29/07/2024
MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and Massachusetts General Hospital (MGH) have developed Mirai, an advanced AI model 🌟 designed to predict breast cancer risk up to five years in advance using mammograms.
Mirai has demonstrated greater accuracy than the widely-used Tyrer-Cuzick model and previous deep learning methods. It achieved a C-index of 0.76 on the MGH test set, surpassing both the Tyrer-Cuzick model (0.64) and other AI models (0.72). 📊
The AI was trained on an extensive dataset from Massachusetts General Hospital (MGH) and validated on additional datasets from Karolinska University Hospital in Sweden 🇸🇪 and Chang Gung Memorial Hospital (CGMH) in Taiwan 🇹🇼.
Mirai outperformed existing risk models, showing superior sensitivity and specificity in identifying patients at high risk of developing breast cancer within five years. The model’s performance was consistent across different racial groups, age ranges, and breast density categories. 🏆
The development of Mirai highlights the transformative potential of AI in breast cancer risk assessment. 🤖💡
👉 Follow for the latest AI updates!
