首页|基于机器学习的卵巢恶性肿瘤预测模型

基于机器学习的卵巢恶性肿瘤预测模型

扫码查看
目前临床使用的检验项目多达1000余项,采用逻辑回归二分类和支持向量机分别构建卵巢恶性肿瘤预测模型,探讨检验项 目与诊断结果的相关性.实验结果表明,2种机器学习算法构建的卵巢恶性肿瘤预测模型具有较高的预测水平,红细胞体积分布宽度和平均血小板体积等非特异性检验项目与卵巢恶性肿瘤诊断结果具有较强的相关性.
Prediction Model of Malignant Ovarian Tumor Based on Machine Learning
There are more than 1000 inspection items currently used in clinical practice.The real laboratory medicine data are used to construct the malignant ovarian tumor prediction model through logistic regression and support vector machine,and to explore the correlation between the inspection items and the diagnosis results.The experimental results show that the malig-nant ovarian tumor prediction model constructed by the two machine learning algorithms has high prediction levels,and non-specific inspection items such as red blood cell distribution width and mean platelet volume have strong correlation with the di-agnosis results of malignant ovarian tumors.

machine learninglogical regression two classificationsupport vector machinemalignant ovarian tumorlaborato-ry medicine

王莹、顾大勇

展开 >

深圳市第二人民医院,检验科,广东,深圳 518000

机器学习 逻辑回归二分类 支持向量机 卵巢恶性肿瘤 医学检验

深圳市科技计划

ZDSYS20210623092001003

2024

微型电脑应用
上海市微型电脑应用学会

微型电脑应用

CSTPCD
影响因子:0.359
ISSN:1007-757X
年,卷(期):2024.40(4)
  • 13