Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Using AI to Provide Early Diagnoses of Parkinson's

DZone's Guide to

Using AI to Provide Early Diagnoses of Parkinson's

Researchers have been figuring out how to detect Parkinson's early on. The algorithm they created is capable of predicting those with Parkinson's with an accuracy of 73%.

· AI Zone
Free Resource

Find out how AI-Fueled APIs from Neura can make interesting products more exciting and engaging. 

I've covered a wide range of projects in the past year or so that utilize AI to trawl medical data to provide earlier diagnoses than would otherwise be possible. The latest example comes via a study from Washington University in St. Louis, whose algorithm scans patient medical records for signs of Parkinson's developing.

This is a crucial breakthrough as it's a condition that is currently impossible to predict until symptoms begin to emerge. The authors wanted to change that, and analyzed over 200,000 Medicare claims to try and provide earlier diagnoses.

"Using this algorithm, electronic medical records could be scanned and physicians could be alerted to the potential that their patients may need to be evaluated for Parkinson's disease," the authors say.

They continue:

"One of the most interesting findings is that people who are going to develop Parkinson's have medical histories that are notably different from those who don't develop the disease. This suggests there are lifelong differences that may permit identification of those likely to develop the disease decades before onset."

Early Signs

The authors believe that people who go on to develop Parkinson's follow a pattern of early tests and diagnoses that can be looked for in their medical records.

Of the population analyzed, nearly 90,000 had been officially diagnosed with Parkinson's. A control group of nearly 120,000 people of a similar age but with no such diagnosis was used to try and spot any differences in the preceding years.

The algorithm was capable of predicting those with Parkinson's with an accuracy of 73%. The paper reveals that many of the claims codes that helped predict the disease referred to problems already known to be associated with Parkinson's such as tremors, posture abnormalities, psychiatric or cognitive dysfunction, gastrointestinal problems, sleep disturbances, fatigue, and trauma, including falls. Other factors associated with the disease included weight loss and multiple forms of chronic kidney disease.

"We want to be able to catch people as early as possible," the authors say. "If I know someone may be in the beginning stages of Parkinson's disease, I would evaluate their gait and balance to determine if they have unrecognized impairments that could lead to falls, or whether they have difficulty performing activities of daily living. Either of these scenarios may benefit from treatment."

In the year and a half prior to diagnosis, patients would undertake a flurry of medical tests on frequent visits to the doctor. Most of these were due to a worsening of symptoms, with an array of tests undertaken to try and find the cause. The team believes their system could reduce the need for such tests and speed up the diagnosis.

To find out how AI-Fueled APIs can increase engagement and retention, download Six Ways to Boost Engagement for Your IoT Device or App with AI today.

Topics:
ai ,healthcare ,algorithm ,predictive analytics

Published at DZone with permission of Adi Gaskell, DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}