计算机应用 ›› 2020, Vol. 40 ›› Issue (5): 1284-1290.DOI: 10.11772/j.issn.1001-9081.2019091673

• 人工智能 • 上一篇    下一篇

动态事件时间数据的多任务Logistic生存预测方法

阮灿华, 林甲祥   

  1. 福建农林大学 计算机与信息学院,福州 350002
  • 收稿日期:2019-10-08 修回日期:2019-12-25 出版日期:2020-05-10 发布日期:2020-05-15
  • 通讯作者: 阮灿华(1980—)
  • 作者简介:阮灿华(1980—),男,福建福安人,实验师,硕士,主要研究方向:生存数据挖掘、智能护理、计算医学、网络安全; 林甲祥(1982—),男,福建泉州人,讲师,博士,主要研究方向:生存数据挖掘、智能护理、计算医学、人工智能。
  • 基金资助:

    福建省自然科学基金面上项目(2018J01644);福建农林大学科技创新专项基金资助项目(CXZX2018033)。

Multi-task Logistic survival prediction method for time-dependent time-to-event data

RUAN Canhua, LIN Jiaxiang   

  1. College of Computer and Information Sciences, Fujian Agriculture and Forestry University, Fuzhou Fujian 350002, China
  • Received:2019-10-08 Revised:2019-12-25 Online:2020-05-10 Published:2020-05-15
  • Contact: RUAN Canhua, born in 1980, M. S., experimentalist. His research interests include survival data mining, intelligent nursing, computational medicine, cybersecurity.
  • About author:RUAN Canhua, born in 1980, M. S., experimentalist. His research interests include survival data mining, intelligent nursing, computational medicine, cybersecurity.LIN Jiaxiang, born in 1982, Ph. D., lecturer. His research interests include survival data mining, intelligent nursing, computational medicine, artificial intelligence.
  • Supported by:

    This work is partially supported by the Surface Program of National Natural Science Foundation of Fujian Province (2018J01644), the Special Fund of Innovative Science and Technology of Fujian Agriculture and Forestry University (CXZX2018033).

摘要:

事件时间数据广泛存在于临床医学研究领域,包含大量复杂的随时间变化的动态风险因子变量。为了对这些动态事件时间数据进行有效分析,克服生存模型参数假设的局限性,提出了一种多任务Logistic生存学习和预测方法。将生存预测转化为一系列不同时间点的多任务二元生存分类问题,利用动态风险因子变量的全部观测值估计累积风险。通过对事件样本和删失样本的全数据学习正则化Logistic回归参数。评估风险因子与事件时间的动态关系,根据生存概率估计事件时间。在多个实际临床数据集上开展的对比实验验证了提出的多任务预测方法对于动态数据不仅具有较强的适用性,而且能够保障预测结果的准确性和可靠性

关键词: 多任务学习, 生存预测, Logistic回归, 事件时间数据, 风险因子变量, 删失

Abstract:

Time-to-event data are ubiquitous in clinical medicine research domain, and include a large number of time-dependent time-dependent risk factor variables. To effectively analyze the time-dependent time-to-event data and to overcome the limitation of parameter hypothesis of the survival model, a multi-task Logistic survival leaning and prediction method was proposed. The survival prediction was transformed into a series of multi-task binary survival classification problems at various time points, and all observations of time-dependent risk factors were used to estimate the cumulative risk. By learning all data of event samples and censored samples, the Logistic regression parameters were regularized. The time-dependent relationships between risk factors and time-to-event were evaluated, and the time-to-event was estimated according to the survival probability. The comparative experiments on multiple real clinical datasets demonstrate the applicability of the proposed multi-task prediction method for time-dependent data and that the method can guarantee the accuracy and reliability of the prediction results.

Key words: multi-task learning, survival prediction, Logistic regression, time-to-event data, risk factor variable, censoring

中图分类号: