Every month, I’ll be highlighting Alzheimer’s research that is going on RIGHT NOW! Harvested from different websites, journals and podcasts, I’ll translate them into understandable English and share them with you. Today: Using math to predict my chances of getting Alzheimer’s? (Why isn’t THIS headline news instead of another probably disappointing “drug” that costs a gazillion dollars and just does the same thing that some different drug did before? (https://breastcancerreaper.blogspot.com/2021/07/alzheimers-research-right-now-14-fda.html))
Two separate studies show similar results in VERY different ways.
In the first, scientists used stem cells in people (taken with their permission, these cells come from an individual – basically cells that aren’t heart or brain or skin or nerve cells. They were able to “turn on the cells” so they became new brain cells in a test tube. The people they started with had either been identified with Alzheimer’s or had passed away from Alzheimer’s complications. The new cells aged, and as they did, they produced the two main suspects in AD – amyloid plaques and Tau tangles – plaques and tangles. They behaved exactly as the cells in the study volunteers had.
The result: they found that there is NO SINGLE THING that causes Alzheimer’s in people. It’s a complicated mixture of genetics, environment, exercise, what you eat, and even what kind of medications you take. Only fifty people took part in the study – but it has yielded some startling information and only the future can tell what else we might learn!
The second discovery? “Researchers from Lithuania developed a deep learning-based method that can predict the possible onset of Alzheimer's disease from brain images with an accuracy of over 99 percent.”
OK, let’s unwrap this a little. What’s “deep learning”? “Deep learning is a…technique that teaches computers to do what comes naturally to humans: learn by example…[it’s behind driverless cars, letting them tell the differences between a stop sign, a pedestrian, and a lamppost. It lets you use Siri or HeyGoogle, or whatever system to voice control your phones, tablets, TVs, and hands-free speakers. Deep learning lets your computer “learn” to perform classification tasks directly from images, text, or sound.”
So, a computer with the deep learning program has learned to predict whether or not I’ll get Alzheimer’s – with nearly 100 % accuracy!
The problem? You have to get an fMRI scan for this to work. What the heck’s an fMRI? “Functional magnetic resonance imaging (fMRI) measures the small changes in blood flow that occur with brain activity…fMRI may detect abnormalities within the brain that cannot be found with other imaging techniques.” It’s the detection of the abnormalities that form the basis of the ALGORITHM (a mathematical formula) that analyzes changes in a person’s brain over time.
Of course, PEOPLE could do this, but it takes lots of time. Computers can do it faster and better. But THAT’S NOT WHERE IT STOPS; working with the computer data, doctors can also look at variables outside of what the computer was programmed to do….
Neither of these techniques – and that’s what they are, sharpened tools of things scientists have been trying to do for yours, combined with cloning techniques and advanced computer techniques – to better understand what goes on inside an AD person’s head. Once they understand what IS happening, they can figure out how to STOP WHAT’S HAPPENING…
And that’s when we get a cure.
Resources: https://medicalxpress.com/news/2021-09-brain-patient-specific-alzheimer-insights-cognitive.html, https://medicalxpress.com/news/2021-09-algorithm-alzheimer-percent-accuracy.html, Deep Learning: https://machinelearningmastery.com/what-is-deep-learning/
Image: https://www.meduniwien.ac.at/web/fileadmin/_processed_/e/1/csm_shutterstock_142671010_4683b6bf13.jpg
No comments:
Post a Comment