Stepwise logistic regression analysis of tumor contour features for breast ultrasound diagnosis
Huihua Kenny Chiang, Chui‐Mei Tiu, Guo-Shian Hung, Shiao‐Chi Wu, T.Y. Chang, Yi‐Hong Chou · 2002
To assist the ultrasound diagnosis of solid breast tumors by using stepwise logistic regression (SLR) analysis of tumor contour features, we reviewed 111 digitized US images of breast tumors. They were 40 benign breast tumors (fibroadenomas), and 71 infiltrative ductal carcinomas. The contour features were calculated by the radial length. A SLR model with contour features was used to classify tumors as benign or malignant. The accuracy of our model with contour features for classifying malignancies was 91.0% (101 of 111 tumors), the sensitivity was 97.2% (69 of 71), the specificity was 80.0% (32 of 40).