基于机器学习方法的肺癌病人生存时间报告

Survival Time Report of Lung Cancer Patients Based on Machine Learning Method

摘要: 肺癌作为发病率及死亡率极高的恶性肿瘤之一，给病人也给家庭带来了无可弥补的巨大的精神压力和经济压力。本文基于R软件自带的survival-lung数据集，利用传统的线性模型、决策树、随机森林以及岭回归四种模型，研究了影响肺癌患者生存时间的因素，对医学工作者制定有效、合理的医疗方案具有现实意义。研究表明：第一，在相关关系研究中，肺癌患者的生存时间与年龄、心电图表现分、最近六个月减重、膳食中消耗的卡路里、医疗机构数等都具有较强的相关关系；第二，通过性别对比研究生存时间，得出女性肺癌患者生存时间高于男性肺癌生存时间；第三，通过生存时间曲线图得出，肺癌患者的生存时间与生存概率呈负相关关系。最后，根据研究结果提出相应的延长肺癌患者生存时间的建议。

Abstract: As one of the malignant tumors with high incidence rate and mortality, lung cancer has brought irreparable mental and economic pressure to patients and families. Based on the survival-lung data set of R software, this paper studies the factors affecting the survival time of lung cancer patients by using the traditional linear model, decision tree, random forest and ridge regression models, which has practical significance for medical workers to formulate effective and reasonable medical plans. The results show that: First, in the correlation study, the survival time of lung cancer patients has a strong correlation with age, ECG score, wt.loss, calories consumed in diet, and the number of medical institutions; Second, the survival time of female lung cancer patients is higher than that of male lung cancer patients through gender comparison; Thirdly, the survival time curve shows that the survival time of lung cancer patients is negatively correlated with the survival probability. Finally, according to the results of the study, some suggestions were put forward to prolong the survival time of lung cancer patients.