Font Size: a A A

Optimization Of A-LASSO Method And Its Application In Cancer Metastasis Prediction

Posted on:2016-12-11Degree:MasterType:Thesis
Country:ChinaCandidate:C SunFull Text:PDF
GTID:2404330542992395Subject:Applied Statistics
Abstract/Summary:PDF Full Text Request
Purpose of us is to reveal the preoperative lymph node metastasis of colorectal cancer.It is significant for guidance before the operation and reducing the tumor recurrence after surgery.Clinical data for 252 patients with elevated colorectal cancer who underwent radical resection from June 2012 to December 2013 in people's Hospital of Liaoning Province were analyzed retrospectively.The main factors influencing the lymph node metastasis and model parameters were estimated at the same time via Adaptive-Lasso(A-Lasso).A-Lasso results suggest that the abnormal expression in abdominal pain,bloody stools,carcinoembryonic antigen(CEA)and Serum monocyte percentage was statistically significant in colorectal cancer lymphatic metastasis,and the probability of lymph node metastasis of colorectal cancer patients can be analyzed through the model.The result of A-Lasso method can be applied into estimating the possibility of lymph node metastasis for colorectal cancer patients,guidancing the intervention therapy before the operation,helping to improve the cure degree of colorectal surgery,reducing postoperative recurrence,improving the survival rate and survival quality of colorectal cancer patients.
Keywords/Search Tags:Colorectal cancer, Lymph node metastasis, Variable selection, A-Lasso
PDF Full Text Request
Related items