Artificial intelligence-based enhancement of mammography interpretation for early breast cancer diagnosis.

Abstract

Breast cancer is a major global health concern and a leading cause of cancer-related deaths among women. Projections show annual cases will surpass 3 million by 2040, underscoring the urgent need for accurate and accessible diagnostic tools. Although mammography is the most common screening method, its accuracy can be compromised by issues such as tissue superposition and interpreter variability, especially in low-resource settings. "This study developed artificial intelligence-based algorithms to enhance early and accurate breast cancer detection from mammograms. A mixed-method approach was adopted, beginning with a systematic literature review to identify state-of-the-art AI techniques in breast cancer diagnosis. Based on these insights, several deep learning-based algorithms were designed and developed to enhance mammogram quality and support diagnostic classification. The models were trained and evaluated using publicly available mammography datasets. Experimental results demonstrated significant improvements in classification and visualization. Enhanced VGG19 achieved strong sensitivity for malignant case detection, the Cycle GAN framework produced state-of-the-art classification performance with high accuracy and AUC, and Mammogan•14 delivered near-perfect diagnostic accuracy of 0.9849, sensitivity' Of 0.9717, specificity of 1.000, and an AUC of 0.9985. Collectively, all models offered clinically feasible inference times, making them suitable for real-world applications. In conclusion, this research bridges the gap between AI innovation and clinical practice, particularly in settings with limited resources. Future work should focus on hybrid approaches that integrate deep learning with traditional machine learning techniques to strengthen robustness, improve interpretability, and ensure trustworthy deployment in clinical environments.