Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

Background: Breast cancer is the most prevalent malignancy in women and there are many possible reasons for its occurrence. One important issue in different kinds of cancers is the spreading of cancerous cells to other tissues (metastasis). Therefore, we have investigated the factors associated with the prediction of metastasis of breast cancer with data mining tools.Methods: This study is a cross-sectional investigation carried out on breast cancer patients (1074 patients) referred to the oncology center of Seyedoshohada Hospital in Isfahan, Iran, between 1999 and 2009. Data mining methods such as association rules and ordinal logistic regression were applied to analyze these data.Findings: Considering positive ER, PR hormonal receptors and positive P53 gene patients, the possibility of metastasis among patients with high lymph node ratio is 4.77, 1.4 and 4.45 times more than patients with low lymph node ratio, respectively. Considering negative ER, PR hormonal receptors and negative P53 gene patients, the possibility of more metastasis among patients with high lymph node ratio increased by 97%. Considering those with a tumor size of more than 2 cm, the possibility of more metastasis among patients with positive ER and P53 gene, is 1.35, 1.27 times higher than those with a tumor size of less than 2 cm. Among those with negative status, the estimated possibility ratio is 2.75. Conclusion: This study has shown the risk factors of metastatic breast cancer. These factors help the identification of subgroups of patients that have a high risk of metastases. Thus the long-term survival and quality of life of patients will be improved by the better understanding of the metastatic process.

Full-Text [PDF 300 kb]   (1036 Downloads)    

Add your comments about this article : Your username or Email:

---