Mirai FAQ

What do the terms “prospective” and “retrospective” mean?

Before MIRAI is deployed in a hospital, it is “retrospectively” tested on a large dataset consisting of past images taken from patients over the course of 5 years. These images include patients who did not end up developing cancer within 5 years and patients who did end up developing cancer within 5 years. These past images help clinicians ensure the model is working properly before moving on to use the model “prospectively” on real patients.

The map above shows where MIRAI is still being tested retrospectively and where MIRAI is being prospectively tested on real patients.

How does Mirai work?

The current model utilizes your family history, demographics, and other information to create a risk score.  MIRAI works differently and looks at the patterns in your mammogram. MIRAI uses an algorithm (or a set of rules to be followed in a calculation) to model a patient’s risk over time. The algorithm was developed by “teaching” the software tool with over 200,000 mammograms, and then – confirming its accuracy on other sets of mammograms. In this way, the tool gets a lot of experience by looking at many different patients, and then can make better predictions for future patients. MIRAI gives doctors another tool to use to personalize the best care for their patients. If you’re interested in a more in-depth technical explanation, the original research paper is open-access.

Can the software “see things” that the radiologist cannot see?

Extensive training and experience are what allow radiologists to have a good intuition with respect to risk assessment, usually heavily relying on statistics based on risk factors such as age. MIRAI helps the doctors to formalize this intuition, by proposing a systemic and comprehensive computation of risk over time. Moreover, radiologists usually don’t use images to assess risk, but rather to give a diagnosis if they notice something abnormal on the image. On the other hand, MIRAI can identify patterns in the image that may indicate future risk. MIRAI software, though not yet FDA-approved, has been tested in 7 different hospitals across the world in over 100,000 women and has been shown to be more accurate in predicting the risk for developing breast cancer in the short term (over a 3-5 year period) compared to traditional tools. Researchers also found that MIRAI was accurate across patients who were of different races, ages, and breast density categories.

Why are things like family history not included in the tool?  Does this affect the tool’s reliability?

MIRAI has been trained to predict risk factors from using only a mammogram image and uses these factors to give the final risk assessment. The software predicts an individual’s risk equally well, with or without the addition of family history data. Traditional risk factors are therefore captured in the mammogram image but in a smart way, that allows it to still account for them even for patients for which this data is missing.

Does the computer make the decision, or the doctor?

Ultimately, all the decisions are made by doctors, not the computer. The aim of MIRAI is to provide physicians with an even more accurate risk assessment tool than the ones on which they already rely.

Is MIRAI validated for use in all women of all ages as a standard of care?

MIRAI has been retrospectively validated on globally diverse test sets from Massachusetts General Hospital, USA; Novant, USA; Emory, USA; Maccabi-Assuta, Israel; Karolinska, Sweden; Chang Gung Memorial Hospital, Taiwan; and Barretos, Brazil.

The Novant dataset patients were age 40-69 years. Maccabi-Assuta dataset patients were age 30 years or older. The Karolinska dataset patients were age 40-74 years. The CGMH dataset patients were age 45-70 years. Barretos patients were patients age 40 to 69 years.

You can learn more about the multi-institutional validation of Mirai here.

Can I use MIRAI for my own mammogram?

You are welcome to access the code for MIRAI to assess your mammogram for breast cancer risk. Due to a high volume of these types of requests, we are not able to use MIRAI to assess your mammogram on your behalf nor are we able to accept requests for enrollment in a clinical trial.

Is MIRAI being deployed in hospitals for women undergoing breast cancer screening?

MIRAI is currently being prospectively and retrospectively deployed for real patients in several hospitals around the world as part of our hospital network. However, its current usage is focused primarily on research and it is not being used on all patients undergoing breast cancer screening. We hope that future results from the research will allow MIRAI to be integrated into the standard of care.

Is patient privacy protected in MIRAI’s training process?

MIRAI is trained on de-identified patient data, meaning the data has no personally identifiable information such as a person’s name, email address, social security number, address, etc.