TY - GEN
T1 - Related Factors and Risk Prediction of Type 2 Diabetes Complicated with Liver Cancer
AU - Chen, Hui
AU - Xin, Yi
AU - Yang, Yuting
AU - Li, Fei
AU - Cheng, Guoliang
AU - Zhang, Xinxin
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/8
Y1 - 2019/8
N2 - Among liver cancer patients, diabetes is one of the most common complications. This paper uses the Diabetes Dataset of the National Clinical Medical Science Data Center to explore the factors affecting liver cancer in patients with type 2 diabetes mellitus (T2DM) and to establish two control models. Control Model 1 is a comparison of T2DM combined liver cancer patients and T2DM patients without cancer. Control Model 2 is a comparison of T2DM combined liver cancer patients and T2DM combined other cancer patients. Using SPSS19.0 software, logistic regression analysis was used to screen related factors, predictive models were established for risk prediction, and cross-validation was used to evaluate model performance. In control model 1, the eight factors of gender, age, aspartate aminotransferase, direct bilirubin, \gamma-glutamyltransferase, triglyceride, total cholesterol, and high-density lipoprotein cholesterol are independent risks, which have statistical significance in improving the goodness of fit. The average ROC area of the test set of Model 1 was 0.925, the sensitivity was 78.68%, the specificity was 90.12%, and the accuracy rate was 84.50%, which can be used to predict the risk of liver cancer in diabetic patients. In control model 2, the six factors of gender, aspartate aminotransferase, \gamma glutamyltransferase, triglyceride, serum uric acid, and high-density lipoprotein cholesterol are independent risk factors, which have statistical significance in improving the goodness of fit. The test set of Model 2 has an average ROC area of 0.810, a sensitivity of 66.14%, a specificity of 85.54%, and an accuracy rate of 77.20%, which can be used to determine whether a cancer is in the liver if a diabetic patient may have cancer. Combined with the two models, the specific factors for type 2 diabetes complicated with liver cancer are gender, aspartate aminotransferase, gamma glutamic transferase, triglyceride, high-density lipoprotein cholesterol.
AB - Among liver cancer patients, diabetes is one of the most common complications. This paper uses the Diabetes Dataset of the National Clinical Medical Science Data Center to explore the factors affecting liver cancer in patients with type 2 diabetes mellitus (T2DM) and to establish two control models. Control Model 1 is a comparison of T2DM combined liver cancer patients and T2DM patients without cancer. Control Model 2 is a comparison of T2DM combined liver cancer patients and T2DM combined other cancer patients. Using SPSS19.0 software, logistic regression analysis was used to screen related factors, predictive models were established for risk prediction, and cross-validation was used to evaluate model performance. In control model 1, the eight factors of gender, age, aspartate aminotransferase, direct bilirubin, \gamma-glutamyltransferase, triglyceride, total cholesterol, and high-density lipoprotein cholesterol are independent risks, which have statistical significance in improving the goodness of fit. The average ROC area of the test set of Model 1 was 0.925, the sensitivity was 78.68%, the specificity was 90.12%, and the accuracy rate was 84.50%, which can be used to predict the risk of liver cancer in diabetic patients. In control model 2, the six factors of gender, aspartate aminotransferase, \gamma glutamyltransferase, triglyceride, serum uric acid, and high-density lipoprotein cholesterol are independent risk factors, which have statistical significance in improving the goodness of fit. The test set of Model 2 has an average ROC area of 0.810, a sensitivity of 66.14%, a specificity of 85.54%, and an accuracy rate of 77.20%, which can be used to determine whether a cancer is in the liver if a diabetic patient may have cancer. Combined with the two models, the specific factors for type 2 diabetes complicated with liver cancer are gender, aspartate aminotransferase, gamma glutamic transferase, triglyceride, high-density lipoprotein cholesterol.
KW - Data mining
KW - Liver cancer
KW - Logistic regression model
KW - Risk prediction
KW - Type 2 diabetes
UR - http://www.scopus.com/inward/record.url?scp=85072409254&partnerID=8YFLogxK
U2 - 10.1109/ICMA.2019.8816301
DO - 10.1109/ICMA.2019.8816301
M3 - Conference contribution
AN - SCOPUS:85072409254
T3 - Proceedings of 2019 IEEE International Conference on Mechatronics and Automation, ICMA 2019
SP - 2138
EP - 2143
BT - Proceedings of 2019 IEEE International Conference on Mechatronics and Automation, ICMA 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 16th IEEE International Conference on Mechatronics and Automation, ICMA 2019
Y2 - 4 August 2019 through 7 August 2019
ER -
