Can AI Help Save Lives From Breast Cancer? These Scientists Think So

One of the biggest fears we women live with is breast cancer. In the United States, it is the most common cancer among women. It accounts for about 30% (or 1 in 3) of all new female cancers each year.

According to the American Cancer Society, by 2024, there could be 310,720 new cases of invasive breast cancer diagnosis in women. Of these, 42,250 are likely to die.

However, a new artificial intelligence model could radically reduce these numbers.

Forget ChatGPT. AI could save thousands of lives

While the world enjoys making funny pictures or automating processes to make work easier through new artificial intelligence tools, scientists at MIT have developed a model that promises to revolutionize the world.

It is a machine-learning model that can identify the stage of breast cancer early. This new development could help reduce overtreatment and prevent invasive cancer from progressing.

Early diagnosis is a gamechanger

MIT scientists have worked with a specific type of cancer, ductal carcinoma in situ (DCIS). This is a type of pre-invasive tumor that sometimes progresses to a very deadly form of breast cancer. DCIS accounts for about 25% of all breast cancer diagnoses.

It is difficult for physicians to determine the type and stage of DCIS, so patients with this diagnosis are often overtreated. To solve this problem, an interdisciplinary team of MIT and ETH Zurich researchers has developed an artificial intelligence model capable of identifying the different stages of DCIS from an inexpensive and easy-to-obtain breast tissue image.

MIT’s efforts are predictive but accurate

Their model demonstrates that both the state and arrangement of cells in a tissue sample are essential in determining the stage of DCIS.

The researchers’ ease of obtaining the tissue images helped them create one of the largest data sets of its kind. So, they decided to train and test their model. As the team explained, when they compared their predictions with the findings of a pathologist, they observed a clear match in many cases.

New developments in AI could change the future of breast cancer treatment

The new MIT model could serve as a tool to speed up breast cancer diagnosis. It could avoid laborious testing and give doctors more time to evaluate cases in which it is unclear whether DCIS will become invasive.

“We took the first step in understanding that we should be looking at the spatial organization of cells when diagnosing DCIS, and now we have developed a technique that is scalable,” said Caroline Uhler, a professor in the Department of Electrical Engineering and Computer Science (EECS) and the Institute for Data, Systems, and Society (IDSS). “From here, we really need a prospective study. Working with a hospital and getting this all the way to the clinic will be an important step forward.”

What does this mean for Latinas?

As explained by the Breast Cancer Research Foundation, while there has been an overall 40% decline in breast cancer deaths over the last 30 years in the U.S., there is a persistent mortality gap between racial groups in the U.S.

Now, while in the U.S., the incidence of breast cancer in Latinas is 28% lower than in non-Latino white women, Latinas are more likely to be diagnosed at a younger age and with more aggressive diseases such as triple-negative breast cancer, which has fewer targeted treatments.

We are also diagnosed at more advanced stages and are about 30% more likely to die from breast cancer than our non-Latino white counterparts.

That is why achievements like MIT’s could facilitate early diagnosis and save millions of lives.