DayFR Euro

New AI promises to better predict the course of breast cancer

An important breakthrough could soon improve the treatment of breast cancer thanks to artificial intelligence (AI). In collaboration with medical institutions in the United States, researchers from the startup Ataraxis AI have in fact developed an AI model that predicts the progression of this disease with greater precision than current tests used in hospitals. If their results are confirmed, this technology could transform the way doctors evaluate and treat breast cancer by enabling a more targeted and personalized approach.

The challenge of prediction in breast cancer

The cancer be you is one of the most common forms of cancer in women, although it can also affect men. It grows from cells in the ducts or lobules of the breast and when not detected early, it can spread to lymph nodes and other parts of the body.

There are many types of breast cancer and not all behave the same: some progress quickly, while others progress more slowly. This diversity complicates medical decisionsbecause doctors must determine as accurately as possible the rate of progression of a cancer to administer the most appropriate treatment. However, an inaccurate assessment of the rate of progression can lead to either heavier treatments than necessary or insufficient care.

Tests exist to assess the level of aggressiveness of a cancer and adapt care to the case of each patient. One of the best known, theOncotype DXuses genetic data to estimate the risk of recurrence or the probability of cancer progression. However, these tests remain limited in precision and can give variable results, which leaves a margin of uncertainty for doctors.

The AI ​​model developed by Ataraxis AI

This is whereAtaraxis AI intervenes. Founded recently, this biotechnology startup has collaborated with several hospitals to access vast databases including images of tumors and detailed statistical information on the progress of breast cancer in different patients. On this basis, they developed a machine learning model capable of analyzing complex sets of data, allowing it to give a more accurate estimate of the rate of progression of a cancer in a given patient.

The Ataraxis AI model is distinguished by a unique approach. Instead of relying on a single algorithm, the team designed several distinct AI models that work in complementary ways. Each model uses different methods to interpret data, but produces the same types of results. By aggregating these results, Ataraxis AI obtains a more reliable average rating. This aggregation method is supposed to minimize errors, which results in greater accuracy.

Credits: Motortion/iStock

Promising test results

The first tests on the data of 3 500 patients show that this AI model could really be a game-changer. The Ataraxis AI team tested their model against results from standard tools, such as Oncotype DX, and found that their model was up to 30% more precise. This difference is significant because it means that doctors could have a more accurate picture of the rate of progression of each patient’s cancer, and therefore prescribe more appropriate treatments.

These promising results will still need to be confirmed in real-world clinical trials before the technology can be widely used. However, if this is the case, then the Ataraxis AI model could not only improve breast cancer managementbut also inspire other AI applications in oncology. Indeed, a model capable of predicting the evolution of a tumor with such precision could be adapted to other types of cancer where the aggressiveness and rate of progression also vary from patient to patient. This would contribute to more precise medicine where each patient would benefit from personalized treatment and monitoring adapted to their specific needs.

Ataraxis AI researchers are already developing other tools based on this model with the aim of supporting doctors in their decision-making. They also plan to make their model available as software for hospitals next year, which could mark a turning point in the monitoring of breast cancer patients.

-

Related News :