分析工具

• 批准号: 10878034
• 负责人: Stephanie Chow Garbern
• 金额: $ 16.38万
• 依托单位: RHODE ISLAND HOSPITAL
• 依托单位国家: 美国
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-08-10 至 2025-04-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10878034
• 关键词: Affect; Area; Award; Bolus Infusion; Cellular Phone; Cessation of life; Child; Childhood; Clinical; Critical Care; Dangerousness; Data; Development; Diagnostic; Diarrhea; Disease; Equipment; Geographic Locations; Healthcare Systems; IV Fluid; Immune response; Infection; Laboratories; Life; Liquid substance; Malaria; Malnutrition; Monitor; Multiple Organ Failure; Normal saline; Organ failure; Parents; Patient-Focused Outcomes; Physiological; Pneumonia; Population; Practice Management; Research; Resource-limited setting; Resources; Sepsis; Severities; Subgroup; Technology; Time; Veins; acute infection; analytical tool; antimicrobial drug; improved outcome; individualized medicine; intravenous administration; low and middle-income countries; low income country; mHealth; machine learning model; mortality; mortality risk; novel; nutrition; pediatric patients; pediatric sepsis; predictive modeling; prevent; septic patients; temporal measurement; tool; transmission process; treatment response; treatment strategy; wearable device

PROJECT SUMMARY Sepsis is a life-threatening condition caused by an imbalanced immune response to infection, leading to organ failure and death if untreated. Low- and middle-income countries (LMICs) bear a disproportionate burden of childhood deaths due to sepsis resulting from acute infections like pneumonia, diarrhea, and malaria. Timely recognition and treatment of sepsis in children are crucial for altering the disease course and preventing mortality. There is immense potential for novel clinical tools that can expedite the identification of severe sepsis, multiple organ dysfunction syndrome (MODS), and predict the highest mortality risk. These tools can also effectively monitor treatment responses, decreasing the need for expensive laboratory diagnostics and critical care monitoring equipment used in high-resource settings. While early initiation of antimicrobial drugs is well-established in sepsis treatment, the guidance for administration of intravenous fluids (IVF) for treatment of severe sepsis can vary depending on the geographical region and even nutrition status. The 2020 Surviving Sepsis Campaign advises that children who reside in areas with resource-limited healthcare systems receive lower initial fluid volumes based for sepsis treatment based on a study with narrow generalizability that demonstrated increased mortality rates among pediatric patients who received normal saline boluses for severe sepsis. Children with malnutrition and severe sepsis are also fluid restricted. This project aims to leverage predictive modeling and mobile health (mHealth) tools to evaluate the impact of IVF administration on sepsis severity in pediatric patients. By utilizing machine learning models developed under the parent award, real-time analysis of IVF effects on sepsis severity can be performed in LMIC settings. Utilizing continuous vital sign parameters transmitted from a wearable device to a mobile phone, real-time measurement of IVF impact on sepsis severity can be determined. This research can shed light on the benefits or harms of fluid administration in LMIC children with severe sepsis, including specific populations like malnourished children. Given the limited studies on appropriate fluid volumes for this subgroup, this research offers an opportunity to determine safe and optimal fluid administration practices. Through this research, we aim to contribute to the understanding of IVF administration effects on sepsis severity in children, particularly in LMIC settings. The findings can guide fluid management practices, inform tailored treatment strategies for malnourished patients, and improve outcomes for pediatric sepsis patients worldwide.

项目摘要 败血症是一种威胁生命的疾病，由对感染的免疫反应不平衡，导致器官 失败和死亡，如果不加以治疗。低收入和中等收入国家（LMIC）承担着不成比例的负担 由于肺炎，腹泻和疟疾等急性感染引起的败血症而导致的童年死亡。及时 儿童败血症的识别和治疗对于改变疾病病程并防止 死亡。新型临床工具具有巨大的潜力，可以加快严重的识别 败血症，多器官功能障碍综合征（MOD），并预测死亡率最高的风险。这些工具可以 还有效地监测治疗反应，减少对昂贵的实验室诊断的需求和 高资源设置中使用的重症监护监测设备。 虽然早期开始抗菌药物在败血症治疗中已经建立了良好的 给予静脉输液（IVF）治疗严重败血症可能会因 地理区域甚至营养状况。 2020年幸存的败血症运动建议的孩子 居住在具有资源有限的医疗保健系统的地区，基于败血症的初始流体量较低 基于一项狭窄概括的研究的治疗，表明死亡率提高 接受严重败血症的正常盐水大剂量的小儿患者。营养不良和严重的孩子 败血症也受到限制。 该项目旨在利用预测性建模和移动健康（MHealth）工具来评估 小儿患者的脓毒症严重程度给药。通过利用机器学习模型开发 根据父母奖，可以在LMIC环境中对IVF影响对败血症严重程度的影响实时分析。 利用从可穿戴设备传输到手机的连续生命体征参数，实时 可以确定IVF对败血症严重程度的影响。这项研究可以阐明好处 或在患有严重败血症的LMIC儿童中流体给药的伤害，包括特定人群 营养不良的孩子。考虑到该亚组的适当流体量的有限研究，这项研究 提供了确定安全和最佳流体给药实践的机会。 通过这项研究，我们旨在为理解IVF给药的影响做出贡献 儿童的严重程度，特别是在LMIC环境中。这些发现可以指导流体管理实践，告知 营养不良的患者量身定制的治疗策略，并改善儿科败血症患者的预后 全世界。