USA: Duke University researchers have developed an artificial intelligence (AI) model that can predict five-year breast cancer risk from mammograms, according to a study published in Radiology.
The researchers revealed that their deep learning-based algorithm, a simplified offshoot of Mirai, performed well in predicting cancer risk by evaluating breast asymmetry features on mammograms.
"Localized bilateral dissimilarity, an imaging marker for breast cancer risk, approximated the predictive power of Mirai and was a key to Mirai’s reasoning," the study stated.

While attending regular screening mammogram appointments is crucial in reducing the risk of dying from breast cancer, pinpointing which women face a higher risk of developing the disease can be tricky.
AI can be helpful in this area. Mirai is a deep learning-based algorithm that has shown that it can help predict breast cancer. However, not much is known about Mirai’s reasoning process. The researchers cautioned that with this in mind, the algorithm could be relied on too heavily by radiologists and lead to incorrect diagnoses.
Jon Donnelly, Duke University, Durham, NC, and colleagues aimed to identify whether bilateral dissimilarity underpins Mirai’s reasoning process. For this purpose, they developed AsymMirai, which they highlighted is easier and simpler to understand than Mirai.
The algorithm was built on the front-end deep learning portion of Mirai and was developed with local bilateral dissimilarity as an interpretable module. This module examines tissue differences between the left and right breasts.
The researchers compared 210,067 mammograms from 81,824 patients in the EMBED (EMory BrEast imaging Dataset) from 2013 to 2020 using both Mirai and AsymMirai models.
Pearson correlation coefficients were computed between the risk scores of Mirai and AsymMirai. Subgroup analysis was performed in patients with consistent AsymMirai’s year-over-year reasoning. Mirai and AsymMirai risk scores were compared using the area under the receiver operating characteristic curve (AUC).
The study included screening mammograms (n = 210 067) from 81,824 patients (mean age, 59.4 years).
The study revealed the following findings:
Deep learning–-extracted bilateral dissimilarity produced similar risk scores to those of Mirai (1-year risk prediction, r = 0.6832; 4–5-year prediction, r = 0.6988) and achieved similar performance as Mirai.
For AsymMirai, the 1-year breast cancer risk AUC was 0.79 (Mirai, 0.84), and the 5-year risk AUC was 0.66 (Mirai, 0.71).
In a subgroup of 183 patients for whom AsymMirai repeatedly highlighted the same tissue over time, AsymMirai achieved a 3-year AUC of 0.92.
"The study represents not a potential contribution of artificial intelligence to the existing body of knowledge in breast cancer risk prediction assessment," Vivianne Freitas, MD, from the Joint Department of Medical Imaging in Toronto, Canada, wrote in an accompanying editorial.
Along with that, it shows that AI models “can be understandable and effective, serving as a bridge and narrowing the divide between the intricate world of AI algorithms and their real-world clinical application,” she wrote.
Reference:
Donnelly J, Moffett L, Barnett AJ, Trivedi H, Schwartz F, Lo J, Rudin C. AsymMirai: Interpretable Mammography-based Deep Learning Model for 1-5-year Breast Cancer Risk Prediction. Radiology. 2024 Mar;310(3):e232780. doi: 10.1148/radiol.232780. PMID: 38501952.