AI predicts and detects Dementia
AI predicts and detects dementia
An advanced AI model has been developed to analyze data from cognitive assessments and MRI scans, offering a revolutionary alternative to traditional, costly, and invasive procedures such as PET scans and spinal taps. This innovative tool categorizes patients into three distinct groups: those likely to remain stable, those who may progress slowly, and those at risk of rapid decline. Remarkably, the AI has demonstrated an 82% accuracy rate in identifying cases that will progress to Alzheimer’s and an 81% accuracy rate in recognizing cases that will remain stable. This significant reduction in misdiagnosis rates has been achieved through the validation of the AI’s predictions using six years of follow-up data and extensive testing across memory clinics in various countries, confirming its global applicability.
Why it matters: With a rapidly aging global population, the incidence of dementia is projected to triple over the next 50 years. Early detection is crucial in determining the effectiveness of treatment. AI’s predictive capabilities herald a new era of proactive treatment, potentially transforming the lives of those experiencing cognitive decline.
Dementia represents a substantial global healthcare challenge, affecting over 55 million people worldwide and costing an estimated $820 billion annually. This burden is expected to increase threefold in the next 50 years. Alzheimer’s disease (AD) is the leading cause of dementia, accounting for 60-80% of cases.
Early prediction of AD onset is critical for effective clinical management and treatment. Recent positive phase three clinical trial results emphasize the necessity for early detection to maximize the effectiveness of treatments. However, there is still a lack of effective tools for early dementia diagnosis and prognosis.
Standard memory tests are often not sensitive enough at early disease stages, and most patients do not have access to more specific diagnostic tools like positron emission tomography (PET) scans or lumbar punctures for cerebrospinal fluid biomarkers. These invasive and expensive biomarkers are not part of routine clinical practice, resulting in significant healthcare disparities. Consequently, up to a third of patients may be misdiagnosed, and others may be diagnosed too late for treatments to be effective.
Advancements in analytical techniques are providing a pivotal shift in addressing these challenges by improving early prediction and prognosis of dementia through lower-cost, less invasive assessments.
The success of Artificial Intelligence (AI) models, particularly those based on machine learning (ML) algorithms, in stratifying individuals shows great promise. However, translating these models to clinical practice faces several hurdles. Firstly, ML models developed using research cohort data may not accurately represent the demographics and medical comorbidities found in clinical populations.
Secondly, research cohort data are often rich and structured, whereas real-world data are collected using less sensitive measures and can be incomplete and inconsistent due to the lack of standardized assessment methods across healthcare providers. This discrepancy can lead to ML models failing to generalize effectively when applied to real-world patient data.
Thirdly, many existing ML models focus on stratifying individuals in a cross-sectional manner, without accounting for the need to predict individualized health trajectories. This is crucial for the prognosis of neurodegenerative disorders, which span a continuum from health to disease.
By addressing these limitations and leveraging AI’s predictive power, healthcare providers can significantly enhance early detection and management of Alzheimer’s disease, ultimately improving patient outcomes and reducing the global burden of dementia.
Researchers at West Virginia University have identified a set of diagnostic metabolic biomarkers that can aid in developing artificial intelligence tools for detecting Alzheimer’s disease in its early stages, as well as determining risk factors and potential treatment interventions.
The study, published in the Journal of the Neurological Sciences, aimed to identify which metabolic biomarkers are most relevant to Alzheimer’s disease and then train an AI model to predict the likelihood of whether the disease has or could develop. Scientists chose the deep learning method of AI for this research due to its versatile approach in predicting complex biological phenomena and its capability to handle vast volumes of data and complex algorithms.
“The deep learning method using artificial neural networks, inspired by the layered structure of the brain’s neurons and their computations, has reached unprecedented prediction performance for complex tasks,” said Kesheng Wang, professor in the WVU School of Nursing, who led the study. “Deep learning techniques have proven to be more accurate for Alzheimer’s disease diagnosis compared to conventional machine learning models.”
In medicine, biomarkers are measurable indicators of the severity or presence of a disease. People typically associate biomarkers with numbers on their bloodwork reports, such as cholesterol or glucose levels. Metabolic biomarkers, however, exist in the molecules of cells, tissues, and body fluids, reflecting the interaction between genes and lifestyle factors like diet and environment. At this molecular level, scientists can better understand changes in a person’s health and their risks of developing diseases.
“Alzheimer’s disease may start years or even decades before clinical symptoms appear. Therefore, it is crucial to identify predictive biomarkers in the preclinical stage to develop strategies that can prevent the progression of the disease,” Wang emphasized. Early detection of Alzheimer’s disease is also vital for drug development and application, as well as for diagnostic and therapeutic approaches to prevent loss of function and diminished longevity.
For the study, data from the Alzheimer’s Disease Neuroimaging Initiative was obtained from 78 people diagnosed with Alzheimer’s disease and 99 people with normal cognitive function, ranging in age from 75 to 82. Using LASSO software, researchers imported 150 metabolic biomarkers and identified 21 as most relevant to Alzheimer’s disease.
“The metabolites are part of the glucose, amino acid, and lipid metabolisms,” Wang explained. “Some of these metabolites correlate with clinical biomarkers, such as plaques, cognitive measures, and hippocampus volume associated with Alzheimer’s patients.” The hippocampus, often the first area of the brain damaged by Alzheimer’s disease, shows shrinkage in affected individuals.
Researchers then tested multiple deep learning models until they built one that achieved the highest accuracy for assessment. Despite the promising results, Wang noted that studies using deep learning for detecting Alzheimer’s disease are still in the early stages, and further research is needed. He and his team are currently working on a project to integrate data from proteins and metabolism using deep learning methods.
“The metabolic basis of Alzheimer’s disease is still poorly understood, and the relationships between systemic abnormalities in metabolism and Alzheimer’s disease pathogenesis are unclear,” Wang said. “This study shows there is potential to identify metabolic biomarkers that are predictive of Alzheimer’s disease diagnosis and progression.”
Read the full article – HERE
AI Predicts Alzheimer’s Disease Years Before Symptoms Appear
UC San Francisco scientists have developed a method to predict Alzheimer’s disease up to seven years before symptoms emerge by analyzing patient records with machine learning. The conditions that most significantly influenced the prediction were high cholesterol and, for women, osteoporosis.
This research highlights the potential of artificial intelligence (AI) to identify patterns in clinical data that can be used to examine large genetic databases, thereby uncovering the underlying factors driving disease risk. The researchers aspire that this advancement will eventually expedite the diagnosis and treatment of Alzheimer’s and other complex diseases.
“This is a first step towards using AI on routine clinical data, not only to identify risk as early as possible but also to understand the biology behind it,” said the study’s lead author, Alice Tang, an MD/PhD student in the Sirota Lab at UCSF. “The power of this AI approach comes from identifying risk based on combinations of diseases.
Read the full article HERE
Share this post :
Living Longer, Living Better: Stories of Extraordinary Lives in the Marginal Decade
Apollo Neuro: Exploring Its Impact on Mental Health and Well-being
Lion’s Mane Mushrooms
