A new study published in the journal Nature Communications has given hope to Parkinson's patients with an artificial intelligence (AI) blood test that can predict the disease early, before symptoms appear .
Parkinson's disease, the world's second most common neurodegenerative disease affecting about 10 million people, currently has no cure. Early diagnosis plays a key role in controlling the disease and improving the patient's quality of life.
The research team from University College London (UCL) and the University of Göttingen trained a machine learning algorithm to identify eight specific proteins in the blood that are associated with Parkinson's risk.
Testing on 72 people at risk of brain disorders, including Parkinson's, showed that the method accurately predicted 16 cases of developing the disease, even seven years earlier than clinical diagnosis. The overall accuracy of the test reached 79%, showing great potential for practical application.
Professor Kevin Mills, lead author of the study from UCL, emphasized, “we need to start trial treatments before patients show symptoms.” In addition, Dr. Jenny Hällqvist, co-author of the study, also agreed that it is necessary to protect nerve cells from the beginning, instead of waiting until it is too late.
The study is considered by experts to be an important step forward in finding a simple, non-invasive and user-friendly testing method. However, more large-scale experiments are needed to verify the accuracy of this method.
Not stopping there, scientists are planning to create a simpler test, requiring only a drop of blood on a card to send to the analysis laboratory, to predict Parkinson's disease even earlier.
Although there are still many challenges, AI blood testing is expected to open new directions in early diagnosis, monitoring treatment effectiveness and developing breakthrough treatments for Parkinson's disease, bringing hope. Hope for millions of sick people around the world.