Multi-institutional validation of a multi-modal machine learning algorithm to predict and reduce acute care during cancer therapy

对多模式机器学习算法进行多机构验证，以预测和减少癌症治疗期间的急性护理

• 批准号: 10587221
• 负责人: Julian Clint Hong
• 金额: $ 38.91万
• 依托单位: UNIVERSITY OF CALIFORNIA, SAN FRANCISCO
• 依托单位国家: 美国
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-03-01 至 2028-02-29
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10587221
• 关键词: Accident and Emergency department; Address; Affect; Apple; Artificial Intelligence; Asian; Black race; Boston; COVID-19 pandemic; California; Caring; Chemotherapy and/or radiation; Clinical; Clinical Data; Clinical Trials; Community Practice; Complement; Computer software; Data; Data Sources; Devices; Disease; Early identification; Ecosystem; Electronic Health Record; Emergency department visit; Ensure; Equity; Evaluation; Future; Geography; Goals; Health; Healthcare; Healthcare Systems; Hospital Departments; Hospitalization; Hospitals; Informatics; Institution; Intervention; Israel; Machine Learning; Malignant Neoplasms; Medical center; Medicare; Modeling; Monitor; Native Americans; Non-Small-Cell Lung Carcinoma; Oncology; Outcome; Outpatients; Patient Care; Patients; Performance; Procedures; Public Health; Quality of life; Race; Radiation therapy; Randomized; Randomized, Controlled Trials; Research; Resources; San Francisco; Supportive care; System; Systemic Therapy; Treatment outcome; Underserved Population; United States; United States Centers for Medicare and Medicaid Services; Universities; Validation; Washington; Work; acute care; cancer care; cancer therapy; chemoradiation; chemotherapy; clinical practice; cohort; comorbidity; cost; data harmonization; fitbit; handheld mobile device; health care delivery; health care settings; health data; health equality; high risk; improved; machine learning algorithm; machine learning model; machine learning prediction; machine learning prediction algorithm; multimodality; patient population; personalized predictions; portability; practice setting; predictive modeling; prospective; randomized, controlled study; remote health care; rural Americans; socioeconomics; success; symptom management; symptom treatment; symptomatic improvement; tool; wearable device

PROJECT ABSTRACT An estimated 650,000 patients with cancer receive systemic therapy or radiation therapy (RT) annually in the United States. Many of these patients undergoing outpatient cancer therapy will require acute care with an emergency department visit or hospital admission due to symptoms from treatment, disease, or comorbidities. This can impact cancer outcomes, patient treatment decisions, and costs to patients and the healthcare system. While there has been much enthusiasm for artificial intelligence and machine learning (ML) to improve healthcare delivery, high quality prospective data are lacking, especially across diverse clinical practice settings. We previously completed one of the first randomized controlled studies in healthcare ML, demonstrating that ML based on EHR data can accurately generate personalized predictions and guide supportive interventions to decrease acute care requirements and costs in patients undergoing RT and chemoradiotherapy (CRT) (NCT04277650). We have also developed a ML model for predicting hospitalizations based on prospective clinical trials of daily step counts collected in patients undergoing CRT. The research objective of this application is to leverage a geographically, racially, socioeconomically, and technically diverse network of healthcare settings and patients to assess and maximize how accurately and equitably these approaches generalize. Our team includes the University of California, San Francisco (UCSF), Duke University, Beth Israel Deaconess Medical Center, Essentia Health in Duluth, MN and Ashland, WI, Washington Hospital in Fremont, CA, Duke Regional Hospital in Durham, NC, and Duke Raleigh Hospital in Raleigh, NC. Specifically, we seek to: (1) prospectively evaluate the validity of an EHR-based acute care prediction ML algorithm across our network and establish a framework for equity, generalizability, and portability and (2) validate our existing patient-generated health data (PGHD; step count) models that predict hospitalization during CRT at a second institution and integrate with our EHR-based ML algorithm to enhance prediction of acute care needs. We hypothesize that our approaches will be accurate across institutions though require adjustments for both generalizability and fairness, and that EHR- and PGHD-based approaches will offer complementary predictive performance. The long-term goal is to develop informatics-based tools that can be broadly and equitably deployed to improve the delivery of cancer care and subsequent treatment outcomes. This research will generate data regarding the generalizability and fairness of EHR- and PGHD-based approaches and a platform for a future multi-institutional randomized controlled trial.

项目摘要 估计每年有650,000名癌症患者接受全身治疗或放射治疗（RT） 美国。这些患者中有许多接受门诊癌症治疗将需要急性护理 由于治疗，疾病或合并症的症状，急诊室就诊或住院。 这可能会影响癌症的结果，患者治疗决策以及对患者和医疗保健的成本 系统。尽管人工智能和机器学习（ML）有很多热情，以改善 缺乏医疗保健服务，高质量的前瞻性数据，尤其是在各种临床实践中 设置。 我们以前完成了医疗保健ML中最早的随机对照研究之一，表明 基于EHR数据的ML可以准确地产生个性化的预测，并指导支持性干预措施 减少接受RT和化学放疗（CRT）患者的急性护理要求和成本 （NCT04277650）。我们还开发了一种ML模型，用于预测基于潜在的住院 接受CRT患者收集的每日步骤计数的临床试验。研究目标 应用是利用地理，种族，社会经济和技术多样化的网络的应用 医疗保健环境和患者，以评估和最大化这些方法的准确和公平程度 概括。我们的团队包括加利福尼亚大学旧金山大学（UCSF），杜克大学，贝丝 明尼苏达州德卢斯市的Essentia Health的执事医疗中心和威斯康星州的Ashland，华盛顿州弗里蒙特的华盛顿医院 CA，北卡罗来纳州达勒姆市的杜克大学地区医院和北卡罗来纳州罗利的杜克·罗利医院。具体来说，我们寻求 至：（1）前瞻性评估基于EHR的急性护理预测ML算法的有效性 网络并建立一个公平性，可推广性和可移植性的框架，以及（2）验证我们现有的 患者生成的健康数据（PGHD;步骤计数）模型，这些模型可预测CRT期间一秒钟的住院 机构并与基于EHR的ML算法集成，以增强急性护理需求的预测。我们 假设我们的方法在整个机构中都将是准确的 概括性和公平性，基于EHR和PGHD的方法将提供互补的预测 表现。 长期目标是开发基于信息学的工具，这些工具可以广泛，公平地部署到 改善癌症护理的分娩和随后的治疗结果。这项研究将生成数据 关于基于EHR和PGHD的方法的普遍性和公平性以及未来的平台 多机构的随机对照试验。