© Gorodenkoff/Shutterstock.com
Artificial intelligence (AI) has the potential to revolutionize many fields; medicine is of course one of them. Algorithms are indeed capable of detecting certain diseases at an ultra-early stage, but one model has gone even further by identifying women at risk of developing breast cancer. And this, several years before it is diagnosed.
Detecting several years before diagnosis
These are the results of a study conducted by an international team of researchers, including experts from the FHI (Norwegian Institute of Public Health) and the Universities of California and Washington. The latter was based on a considerable database: mammograms from 116,495 women who participated in the Norwegian screening program between 2004 and 2018. Among them, 1,607 developed breast cancer during the period studied.
Responsible for analyzing the mammograms, the algorithm was not only able to identify women at high risk of developing breast cancer, but also to precisely locate the breast concerned, four to six years before the actual diagnosis. According to Solveig Hofvind, head of the screening program and the AI project, “ the breast that developed cancer had an AI score about twice as high as the other “.
The particularity of these results also lies in the researchers' approach, as they exploited an existing commercial AI model to carry out the analysis of mammograms.
200% Deposit Bonus up to €3,000 180% First Deposit Bonus up to $20,000
© Angiola Harry/Unsplash
A first test in real conditions
Another study, this time carried out by the University of Lübeck in Germany, tested Artificial Intelligence in a National Breast Cancer Screening Program. A First in Real Conditions.
Thus, the examinations of the 461,818 participants in the program were analyzed independently by two radiologists. For 260,739 of them, at least one of the experts used an AI. The results are unequivocal: the detection rate was 17.6% higher with AI, or one additional case detected per 1,000 women. The tool enabled 204 diagnoses, without an increase in false negatives, reports The Guardian.
This research opens up promising prospects for the development of personalized screening programs, more effective and potentially less expensive. All the more so in a context of a shortage of specialists. Because early diagnosis is vital: the earlier a cancer is detected, the greater the chances of survival.
Breast cancer is the most common cancer in France and is the leading cause of cancer death in women.
- In one study, AI was able to detect women at risk of developing breast cancer several years before diagnosis.
- In another study, radiologists using AI were more effective at spotting breast cancers.
- More and more experts are calling for more widespread use of technology to diagnose breast cancer.
📍 To never miss any Presse-citron news, follow us on Google News and WhatsApp.
[ ]