AI reduces false positives by 37.3% in breast cancer diagnosis | Women’s Health

Despite making up half of the global population, women's health has often been sidelined by traditional healthcare systems. Systemic neglect has led to delayed diagnoses and inadequate care. Artificial intelligence (AI) is emerging as a transformative force in women's healthcare, enhancing clinical decision-making, improving diagnostic accuracy, and guiding treatments for better outcomes.

As we celebrate International Women's Day on March 8th, let's explore how AI is changing the landscape and empowering women's healthcare under the theme of "Rights. Equality. Empowerment."

The Gender Gap in Healthcare

Historically excluded from clinical trials until the 1970s, women's health remains under-researched. Conditions like endometriosis, which affects 1 in 10 women, and polycystic ovary syndrome (PCOS) are often undiagnosed or dismissed. Women also face higher misdiagnosis rates in critical situations, such as heart attacks, with a 66.1% misdiagnosis rate reported by Soliant Health in 2024. These disparities highlight the need for innovative solutions in women's healthcare.

Diagnostic Accuracy and Early Detection

Multiple studies have shown that AI is significantly improving diagnostic accuracy. A study published in Nature Communication showed that radiologist-level AI systems help radiologists decrease their false positive rates by 37.3% in breast ultrasound diagnosis. Similarly, an AI tool tested by the NHS analyzed over 10,000 mammograms, identifying all breast cancer cases, including 11 missed by human doctors, according to BBC News.

AI algorithms, particularly those based on deep learning, excel in interpreting medical images, often surpassing radiologists' precision. A Google Health AI model outperformed radiologists in detecting breast cancer from mammograms, reducing false positives and negatives.

Enhancing Clinical Decision-Making

With the rapid growth of machine learning in medical technology, AI has been increasingly integrated into healthcare in various ways, with numerous related research projects underway around the world. One important aspect is that AI can now support clinical decision-making and reduce gender bias. AI models can analyze medical histories and symptoms, assisting doctors in ruling out potential diagnoses and reducing unnecessary tests.

One such tool, Genetic Transformer (GeneT), is a multimodal gene detection model developed by BGI Genomics. It integrates pathogenic biomarker selection, verification, and whole-genome consultation to enhance diagnostic precision. The study focused on GeneT and rare disease screening is undergoing peer review and has been preprinted on MedRxiv.

"AI can quickly master professional knowledge in a disease field by analyzing literature, open sources, and data from BGI Genomics' collaborations," said Dr. Liang Lungang, Deputy Director of IT Department at BGI Genomics. "This 'knowledge' enables AI to provide actionable insights for clinicians."

BGI Genomics is collaborating with major hospitals to apply first-hand clinical experience to the diagnosis and treatment of rare diseases, ensuring that AI has practical clinical knowledge. With sufficient data input, tools like GeneT can revolutionize the diagnosis and treatment of genetic conditions affecting women.

Predictive Analytics and Personalized Treatment

AI's predictive analytics are revolutionizing treatment by leveraging genetic and clinical data. It can estimate a patient's risk of recurrence, metastasis, and other complications. The predictive analytic feature is invaluable for conditions like colorectal and breast cancer, where early intervention can significantly improve outcomes.

Dr. Lin Cong from the Institute of Intelligent Medical Research (IIMR), featured in BGI Genomics' ChatDNA series, is developing an AI model for colorectal cancer screening. This model not only assists clinicians in diagnosis but also provides prognostic predictions that guide personalized treatments.

"We aim to develop a universally applicable AI model in the next 3–5 years," Dr. Lin said. "It will enable faster, more accurate diagnoses, giving clinicians the tools to make data-driven decisions that can transform patient care."

Guiding Personalized Treatment

AI integrates various data types—medical records, images, and genetic profiles—to provide a comprehensive view of a patient's condition. MIT researchers have also developed an AI model that identifies stages of ductal carcinoma in situ (DCIS) from breast tissue images, showing agreement with pathologists, according to MIT News. MIT's study showed that AI could identify breast tumor stages that are likely to progress to invasive cancer by analyzing chromatin images. AI also predicts how patients will respond to different cancer treatments, helping clinicians choose the most effective treatment plans. Prognostic modeling enables more personalized and proactive cancer management.

AI is not just a tech trend; it's a necessity in women's healthcare. For this International Women's Day, we are together to celebrate the empowerment in women's healthcare by new technology. From enhanced diagnostics and personalized treatments to predictive analytics, AI has the potential to transform women's healthcare. The future is one of continuous innovation and collaboration, aiming to create a more equitable and effective healthcare system for women worldwide.

About BGI Genomics

BGI Genomics, headquartered in Shenzhen, China, is the world's leading integrated solutions provider of precision medicine. Our services cover more than 100 countries and regions, involving more than 2,300 medical institutions. In July 2017, as a subsidiary of BGI Group, BGI Genomics (300676.SZ) was officially listed on the Shenzhen Stock Exchange.