{
  "id": 15706854,
  "title": "Researchers use ChatGPT-style technology to predict stroke risk from a person’s DNA",
  "url": "https://urgent.news/2026/10/20/researchers-use-chatgpt-style-technology-to-predict-stroke-risk-from",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-20T05:32:38.000Z",
  "source": {
    "name": "Euronews",
    "slug": "euronews",
    "url": "https://www.euronews.com/2026/10/20/researchers-use-chatgpt-style-technology-to-predict-stroke-risk-from-a-persons-dna"
  },
  "original_language": "en",
  "account": "Scientists have created a new AI model that uses genetic data to assess a person's risk of ischaemic stroke, the most common type of stroke. The University of Technology Sydney (UTS) research team identified 18 specific genetic variations linked to stroke risk. These genetic factors account for about 38% of an individual's overall stroke risk, according to previous studies. Traditional methods of assessing stroke risk have not traditionally focused on genetics, but this new AI model developed by the UTS team could help improve stroke risk assessments and prompt more personalized preventative measures. The AI model, built using transformers technology similar to that used in large language models like ChatGPT, successfully categorized individuals into low, medium and high genetic risk groups when tested on two major medical datasets containing over 40,000 people. Individuals in the high-risk group were 21% more likely to experience a first-time ischaemic stroke compared to those in the low-risk group. While polygenic risk scores have been used to estimate genetic risk for diseases, this new AI model goes beyond by accounting for how different genetic variations interact with one another and their specific positions along the DNA sequence. Researchers believe this personalized approach could pave the way for targeted treatments and improved overall stroke prevention efforts.",
  "summary": "A new AI tool that identifies genetic risk factors for stroke could open the door to more personalised prevention and treatment.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}