Hybrid Convolution Neural Network in Classification of Cancer in Histopathology Images.

Area of Science:

• Oncology
• Computer Science
• Medical Imaging

Background:

• Breast cancer is a leading cause of death among women globally, with early diagnosis crucial for survival.
• Manual analysis of histopathology images for mitotic cell detection is time-consuming and prone to errors.
• Conventional image processing techniques lack the accuracy and efficiency needed for reliable breast cancer detection.

Purpose of the Study:

• To develop an automated system for accurate and efficient breast cancer detection using histopathology images.
• To improve the accuracy and reduce the time required for mitotic cell identification.
• To leverage deep learning for enhanced breast cancer prognosis.

Main Methods:

• A hybrid deep learning model combining two Convolutional Neural Network (CNN) architectures was developed.
• Histopathology images underwent preprocessing, segmentation (Otsu-based), and feature extraction using CNNs.
• The hybrid CNN model, incorporating model leveraging, was trained and tested on a large dataset (50,000 images each).

Main Results:

• The proposed hybrid CNN model achieved an overall accuracy of 98.9% in classifying mitotic cells.
• The model demonstrated high computational efficiency and accuracy in automatic feature extraction.
• The system effectively differentiates between cancerous and non-cancerous cells, aiding in diagnosis.

Conclusions:

• The hybrid deep learning approach significantly enhances the accuracy and efficiency of breast cancer detection.
• Automated analysis of histopathology images offers a promising tool for improving breast cancer prognosis.
• This technology has the potential to reduce diagnostic time and improve patient outcomes.