Tapping AI to spot AD

Haryana-based research team develops framework for Alzheimer’s disease detection

Updated - January 13, 2019 10:32 am IST

Published - January 13, 2019 12:02 am IST

Picture this about Alzheimer’s disease. It is an affliction of progressive forgetfulness. Things get misplaced and one tends to locate them — for example, car keys being found in the washroom — in unusual places. It may even be hard to track a conversation.

Sorry, what are we talking about? By the time specialists advise a formal diagnosis, it is usually too late. And this is usually after a detailed examination by a medical expert complete with Magnetic Resonance Imaging (MRI) scans and several tests for cognitive impairment.

For years now, researchers have been on the hunt for a way to detect Alzheimer’s in its infancy. Some years ago, researchers started to intensely focus on the hippocampus, a region of the brain critical to forming memories.

In a study published in Alzheimer’s Research and Therapy , in October 2015, researchers established a correlation between a rapid loss of tissue in the hippocampus and an early onset of the disease.

Says Dr. Aaron Bonner-Jackson, one of the researchers, “Overall, we found strong relationships between the memory measures and hippocampal volumes in the sample as a whole, suggesting that these measures may serve as effective indicators of hippocampal size. In particular, performance on the measure of non-verbal memory was most closely related to hippocampal size among the larger sample.” Dr. Bonner-Jackson explained his team’s findings in a blog on BioMed Central. He added, “When we looked at the sample of individuals with Mild Cognitive Impairment, a similar pattern emerged. Specifically, the non-verbal memory task showed much stronger associations with hippocampal volumes than a comparable verbal memory task.”

But insights gleaned from testing hippocampal volumes, while useful, are still too late in the day. There is no effective treatment for the disease and the only hope is that if it is caught early enough, some medical intervention could be made.

Tracking glutathione

In recent years scientists have been intrigued by a low-profile chemical called glutathione. Other than its key role in regulating the growth of cells, glutathione is also one of the body’s core antioxidants. It scavenges on molecules that may whet oxidation (which means an acceleration in cellular and DNA damage).

Therefore, many researchers argue, low glutathione levels in the brain could mean that many molecules responsible for oxidation may be running amok and accelerating cell damage. This could mean they are eating away at the hippocampus and thereby accelerating Alzheimer’s disease.

Tracking glutathione levels in the brain is an extremely involved process and requires MRI as well as specialist analysis. However a team at the National Brain Research Centre (NBRC), Haryana, says that it has got a framework in place that correlates glutathione levels in the brain, pictures of the hippocampus (taken using MRI) and performance on certain standard cognitive tests (for example, memory and motor-coordination). This can then be used to develop an algorithm to calculate the odds of an early onset of Alzheimer’s.

Says Prof. Pravat K. Mandal, lead researcher and senior professor at the NBRC, “What we’re looking at is to be able to determine in half-hour maximum, via a diagnostic test, the patient’s risk for Alzheimer’s.”

However, what Prof. Mandal currently lacks is the troves of data from patients that ‘machine learning’ systems need to draw correlations — and maybe reveal hidden relationships — on how early onset of Alzheimer’s could be detected.

In a paper to be published in the Frontiers in Neurology , Prof. Mandal and his colleagues say, “Large-scale data analysis using brain imaging, metabolic and neuropsychological score provide information about disease progression and identify early diagnostic biomarker. Hence, conceptualisation of analytics is an important step and will help physicists, clinicians, engineers work hand-in-hand to help [develop] an effective tool for early diagnosis or prediction of AD [Alzheimer’s disease].”

Other teams elsewhere are also tapping artificial intelligence (AI) to predict Alzheimer’s. In November 2018, researchers reported in Radiology that a deep learning algorithm developed using imaging data from more than 1,000 Alzheimer’s patients could, reportedly, accurately predict the presence of the disease more than six years before a doctor finalised the diagnosis. A study in 2018 by the Centre for Internet & Society India claimed that AI could help add $957 billion to the Indian economy by 2035, highlighting its potential.


0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in


Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.