Hi HN, I came across this new research from University College London and Goettingen University that seems like a significant step in predictive medicine.
They've developed an AI model that analyzes a panel of eight key proteins in a person's blood. In their study, the model was able to identify patients who would go on to develop Parkinson's disease with 100% accuracy, up to seven years before any motor symptoms appeared.
The "100% accuracy" claim is obviously a very bold one and will need to be tested to verify this. The use of a simple blood test could be a game-changer for clinical trials and, eventually, early treatment. let me ask this: what do we do when we still cannot cure Parkinson's but can predict it?
jfengel · 1d ago
Thank you for that.
My father just died of Parkinson's. (Or at least, with it, but it was causing more and more serious issues.) I've probably got a decade before I would show any symptoms at all.
Meantime, I've been doing my best to be generically active and healthy -- I still run marathons, eat reasonably well, moderate alcohol, no smoking, etc. (Dad didn't smoke, either -- though there are rumors that nicotine might actually help.)
So this test would probably become commercially available right about the time I'd start taking it.
karlperera · 1d ago
Sorry to hear that I really am. We all have to whatever we can right?
andsoitis · 1d ago
There’s no cure for Parkinson’s and treatments target symptoms.
karlperera · 1d ago
That's true but do you think there will be in the near future? What kinds of treatments might be possible?
They've developed an AI model that analyzes a panel of eight key proteins in a person's blood. In their study, the model was able to identify patients who would go on to develop Parkinson's disease with 100% accuracy, up to seven years before any motor symptoms appeared.
The "100% accuracy" claim is obviously a very bold one and will need to be tested to verify this. The use of a simple blood test could be a game-changer for clinical trials and, eventually, early treatment. let me ask this: what do we do when we still cannot cure Parkinson's but can predict it?
My father just died of Parkinson's. (Or at least, with it, but it was causing more and more serious issues.) I've probably got a decade before I would show any symptoms at all.
Meantime, I've been doing my best to be generically active and healthy -- I still run marathons, eat reasonably well, moderate alcohol, no smoking, etc. (Dad didn't smoke, either -- though there are rumors that nicotine might actually help.)
So this test would probably become commercially available right about the time I'd start taking it.