Northeastern University Researchers Unveil Groundbreaking AI for Breast Cancer Detection

Overall, the average risk of a woman in the United States developing breast cancer sometime in her life is about 13%. This means there is a 1 in 8 chance they will develop breast cancer.

Earlier this year, researchers at Northeastern University introduced a revolutionary web-based AI tool designed to enhance the speed and accuracy of prostate cancer diagnoses. Building on this success, the team, led by bioengineering professor Saeed Amal, has now developed an advanced AI architecture to detect breast cancer. Remarkably, this new system boasts an impressive accuracy rate of 99.72%.

Breast cancer remains a critical health issue, constituting 30% of new cancer cases among women annually. The American Cancer Society predicts that in 2024, approximately 42,500 women will succumb to this disease. This new AI tool represents a significant leap forward in the early detection and treatment of breast cancer, potentially saving thousands of lives each year.

The AI developed by Amal and his team leverages high-resolution images and historical data to identify cancerous patterns. Unlike human diagnosticians, this AI system does not suffer from fatigue, ensuring consistent and reliable detection even after numerous analyses. “The AI can’t miss a tumor in the biopsy and won’t be exhausted after diagnosing 10 or 20 people,” Amal explains. This reliability is crucial for improving diagnostic accuracy and patient outcomes.

This project is part of a broader initiative led by Amal to create an online framework that clinicians can utilize to diagnose a variety of cancers using innovative AI technologies. The goal is to not only improve the speed and accuracy of diagnoses but also to aid in the development of AI models capable of diagnosing rare and uncommon cancers with limited patient data. According to Amal, this new tool has the potential to “redefine digital pathology.”

For the breast cancer detection project, researchers utilized publicly available datasets, specifically the Breast Cancer Histopathological Database, which contains images of both malignant and benign breast tissues. The findings, recently published in the journal Cancers, highlight the significant advancements made in AI-driven cancer diagnostics.

The advancements made by Professor Amal and his team at Northeastern University represent a monumental step forward in the fight against cancer. This new AI tool not only exemplifies the potential of technology to transform healthcare but also sets a new standard for accuracy and efficiency in cancer diagnostics. The future of digital pathology looks promising, with significant implications for patient care and treatment outcomes.

Citations and To learn more, read this article! : https://news.northeastern.edu/2024/06/24/ai-breast-cancer-diagnosis/