{
  "id": 5354638,
  "title": "Routine clinical data help AI predict rapid Parkinson's decline three to five years in advance",
  "url": "https://urgent.news/2026/09/03/routine-clinical-data-help-ai-predict-rapid-parkinsons-decline-three",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-03T15:40:02.000Z",
  "source": {
    "name": "Medical Xpress",
    "slug": "medical-xpress",
    "url": "https://medicalxpress.com/news/2026-09-routine-clinical-ai-rapid-parkinson.html"
  },
  "original_language": "en",
  "account": "A new study led by researchers at the University of Miami suggests that artificial intelligence (AI) may be able to predict which Parkinson's disease patients are more likely to experience rapid cognitive or motor decline within the next three to five years. Published in npj Parkinson's Disease, the research indicates that machine-learning models could use clinical measurements already collected during routine patient visits to make these predictions, rather than advanced brain imaging techniques.\n\nIhtsham ul Haq, M.D., a professor of neurology at the University of Miami Miller School of Medicine and senior author of the study, emphasized the difficulty in knowing how Parkinson's disease will progress. The interdisciplinary team, including radiologists, computer scientists, and AI experts, aimed to utilize AI's capability to analyze various types of information simultaneously to identify patients at risk of more rapid decline.\n\nYelena Yesha, Ph.D., a professor in computer science and a faculty member with a secondary appointment in radiology, collaborated on the project, highlighting the interdisciplinary approach. The study involved 1,602 participants from the Parkinson's Progression Markers Initiative and was tested on an independent validation cohort of 541 patients. The researchers trained machine-learning models using MRI scans and detailed clinical data, finding that structural MRI data provided relatively little additional predictive value beyond information obtained through routine clinical evaluations.\n\nThe most informative predictors for motor decline were the results of a synuclein seed amplification assay (SAA), which detects abnormal alpha-synuclein biology, and the rate at which a patient's MDS-UPDRS motor score worsened. For cognitive decline, the study also identified key predictors but did not specify them in the source material. The researchers concluded that clinical information, rather than MRI scans, offered the strongest predictive power for future Parkinson's disease decline.",
  "summary": "For patients diagnosed with Parkinson's disease, one of the greatest uncertainties is what comes next. Some people experience relatively mild symptoms for years. Others face a much steeper course marked by worsening movement problems, cognitive decline or both.",
  "key_points": [
    "AI predicts rapid Parkinson's decline within 3-5 years using routine clinical data",
    "Machine-learning models analyze structural MRI and clinical evaluations",
    "Synuclein seed amplification assay and MDS-UPDRS motor score are key predictors"
  ],
  "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."
}