Abstract: A new AI tool predicts Alzheimer's progression with 82% accuracy using cognitive tests and MRI scans, outperforming existing methods. This tool can reduce the need for expensive tests and improve early intervention.
Alzheimer's disease is the leading cause of dementia, affecting more than 55 million people worldwide.
Important facts:
- The AI tool correctly identified Alzheimer's progression in 82% of cases.
- It uses non-invasive, low-cost data for predictions.
- It can divide patients into groups based on the speed of disease progression.
Source: University of Cambridge
Cambridge scientists have developed an artificial intelligence tool that can predict in four out of five cases whether people with early symptoms of dementia will stay stable or develop Alzheimer's disease.
The team says the new approach could reduce the need for invasive and expensive diagnostic tests while improving early treatment outcomes, allowing interventions such as lifestyle changes or new drugs to work best. .
Dementia is a major global health care challenge, affecting more than 55 million people worldwide with an annual cost of $820 billion. The number of cases is expected to nearly triple over the next 50 years.
The leading cause of dementia is Alzheimer's disease, which accounts for 60-80% of cases. Early detection is critical because this is when treatment is likely to be most effective, yet early detection and diagnosis of dementia requires invasive or expensive tests such as positron emission tomography (PET) scans or lumbar punctures. Can't be valid without usage, which they aren't. Available at all memory clinics.
As a result, up to a third of patients may be misdiagnosed and others may be diagnosed too late for treatment to be effective.
A team led by scientists from the University of Cambridge's Department of Psychology has developed a machine learning model capable of predicting whether a person with mild memory and thinking problems will progress to Alzheimer's disease.
In research published in eClinical Medicinethey show that it is more accurate than existing clinical diagnostic tools.
To build their model, the researchers used data from routinely collected, noninvasive, and low-cost patients—cognitive tests and structural MRI scans that show gray matter atrophy—from 400 patients. from more than 100 individuals who were part of a research group in the United States.
They then tested the model using real-world patient data from another 600 participants in the US cohort and, importantly, longitudinal data from 900 people from memory clinics in the UK and Singapore.
The algorithm was able to distinguish between people with stable mild cognitive impairment and those who progressed to Alzheimer's disease within a three-year period. It was able to correctly identify individuals who developed Alzheimer's in 82% of cases and correctly identify those who did not in 81% of cases using only cognitive tests and MRI scans.
The algorithm was nearly three times more accurate at predicting Alzheimer's progression than the current standard of care. i.e. standardized clinical markers (such as gray matter atrophy or cognitive scores) or clinical assessment. This suggests that the model can significantly reduce misdiagnosis.
The model also allowed researchers to divide people with Alzheimer's disease into three groups using data from each person's first visit to a memory clinic: those whose symptoms would remain stable (about 50 percent); participants), those who will gradually progress to Alzheimer's. 35%) and those who will develop more rapidly (the remaining 15%).
These predictions were confirmed when looking at follow-up data over six years. This is important because it can help identify people who are at an early enough stage to benefit from new treatments, as well as those who need closer monitoring because they have The condition is likely to deteriorate rapidly.
Importantly, the 50% of people who have symptoms similar to memory loss but remain stable, would be better directed to a different medical pathway because their symptoms may be due to causes other than dementia. are, such as anxiety or depression.
Senior author Professor Zoe Kortzi, from the Department of Psychology at the University of Cambridge, said: “We have created a tool that, despite only using data from cognitive tests and MRI scans, is currently able to predict Methods are more sensitive to whether or not someone will progress from mild symptoms to Alzheimer's—and if so, whether the progression will be rapid or slow.
“This has the potential to significantly improve patient well-being, showing us who needs the closest care, while addressing the patient distress we predict. “That they will remain stable. At a time of severe pressure on health care resources, it will also help eliminate the need for unnecessary invasive and expensive diagnostic tests.”
While the researchers tested the algorithm on data from one research group, it was validated using independent data that included nearly 900 people who attended memory clinics in the UK and Singapore.
