JMIR Publications

ABSTRACT

Background:

Breast cancer is one of the most prevalent cancers affecting women worldwide. Early detection significantly reduces mortality rates. Traditional diagnostic methods, such as mammograms and biopsies, require expert interpretation, which can be time-consuming and subject to variability. Artificial intelligence (AI) has been applied to automate detection and classification, but achieving high accuracy and reliability, particularly in distinguishing benign from malignant tumors, remains a challenge.

Objective:

This study aims to develop an automated breast cancer classification model by integrating K-means clustering for unsupervised analysis, Histogram of Oriented Gradients (HOG) for feature extraction, and a Convolutional Neural Network (CNN) for classification.

Methods:

A dataset of 1280 breast cancer images, evenly split between benign and malignant cases, was used. K-means clustering was applied to group images based on visual similarities, followed by Principal Component Analysis (PCA) for dimensionality reduction and visualization. HOG was then used to extract edge-based features, which were fed into a CNN for classification.

Results:

The CNN achieved a classification accuracy of 98%, with precision, recall, and F1-score values of 0.98 for both benign and malignant cases. K-means clustering revealed distinct groups corresponding to benign and malignant tumors, indicating natural separability in the image data.

Conclusions:

The combination of HOG feature extraction and CNN-based classification demonstrated high performance in breast cancer detection. The model offers a promising framework for automated diagnosis, with potential clinical applications to assist radiologists in identifying malignant tumors more efficiently. Future research will explore additional imaging modalities and real-world clinical testing.