Science: AI and diagnostics

AI (artificial intelligence) has the capacity to analyze gigantic piles of data to look for weak signals in tons of noise. Humans simply cannot do that, but AI can. One area of science where there usually is huge amounts of noise that hides weak signals is in medicine. 

One of the really hard places to pick out weak signals is in diagnosis of diseases that are slow to develop. Researchers all over the place are adapting AI to sift through massive data piles to look for signals of disease years before serious symptoms develop. STD writes about really encouraging data related to early detection of Alzheimer's Disease based on an interesting source of data, speech patterns: 
New AI Uses Speech Patterns To Predict 
Alzheimer’s Disease With Astonishing Accuracy

A machine learning model created by Boston University can predict Alzheimer’s disease progression in patients with mild cognitive impairment by analyzing speech, with an accuracy of 78.5%. This tool leverages data from the Framingham Heart Study and offers a more accessible approach to diagnosing Alzheimer’s, potentially allowing for earlier interventions.

Trying to figure out whether someone has Alzheimer’s disease usually involves a battery of assessments—interviews, brain imaging, blood and cerebrospinal fluid tests. But, by then, it’s probably already too late: memories have started slipping away, long established personality traits have begun subtly shifting. If caught early, new pioneering treatments can slow the disease’s remorseless progression, but there’s no surefire way to predict who will develop the dementia associated with Alzheimer’s.

Their model can predict, with an accuracy rate of 78.5%, whether someone with mild cognitive impairment is likely to remain stable over the next six years—or fall into the dementia associated with Alzheimer’s disease. While allowing clinicians to peer into the future and make earlier diagnoses, the researchers say their work could also help make cognitive impairment screening more accessible by automating parts of the process—no expensive lab tests, imaging exams, or even office visits required. The model is powered by machine learning, a subset of AI where computer scientists teach a program to independently analyze data.

According to Alzheimer’s Disease International, the majority of people with dementia worldwide never receive a formal diagnosis, leaving them shut off from treatment and care.

Rhoda Au, a coauthor on the paper, says AI has the power to create “equal opportunity science and healthcare.” The study builds on the same team’s previous work, where they found AI could accurately detect cognitive impairment using voice recordings.  
“Digital is the new blood,” says Au. “You can collect it, analyze it for what is known today, store it, and reanalyze it for whatever new emerges tomorrow.”

MCI = mild cognitive impairment









By Germaine: Science nerd


Comments

Post a Comment

Popular posts from this blog

That Uplifting Tweet You Just Shared? A Russian Troll Sent It

Image
Here’s what Russia’s 2020 disinformation operations look like, according to two experts on social media and propaganda. By   DARREN LINVILL   &   PATRICK WARREN Internet trolls don’t troll. Not the professionals at least. Professional trolls don’t go on social media to antagonize liberals or belittle conservatives. They are not narrow minded, drunk or angry. They don’t lack basic English language skills. They certainly aren’t  “somebody sitting on their bed that weighs 400 pounds,”  as the president once put it. Your stereotypical trolls do exist on social media, but the amateurs aren’t a threat to Western democracy. Professional trolls, on the other hand, are the tip of the spear in the new digital, ideological battleground. To combat the threat they pose, we must first understand them — and take them seriously. MORE: https://www.rollingstone.com/politics/politics-features/russia-troll-2020-election-interference-twitter-916482/ P...
Read more

The Nightmare Scenario That Keeps Election Lawyers Up At Night -- And Could Hand Trump A Second Term

Image
The Nightmare Scenario That Keeps Election Lawyers Up At Night -- And Could Hand Trump A Second Term   (Photo: Photos: Getty Images) Americans will almost certainly go to bed on Nov. 3 without knowing who won the presidential election. Since millions of people will vote by mail, constraints on time and resources will slow ballot counting into potentially a weeks-long process. Voting patterns suggest it’s likely that President  Donald Trump  could end Election Day in the lead in certain key states, only to be overtaken by Democratic opponent  Joe Biden  when more votes are tallied.  This could create a nightmare scenario during the three months stretching from Election Day to the Jan. 20 inauguration: a battle on the state and congressional level over who is the legitimate winner. This could include Congress reconvening on Jan. 6, presided over by Vice President  Mike Pence , with no consensus over its potential role in choosing the next president. This...
Read more

When Life Hands You Lemons

Image
 It's like this folks: sometimes your mind goes blank. Happens more than usual with me of course. Take this forum for instance. I wanted to post something today and couldn't think of a topic. At least not a non-sexual one and we know sexual ones lately have been frowned upon.  Then along comes one of our regulars and complains about something. Well, they might not say it was a complaint, just an observation, but it was a complaint. A minor one. It has something to do about being able to make something out of nothing. Sort of like me this morning, how to make a thread out of an empty head? But they (notice I am using the correct pronoun, not saying he or she), by what they said got me thinking. What I thought was: Not sure if that is the actual saying, if I remember correctly, I remember the saying going something like this: Umm, no, I don't think the saying I am thinking of goes that way either, maybe ONE of you good folks can remind me how the saying goes. Either way, when...
Read more