An Innovative Framework for the Detection and Classification of Breast Cancer Disease Using Logistic Regression Compared with Back Propagation Neural Network
K. Reema Sekhar, Ashley Thomas · 2025
The primary objective of the study is to identify breast cancer diseases with an improved accuracy rate using logistic regression (LR) classifiers in comparison with back propagation neural network (BPNN) classifiers. The research dataset used in the study was sourced from the UCI database. The sample size of different segmentation algorithms for breast cancer image processing detection with improved quality taken samples was 50 (Group 1=25 and Group 2=25) (G-power=0.08). A comparative analysis of the detection rate in different segmentation algorithms for breast cancer diseases image processing detection is performed by LR and BPNN whereas the number of samples was kept as 25 (N=25) for each test group for both the techniques. The detection rate of the LR algorithm is 96.40% whereas the result of the BPNN algorithm detection rate is 88.60%. The significance level of the investigation was determined to be p = 0.001 (p<0.05). This suggests that there is a statistically significant difference among the test groups. The LR algorithm portrays a better performance (96.40%) in detecting breast cancer when compared to the BPNN algorithm (88.60%).