Metastasis prediction based on data mining technique in breast cancer patients

Identifying the metastatic process and its effective factors will help to improve the patient s long-term survival. The purpose of this study was to investigate and identify the factors affecting the prediction ofmetastatic breast cancer using data mining tools. Data mining is a tool for discovery of knowledge from a big data that has been used today in a variety of fields. Diagnosis and prediction of disease in medical science is a growing field of data mining application. In this research, we analyzed 2025 records could be used after data preparation. Then, CHAID algorithm which is a decision tree algorithm were used to discover patterns that predict variables that affect metastasis in a patient. CHAID algorithm were implemented and the level of confidence were calculated, confidence level for the CHAID was 94.24.Based on the results Stage, type of surgery, type of cancer based on pathology and Her2 are the most important predictors of metastasis. The confidence level showed that CHAID is an appropriate algorithms to predict metastasis in breast cancer survivals. In this study we explored patterns that help predict variables that affect metastasis in a patient using the CHAID algorithm by Clementine12software,

کلیدواژه ها:

Breast Cancer ، Cell and Cancer ، Cancer Risks ، Multidisciplinary Cancer Research ، Cancer Treatment and Management