ImportanceToxic effects of concurrent chemoradiotherapy (CRT) can cause treatment interruptions and hospitalizations, reducing treatment efficacy and increasing health care costs. Physical activity monitoring may enable early identification of patients at high risk for hospitalization who may benefit from proactive intervention.ObjectiveTo develop and validate machine learning (ML) approaches based on daily step counts collected by wearable devices on prospective trials to predict hospitalizations during CRT.Design, Setting, and ParticipantsThis study included patients with a variety of cancers enrolled from June 2015 to August 2018 on 3 prospective, single-institution trials of activity monitoring using wearable devices during CRT. Patients were followed up during and 1 month following CRT. Training and validation cohorts were generated temporally, stratifying for cancer diagnosis (70:30). Random forest, neural network, and elastic net–regularized logistic regression (EN) were trained to predict short-term hospitalization risk based on a combination of clinical characteristics and the preceding 2 weeks of activity data. To predict outcomes of activity data, models based only on activity-monitoring features and only on clinical features were trained and evaluated. Data analysis was completed from January 2022 to March 2023.Main Outcomes and MeasuresModel performance was evaluated in terms of the receiver operating characteristic area under curve (ROC AUC) in the stratified temporal validation cohort.ResultsStep counts from 214 patients (median [range] age, 61 [53-68] years; 113 [52.8%] male) were included. EN based on step counts and clinical features had high predictive ability (ROC AUC, 0.83; 95% CI, 0.66-0.92), outperforming random forest (ROC AUC, 0.76; 95% CI, 0.56-0.87; P = .02) and neural network (ROC AUC, 0.80; 95% CI, 0.71-0.88; P = .36). In an ablation study, the EN model based on only step counts demonstrated greater predictive ability than the EN model with step counts and clinical features (ROC AUC, 0.85; 95% CI, 0.70-0.93; P = .09). Both models outperformed the EN model trained on only clinical features (ROC AUC, 0.53; 95% CI, 0.31-0.66; P < .001).Conclusions and RelevanceThis study developed and validated a ML model based on activity-monitoring data collected during prospective clinical trials. Patient-generated health data have the potential to advance predictive ability of ML approaches. The resulting model from this study will be evaluated in an upcoming multi-institutional, cooperative group randomized trial.

中文翻译：

---

基于机器学习的每日步数预测放化疗期间的住院情况


重要性同步放化疗 (CRT) 的毒性作用可能导致治疗中断和住院治疗，降低治疗效果并增加医疗费用。体力活动监测可以尽早识别住院高风险患者，这些患者可能会从主动干预中受益。目标根据可穿戴设备在前瞻性试验中收集的每日步数来开发和验证机器学习 (ML) 方法，以预测 CRT 期间的住院情况。设计、背景和参与者这项研究包括 2015 年 6 月至 2018 年 8 月期间入组的患有多种癌症的患者，参加了 3 项在 CRT 期间使用可穿戴设备进行活动监测的前瞻性、单机构试验。在 CRT 期间和术后 1 个月对患者进行随访。训练和验证队列是临时生成的，对癌症诊断进行分层 (70:30)。训练随机森林、神经网络和弹性网络正则逻辑回归 (EN)，根据临床特征和前 2 周的活动数据预测短期住院风险。为了预测活动数据的结果，训练和评估了仅基于活动监测特征和临床特征的模型。数据分析于 2022 年 1 月至 2023 年 3 月完成。主要结果和措施根据分层时间验证队列中的受试者工作特征曲线下面积 (ROC AUC) 评估模型性能。结果来自 214 名患者的步数计数（中位[范围]年龄） ，61 [53-68] 岁；113 [52.8%] 男性）。基于步数和临床特征的 EN 具有较高的预测能力（ROC AUC，0.83；95% CI，0.66-0.92），优于随机森林（ROC AUC，0.76；95% CI，0.56-0.87；磷= .02）和神经网络（ROC AUC，0.80；95% CI，0.71-0.88；磷= .36)。在一项消融研究中，仅基于步数的 EN 模型表现出比具有步数和临床特征的 EN 模型更高的预测能力（ROC AUC，0.85；95% CI，0.70-0.93；磷= .09）。两种模型均优于仅针对临床特征训练的 EN 模型（ROC AUC，0.53；95% CI，0.31-0.66；磷< .001).结论和相关性本研究根据前瞻性临床试验期间收集的活动监测数据开发并验证了 ML 模型。患者生成的健康数据有可能提高机器学习方法的预测能力。这项研究得出的模型将在即将进行的多机构合作组随机试验中进行评估。