editor
Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and the Jameel Clinic have developed an advanced AI algorithm named “Mirai,” designed to predict breast cancer risk with high accuracy. This system, which leverages deep learning, has shown robust performance across diverse datasets from the U.S., Europe, and Asia, predicting breast cancer risk up to five years in advance.
Mirai stands out by its ability to perform consistently across various populations and clinical environments, improving on existing models by capturing unique risk factors. It considers clinical data such as age and family history, and employs an additive-hazard layer to model risk over time. Validated on datasets from Massachusetts General Hospital (MGH), Karolinska Institute in Sweden, and Chang Gung Memorial Hospital in Taiwan, Mirai outperformed previous methods, identifying nearly twice as many future cancer diagnoses compared to current standards.
Also Read: South Korea to Pile $74.6 BN into AI Chips
The algorithm’s consistent performance across different races, age groups, and breast density categories marks a significant advancement in breast cancer detection. Researchers are now collaborating with hospitals worldwide to integrate Mirai into clinical care, aiming to enhance early detection and reduce unnecessary screenings.
Recent updates have highlighted Mirai’s ability to maintain its accuracy across globally diverse test sets, confirming its generalizability and effectiveness. For example, the algorithm demonstrated concordance indices of 0.76 for the MGH test set, 0.81 for the Karolinska test set, and 0.79 for the Chang Gung Memorial Hospital test set. These findings underscore Mirai’s potential to offer broad and equitable improvements in breast cancer care globally
I like the efforts you have put in this, appreciate it for all the great posts.