利用迁移学习增强癌症与生育数据中缺失值下的闭经状态预测
机器学习
2020-12-04 v1
摘要
为健康与医疗问题收集充足的带标签训练数据是困难的(Antropova 等,2018)。此外,缺失值在健康与医疗数据集中不可避免,而应对由样本不足与缺失所带来的问题并非易事(Snell 等,2017;Sterne 等,2009)。然而,机器学习算法已在诸多现实医疗问题中取得显著成功,例如回归与分类,这些技术或可作为解决上述问题的途径。
引用
@article{arxiv.2012.01974,
title = {Transfer learning to enhance amenorrhea status prediction in cancer and fertility data with missing values},
author = {Xuetong Wu and Hadi Akbarzadeh Khorshidi and Uwe Aickelin and Zobaida Edib and Michelle Peate},
journal= {arXiv preprint arXiv:2012.01974},
year = {2020}
}
备注
Artificial Intelligence: Applications in Healthcare Delivery, chapter 13