多类病人的结直肠癌联合筛查策略优化

Joint screening strategy optimization for heterogeneous patients with colorectal cancer

针对早期结直肠癌病人,考虑了资源约束下的多类病人联合筛查策略优化问题.建立了该问题的马尔科夫决策模型(MDP)用于求解最优动态筛查策略,提出了基于混合整数规划(MIP)的近似模型求解更大规模的问题,分析了不同资源水平与人群特征下的最优联合筛查策略,并对比了不同情景下各种筛查策略的效果.实验结果表明,联合筛查策略明显优于单独筛查策略,能够帮助提升健康人群的预期质量调整生命年(QALYs)达到1.84%,并且提升幅度随资源的紧张程度、疾病的严重程度、男性和老年人群的比例上升而增大.

This paper investigates optimizing the combined screening strategy with resource constraints.First,the problem is formulated as a Markov decision process(MDP)model that jointly determines the colonoscopy exam decisions for multi-category patients with different risk profiles of colorectal cancer.Then,for computa-tional efficiency,this study proposes an approximate model based on mixed integer programming(MIP),which can efficiently solve large-scale problem instances.The analysis in this paper explores the screening strategies and evaluates their performance under varying resource levels and population characteristics.Numerical results show that the joint screening strategy significantly outperforms the individual screening strategy in terms of the expected quality-adjusted life year improved by 1.84%.Moreover,increasing improvements are observed as the resource limitations become more severe,the disease intensity worsens,and the percentage of men and older patients in the population increases.