K-Means Clustering and Classification of Breast Cancer Images Using Histogram of Oriented Gradients Features and...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Breast cancer is a leading cause of cancer death in women globally.
• Early diagnosis significantly improves mortality rates.
• Conventional diagnostic methods like mammograms require expert interpretation, which can be subjective and prone to errors.

Purpose of the Study:

• To develop an innovative hybrid technique for classifying breast cancer images.
• To enhance the accuracy and reliability of automated breast cancer detection systems.

Main Methods:

• A dataset of 2788 breast cancer images (1480 benign, 1308 malignant) was utilized.
• A three-stage hybrid approach was employed: K-means clustering for unsupervised grouping, Histogram of Oriented Gradients (HOG) for feature extraction, and Convolutional Neural Network (CNN) for classification.
• Performance was evaluated using accuracy, precision, recall, and F1-score.

Main Results:

• The hybrid model achieved a high classification accuracy of 98%.
• Precision, recall, and F1-scores were 0.98 for both benign and malignant classifications.
• K-means clustering successfully identified distinct groupings for benign and malignant tumors.

Conclusions:

• The combination of HOG features and CNN classification demonstrates excellent performance for breast cancer detection.
• This automated approach shows potential for clinical application to aid radiologists in efficient malignant tumor identification.
• Future work will involve exploring additional imaging modalities and clinical validation.