A Breast Cancer Image Classification Algorithm with 2c Multiclass Support Vector Machine

Abstract

Breast cancer is the most frequent type of cancer in women; however, early identification has reduced the mortality rate associated with the condition. Studies have demonstrated that the earlier this sickness is detected by mammography, the lower the death rate. Breast mammography is a critical technique in the early identification of breast cancer since it can detect abnormalities in the breast months or years before a patient is aware of the presence of such abnormalities. Mammography is a type of breast scanning used in medical imaging that involves using x-rays to image the breasts. It is a method that produces high-resolution digital pictures of the breasts known as mammography. Immediately following the capture of digital images and transmission of those images to a piece of high-tech digital mammography equipment, our radiologists evaluate the photos to establish the specific position and degree of the sickness in the breast. When compared to the many classifiers typically used in the literature, the suggested Multiclass Support Vector Machine (MSVM) approach produces promising results, according to the authors. This method may pave the way for developing more advanced statistical characteristics based on most cancer prognostic models shortly. It is demonstrated in this paper that the suggested 2C algorithm with MSVM outperforms a decision tree model in terms of accuracy, which follows prior findings. According to our findings, new screening mammography technologies can increase the accuracy and accessibility of screening mammography around the world.

Enhancing Breast Cancer Data Classification Performance in Medical Data by Employing GRU-LMA Deep Learning Classifier for Accurate Diagnosis

“…K-Means GRI (20) method was developed with the objective of minimizing the error rate and training time, but the levels were not achieved due to high-order datasets. The 2C-Multiclass SVM (21) model is trained for tumor tissue detection within the MRI dataset for efficient classification and to enhance images in two classes. Both classes are extracted, and all the histogram features are selected.…”