The research team included Doctoral Researcher Gudhe Raju, Professor Arto Mannermaa, and Senior Researcher Hamid Behravan, who emphasized the necessity of building trust among healthcare professionals through rigorous testing for AI integration.

This study utilized a multi-view deep evidential fusion approach, integrating Dempster-Shafer evidential theory and subjective logic for a comprehensive analysis of mammogram images.

Researchers at the University of Eastern Finland have developed a novel AI-based algorithm known as MV-DEFEAT, designed to enhance mammogram density assessment.

The study aimed to improve breast cancer risk stratification by combining BI-RADS density scores with a deep-learning-based texture score.

The MV-DEFEAT algorithm demonstrated robust performance across various datasets, showcasing its adaptability to diverse patient demographics.

Using the VinDr-Mammo dataset of over 10,000 mammograms, MV-DEFEAT achieved a significant 50.78% improvement in screening accuracy, effectively distinguishing benign from malignant tumors.

The findings underscore the importance of AI in medical diagnostics, suggesting that integrating texture analysis into mammography screening could enhance early cancer detection and improve patient outcomes.

Deep-learning-based texture analysis was shown to enhance risk assessment beyond what density alone can provide, indicating its potential clinical utility.

While high breast tissue density is linked to a greater risk of breast cancer, the study found that mammographic density does not always correlate with high cancer rates, highlighting the need for comprehensive evaluations.

Screen-detected cancer rates varied significantly across different BI-RADS and texture score combinations, emphasizing the necessity for tailored screening strategies.

The study revealed that interval cancer rates were significantly higher in the BI-RADS 4 category, indicating a pressing need for improved screening protocols for women with high-density breasts.

The research team stressed the importance of ongoing refinement and validation of MV-DEFEAT to ensure its reliability in clinical settings.