In the UK, patients were recruited using quantitative MRI in the NHS Memory Clinics Study (QMIN-MC) led by study co-author Dr Timothy Rittman at Cambridge University Hospitals NHS Trust and Cambridgeshire and Peterborough NHS Foundation Trust (CPFT). of
This shows that it should be applicable in a real-world patient, clinical setting, the researchers say.
Dr Ben Underwood, Honorary Consultant Psychiatrist at CPFT and Assistant Professor in the Department of Psychiatry at Cambridge University, said, “Memory problems are common as we age. Uncertainty about whether these may be the first signs of dementia can be very worrying for people and their families, as well as frustrating for doctors who prefer to give definitive answers. .
“The fact that we may be able to reduce this uncertainty with the information we already have is exciting and will become even more important as new treatments emerge.”
“AI models are only as good as the data they're trained on. To ensure we have the potential to be adopted in healthcare settings, we not only Trained and tested on data routinely collected from research groups but also from patients in real memory clinics, this shows that it will generalize to real-world settings.
The team now hopes to extend their model to other forms of dementia, such as vascular dementia and frontotemporal dementia, and using different types of data, such as markers from blood tests.
Professor Cortzi added, “If we are going to tackle the growing health challenge presented by dementia, we will need better tools to identify and intervene at the earliest possible stage. .
“Our vision is to expand our AI tool to help clinicians assign the right person to the right diagnostic and treatment pathway at the right time. Our tool can help match the right patients to clinical trials, modify disease accelerates the discovery of new drugs to treat
About this AI and Alzheimer's disease research news
the author: Ben Underwood
Source: University of Cambridge
contact: Ben Underwood – University of Cambridge
Image: Photo credit to Neuroscience News.
Original research: Open access.
“Robust and Interpretable AI Guided Markers for Early Prediction of Dementia in Real-World Clinical Settings” by Ben Underwood et al. eClinical Medicine
Abstract
Robust and interpretable AI guidance markers for early prediction of dementia in real-world clinical settings
background
Early prediction of dementia has major implications for clinical management and patient outcomes. Nevertheless, we still lack sensitive tools to classify patients early, resulting in patients being underdiagnosed or misdiagnosed. Despite the rapid expansion of machine learning models for dementia prediction, limited model interpretability and generalizability hinder translation to the clinic.
methods
We construct a robust and interpretable predictive prognostic model (PPM) and demonstrate its clinical utility using real-world, routinely collected, noninvasive, and low-cost (cognitive tests, structural MRI) patient data. Correct using To increase scalability and generalizability in the clinic, we: 1) train PPM with clinically relevant predictors (cognitive tests, gray matter atrophy) that are common in research and clinical cohorts, 2) independent multi- Centers test PPM's predictions with real-world data. Memory clinics from all countries (UK, Singapore).
Results
PPM strongly predicts (accuracy: 81.66%, AUC: 0.84, sensitivity: 82.38%, specificity: 80.94%) whether patients with early stage disease (MCI) will remain stable or develop Alzheimer's disease (AD). Will move towards. PPM generalizes real-world patient data from research in memory clinics and its predictions are validated against longitudinal clinical outcomes. PPM allows us to derive an individual AI-guided multimodal marker (i.e., predictive prognostic index) that correlates with standard clinical markers (gray matter atrophy, cognitive scores; PPM-derived markers: hazard ratio = 3.42, p = 3.40, p ) or clinical diagnosis (PPM-derived marker: hazard ratio = 2.84, p < 0.01), reducing misdiagnosis.
Interpretation
Our results provide evidence for a robust and definable clinical AI-guided marker for early prediction of dementia that is validated against longitudinal, multicenter patient data across countries, and for adoption in clinical practice. Has a strong ability.
Funding
Wellcome Trust, Royal Society, Alzheimer's Research UK, Alzheimer's Drug Discovery Foundation Diagnostic Accelerator, Alan Turing Institute.