PDF(Pubmed)
Abstract：
BACKGROUND: The current guidelines for managing screen-detected pulmonary nodules offer rule-based recommendations for immediate diagnostic work-up or follow-up at intervals of 3, 6, or 12 months. Customized visit plans are lacking.
OBJECTIVE: To develop individualized screening schedules using reinforcement learning (RL) and evaluate the effectiveness of RL-based policy models.
METHODS: Using a nested case-control design, we retrospectively identified 308 patients with cancer who had positive screening results in at least two screening rounds in the National Lung Screening Trial. We established a control group that included cancer-free patients with nodules, matched (1:1) according to the year of cancer diagnosis. By generating 10,164 sequence decision episodes, we trained RL-based policy models, incorporating nodule diameter alone, combined with nodule appearance (attenuation and margin) and/or patient information (age, sex, smoking status, pack-years, and family history). We calculated rates of misdiagnosis, missed diagnosis, and delayed diagnosis, and compared the performance of RL-based policy models with rule-based follow-up protocols (National Comprehensive Cancer Network guideline; China Guideline for the Screening and Early Detection of Lung Cancer).
RESULTS: We identified significant interactions between certain variables (e.g., nodule shape and patient smoking pack-years, beyond those considered in guideline protocols) and the selection of follow-up testing intervals, thereby impacting the quality of the decision sequence. In validation, one RL-based policy model achieved rates of 12.3% for misdiagnosis, 9.7% for missed diagnosis, and 11.7% for delayed diagnosis. Compared with the two rule-based protocols, the three best-performing RL-based policy models consistently demonstrated optimal performance for specific patient subgroups based on disease characteristics (benign or malignant), nodule phenotypes (size, shape, and attenuation), and individual attributes.
CONCLUSIONS: This study highlights the potential of using an RL-based approach that is both clinically interpretable and performance-robust to develop personalized lung cancer screening schedules. Our findings present opportunities for enhancing the current cancer screening system.
OBJECTIVE: To develop individualized screening schedules using reinforcement learning (RL) and evaluate the effectiveness of RL-based policy models.
METHODS: Using a nested case-control design, we retrospectively identified 308 patients with cancer who had positive screening results in at least two screening rounds in the National Lung Screening Trial. We established a control group that included cancer-free patients with nodules, matched (1:1) according to the year of cancer diagnosis. By generating 10,164 sequence decision episodes, we trained RL-based policy models, incorporating nodule diameter alone, combined with nodule appearance (attenuation and margin) and/or patient information (age, sex, smoking status, pack-years, and family history). We calculated rates of misdiagnosis, missed diagnosis, and delayed diagnosis, and compared the performance of RL-based policy models with rule-based follow-up protocols (National Comprehensive Cancer Network guideline; China Guideline for the Screening and Early Detection of Lung Cancer).
RESULTS: We identified significant interactions between certain variables (e.g., nodule shape and patient smoking pack-years, beyond those considered in guideline protocols) and the selection of follow-up testing intervals, thereby impacting the quality of the decision sequence. In validation, one RL-based policy model achieved rates of 12.3% for misdiagnosis, 9.7% for missed diagnosis, and 11.7% for delayed diagnosis. Compared with the two rule-based protocols, the three best-performing RL-based policy models consistently demonstrated optimal performance for specific patient subgroups based on disease characteristics (benign or malignant), nodule phenotypes (size, shape, and attenuation), and individual attributes.
CONCLUSIONS: This study highlights the potential of using an RL-based approach that is both clinically interpretable and performance-robust to develop personalized lung cancer screening schedules. Our findings present opportunities for enhancing the current cancer screening system.
摘要：
背景：当前的管理屏幕检测到的肺结节的指南提供了基于规则的建议，以立即进行诊断检查或每隔3、6或12个月进行随访。缺乏定制的访问计划。
目的：使用强化学习（RL）制定个性化的筛查计划，并评估基于RL的政策模型的有效性。
方法：使用嵌套的案例控制设计，我们回顾性地确定了308例癌症患者,这些患者在国家肺癌筛查试验的至少两轮筛查中筛查结果为阳性.我们建立了一个对照组，包括没有癌症的结节患者，根据癌症诊断年份匹配（1：1）。通过生成10,164个序列决策事件，我们训练了基于RL的策略模型，仅包含结节直径，结合结节外观（衰减和边缘）和/或患者信息（年龄，性别,吸烟状况,包年，和家族史）。我们计算了误诊率，漏诊,和延迟诊断，并比较了基于RL的政策模型和基于规则的随访协议（国家综合癌症网络指南；中国肺癌筛查和早期检测指南）的性能。
结果：我们确定了某些变量之间的显着相互作用（例如，结节形状和患者吸烟包年，超出指南协议中考虑的范围)和后续测试间隔的选择，从而影响决策序列的质量。在验证中，一个基于RL的政策模型的误诊率为12.3%，9.7%为漏诊，延迟诊断为11.7%。与两种基于规则的协议相比，三个性能最佳的基于RL的策略模型一致地证明了基于疾病特征（良性或恶性）的特定患者亚组的最佳性能，结节表型(大小，形状,和衰减)，和个人属性。
结论：这项研究强调了使用基于RL的方法的潜力，该方法在临床上可解释且性能稳健，以开发个性化的肺癌筛查时间表。我们的发现为增强当前的癌症筛查系统提供了机会。
目的：使用强化学习（RL）制定个性化的筛查计划，并评估基于RL的政策模型的有效性。
方法：使用嵌套的案例控制设计，我们回顾性地确定了308例癌症患者,这些患者在国家肺癌筛查试验的至少两轮筛查中筛查结果为阳性.我们建立了一个对照组，包括没有癌症的结节患者，根据癌症诊断年份匹配（1：1）。通过生成10,164个序列决策事件，我们训练了基于RL的策略模型，仅包含结节直径，结合结节外观（衰减和边缘）和/或患者信息（年龄，性别,吸烟状况,包年，和家族史）。我们计算了误诊率，漏诊,和延迟诊断，并比较了基于RL的政策模型和基于规则的随访协议（国家综合癌症网络指南；中国肺癌筛查和早期检测指南）的性能。
结果：我们确定了某些变量之间的显着相互作用（例如，结节形状和患者吸烟包年，超出指南协议中考虑的范围)和后续测试间隔的选择，从而影响决策序列的质量。在验证中，一个基于RL的政策模型的误诊率为12.3%，9.7%为漏诊，延迟诊断为11.7%。与两种基于规则的协议相比，三个性能最佳的基于RL的策略模型一致地证明了基于疾病特征（良性或恶性）的特定患者亚组的最佳性能，结节表型(大小，形状,和衰减)，和个人属性。
结论：这项研究强调了使用基于RL的方法的潜力，该方法在临床上可解释且性能稳健，以开发个性化的肺癌筛查时间表。我们的发现为增强当前的癌症筛查系统提供了机会。
