Ehtemam Houriyeh, Montazeri Mitra, Khajouei Reza, Hosseini Raziyeh, Nemati Ali, Maazed Vahid
Health Services Management Research Center, Institute for Futures Studies in Health, Kerman University of Medical Sciences, Kerman, Iran.
Modeling in Health Research Center, Institute for Futures Studies in Health, Kerman University of Medical Sciences, Kerman, Iran.
Iran J Public Health. 2017 Nov;46(11):1563-1571.
Breast cancer is one of the most common cancers with a high mortality rate among women. Prognosis and early diagnosis of breast cancer among women society reduce considerable rate of their mortality. Nowadays, due to this illness, try to be setting up intelligent systems, which can predict and early diagnose this cancer, and reduce mortality of women society.
Overall, 208 samples were collected from 2014 to 2015 from two oncologist offices and Javadalaemeh Clinic in Kerman, southeastern Iran. Data source was medical records of patients, then 64 data mining models in MATLAB and WEKA software were used, eventually these measured precision and accuracy of data mining models.
Among 64 data mining models, Bayes-Net model had 95.67% of accuracy and 95.70% of precision; therefore, was introduced as the best model for prognosis and diagnosis of breast cancer.
Intelligent and reliable data mining models are proposed. Hence, these models are recommended as a useful tool for breast cancer prediction as well as medical decision-making.
乳腺癌是女性中最常见的癌症之一，死亡率很高。女性群体中乳腺癌的预后和早期诊断可显著降低其死亡率。如今，针对这种疾病，人们试图建立智能系统，以预测和早期诊断这种癌症，并降低女性群体的死亡率。
总体而言，2014年至2015年期间，从伊朗东南部克尔曼的两个肿瘤学家办公室和贾瓦达拉梅诊所收集了208个样本。数据来源为患者的病历，然后在MATLAB和WEKA软件中使用了64种数据挖掘模型，最终这些模型测量了数据挖掘模型的精度和准确性。
在64种数据挖掘模型中，贝叶斯网络模型的准确率为95.67%，精确率为95.70%；因此，该模型被推荐为乳腺癌预后和诊断的最佳模型。
提出了智能且可靠的数据挖掘模型。因此，推荐这些模型作为乳腺癌预测以及医疗决策的有用工具。
