September 19, 2024

Athens News

News in English from Greece

Artificial intelligence system capable of diagnosing over 1000 diseases at an early stage


An artificial intelligence system called MILTON analyzes patients' lab results to identify patterns and then detects early signs of disease before symptoms develop, according to a new study.

Pharmaceutical giant AstraZeneca, which developed an artificial intelligence toolsaid it would speed up the development of more effective and targeted treatments. It could also lead to the development of more precise diagnostic tools to initiate preventive treatment.

«Many diseases only become clinically apparent when they have progressed significantly. But there are often changes in the blood that are not detectable by simple tests. The AI ​​tool can detect “signals” that predict the development of diseases such as Alzheimer’s, chronic obstructive pulmonary disease, kidney disease and others.“,” Slav Petrovsky, who led the study, told Sky News.

Scientists studied the genetic data of 484,230 volunteers from the UK Biobank. The MILTON artificial intelligence system analyzed data related to 67 classic laboratory tests and analyses. She also analyzed plasma samples from 46,327 volunteers, recording levels of 3,000 proteins (Blood plasma proteins play a critical role in many body functions, including the immune and hormonal systems).

An artificial intelligence tool has discovered subtle changes in the data and matched them to specific diseases that the volunteers had been diagnosed with 10 or more years earlier. Its predictive performance was rated “excellent” for 121 diseases and “highly predictive” for another 1,091, according to a study published in the journal Nature Genetics. While the biomarkers collected were predictive, adding analysis of 3,000 proteins significantly increased the system’s ability to flag patients at risk.

AstraZeneca emphasizes that The MILTON system is currently a research tooland more testing is needed before doctors can use it for clinical purposes.

«We have to be careful when we say we can predict disease. What we mean is that we can give you a slightly better idea of ​​your chances of developing the disease, but there are still many unknowns.” said Professor Tim Frayling, professor of human genetics at the University of Exeter in the UK.

«This approach is likely to have a greater impact on improving our understanding of how diseases progress in order to find ways to prevent them or treat them at an early stage when the disease can be reversed.“,” he added.

Professor Dusko Ilic, from the Department of Stem Cells at King's College London, said: MILTON's predictive power is “remarkable” because she opens the way for early intervention, more personalized treatments and lower healthcare costsHowever, he noted that concerned about the ethics of using this particular tool – artificial intelligence.

«The powerful predictive capabilities of this tool, if left unregulated, could be misused by insurance companies or employers to assessments of people without their knowledge and consent. This could lead to discrimination and invasion of privacy. Strict rules and oversight will be critical to ensure the system is used ethically and responsibly.“,” he noted.



Source link

Verified by MonsterInsights