A devastating chronic neurodegenerative disease, Alzheimer’s disease (AD) currently affects around 5.5 million people in the United States alone. Causing progressive mental deterioration, it ultimately advances to impact basic bodily functions such as walking and swallowing.
Looking for a way to help, researchers at the University of Bari and Istituto Nazionale di Fisica Nucleare in Italy have developed new machine learning AI technology that may help identify Alzheimer’s a decade before doctors usually can, by way of non-invasive MRI brain scans. An early diagnosis — before any of the symptoms a doctor might recognize become apparent — could give patients a chance to make changes to their lifestyle which may slow Alzheimer’s progression.
“We used publicly available data, consisting of 67 brain MRI scans from the Alzheimer’s Disease Neuroimaging Initiative, including healthy controls and AD patients,” Nicola Amoroso, one of the lead researchers on the project, told Digital Trends. “We used this cohort to feed [our] artificial intelligence, then an independent test of about 148 subjects — including controls,
Alzheimer’s disease and mild cognitive impairment (MCI) subjects — was performed. According to our results, it is possible to distinguish a healthy brain from one with Alzheimer’s with an accuracy of 86 per cent. Crucially, it is also possible to detect the difference between healthy brains and those with MCI with an accuracy of 84 per cent.”
This isn’t the first similar study that involves using cutting-edge technology to help diagnose Alzheimer’s and other neurodegenerative diseases. Researchers at VU University Medical Centre in Amsterdam have also been using MRI scans to try and carry out similar early diagnosis of Alzheimer’s. Another intriguing high tech approach is one being taken at Cedars-Sinai Medical Center, University of Southern California (USC), and University of California, Los Angeles (UCLA), where researchers are working to develop an early diagnosis eye test for Alzheimer’s.
“Our goal is to use our approach for other pathologies,” Nicola Amoroso continued. “In particular, we are now investigating Parkinson’s disease, and preliminary results are really encouraging. It would be very important to support studies and clinical trials to let emerge novel preventive or disease modifying therapies.”
You can read a research paper on the University of Bari’s machine learning project here. With millions of people who could benefit from the research, they have our total support.