Title: ai Detection Tool Outperforms Human Radiologists in Identifying Breast Cancer Cases
An innovative ai tool named Mia has demonstrated the ability to identify signs of breast cancer missed by human radiologists. In a pilot study conducted in the UK, Mia analyzed mammograms of over 10,000 women, successfully detecting all cancer cases with symptoms and an additional 11 undetected cases.
Participants and Performance
Most of the participants were cancer-free, but only 81 chose not to let their scans be assessed by Mia. The ai system was trained on over 6,000 previous breast cancer cases and boasted an accuracy rate of 81.6% in predicting the presence of cancer and a false negative rate of 27.1%, ruling it out in 72.9% of cases when evaluated on new participants.
Importance and Future Developments
Breast cancer is the most frequent cancer in women, with approximately two million new cases diagnosed globally each year. Although survival rates have improved significantly due to early detection and advanced treatments, many patients still experience adverse side effects like lymphoedema following surgical intervention and radiotherapy.
Predicting Side Effects
To mitigate these consequences, researchers are currently developing Mia to predict a patient’s risk of side effects up to three years post-treatment. This would enable doctors to personalize care with alternative therapies or additional supportive measures for high-risk patients.
Clinical Trial
The research team intends to enroll 780 breast cancer patients in a clinical trial called Pre-Act, aiming to prospectively validate the ai risk prediction model over a two-year follow-up period. The ultimate objective is an ai system that can provide a comprehensive evaluation of a patient’s prognosis and treatment needs.
For more information on upcoming enterprise technology events and webinars, visit TechForge.
