If there’s one area where AI can truly have an unprecedented positive impact, it is healthcare — especially when it comes to the diagnosis and treatment of currently incurable diseases such as dementia.
The condition affects over 55 million people worldwide, with nearly 10 million new cases every year. Dementia’s most common type, Alzheimer’s disease, contributes to 60%-70% of all cases.
Globally, the cost of the disorder on healthcare systems reached $1.3tn in 2019.
The psychological cost is even higher. Suffering from the disease may trigger depression and anxiety. Let alone the indescribable emotional impact of seeing a loved one “disappear” right in front of your eyes.
Fortunately, advanced technologies offer a much-needed ray of hope — and European universities are rising to the task. Here are three ambitious initiatives that could provide us with new weapons in the fight against dementia:
Predicting Alzheimer’s development and progression
Researchers at the University of Cambridge have developed a machine learning model that can predict if and how fast an individual with mild memory and thinking problems will develop Alzheimer’s.
The team built the model using cognitive tests and MRI scans showing grey matter atrophy (i.e. the death of nerve cells in the brain) from 400 patients who were part of a research group in the US.
They then tested the model on an additional amount of data from 600 more participants in the US group and 900 individuals from memory clinics in the UK and Singapore.
The algorithm accurately identified those who would develop Alzheimer’s within three years in 82% of cases, and those who wouldn’t in 81% of cases.
It was also able to track the disease’s progression rate, providing valuable insights into the most suitable course of treatment. This way, it can reduce the need for costly and invasive testing methods such as positron emission tomography (PET) scans or lumbar puncture (also known as spinal tap).
The scientists validated the AI tool’s predictions with follow-up data over the course of six years. They suggest that their solution is three times more accurate at predicting Alzheimer’s progression than clinical diagnosis or clinical markers such as grey matter atrophy and cognitive scores.
“If we’re going to tackle the growing health challenge presented by dementia, we will need better tools for identifying and intervening at the earliest possible stage,” said Professor Zoe Kourtzi, senior author of the study.
“Our vision is to scale up our AI tool to help clinicians assign the right person at the right time to the right diagnostic and treatment pathway.”
Early dementia diagnosis
Backed with €14mn by the EU’s Horizon Programme, the AI-Mind project is developing two artificial intelligence tools that can enable the early diagnosis of dementia.
It specifically targets the mild cognitive impairment (MCI) stage, where there are no structural brain defects and intervention is still possible.
To achieve this, the 13 partners behind the project are building the AI-Mind Connector and AI-Mind Predictor.
The Connector analyses brain images from EEG data to detect early signs of cognitive changes that could lead to dementia. The Predictor combines this data with cognitive tests and blood analysis to assess the risk of the disorder with a >95% accuracy.
Both these tools will be integrated into a cloud-based diagnostics platform that can support health professionals.
The project’s ultimate goal is an ambitious one: reducing diagnosis time from between two to five years to a single week. This way, it hopes to increase the “dementia-free” period for MCI patients.
AI-Mind kicked off in 2021 and will run until 2026. Among its partners are seven European universities, including Aalto University in Finland, Tallinn University in Estonia, and the Radboud University Medical Centre in the Netherlands.
Tracking down protein clumps
Another use case for AI in battling dementia is deepening our understanding of protein clumps in the body.
For our bodies to function, billions of interactions between proteins and other molecules are taking place inside the cells. But when errors occur in these processes, proteins can clump together and malfunction, leading among others to neurodegenerative disorders such as Alzheimer, among others.
Researchers from the University of Copenhagen have developed an AI algorithm that can spot protein clumping down to a billionth of a metre in microscopy images.
The algorithm can also count the clumps, classify them by shape and size, and monitor how they change over time. This way, it can help scientists understand why these clumps form and, in turn, enable the discovery of new drugs and therapies.
According to the team, the tool automates in a few minutes a process that would take researchers several weeks.
The machine learning algorithm is freely available on the internet as an open-source model.
“As other researchers around the world begin to deploy the tool, it will help create a large library of molecule and protein structures related to various disorders and biology in general,” said Nikos Hatzakis, co-author of the study.
“This will allow us to better understand diseases and try to stop them.”
Get the TNW newsletter
Get the most important tech news in your inbox each week.