While artificial intelligence (AI) and machine learning have the potential to identify patients at risk of dementia, significant improvements are needed before their use in clinical practice, according to the results of a study published in the journal Alzheimer’s and dementia.
As the prevalence of dementia increases worldwide due to increased life expectancy, researchers are looking for ways to delay its onset. Research in this area has focused on treatment and prevention at an early stage, as there is currently no definitive cure for dementia. Additionally, the literature regarding the impact of treating dementia risk factors to help slow the progression of dementia remains weak.
For the study, researchers evaluated how AI and machine learning could help identify risk factors and potential interventions for dementia patients. Machine learning is a subset of AI focused on models and algorithms that allow computers to make decisions or predictions. Machine learning and other risk profiling tools, such as logistic regression, support vector machines, random forests, and gradient-enhanced trees, are not widely used in dementia prevention.
These models can identify potential causes of risk factors and help identify patient variables to facilitate clinical trial recruitment, reduce trial costs, and accelerate treatment discovery.
“
Multidisciplinary collaboration is needed to harness the potential of ML, maximize the use of available resources and access to data, and improve traditional approaches to advance dementia prevention research.
In the Lancet Commission on Dementia Prevention, Intervention and Care, first published in 2017, some risk factors identified for the development of dementia included fewer years of education, smoking, depressionphysical inactivity, social isolation and type 2 diabetes. In the 2020 updated report, the Lancet Commission added alcohol, traumatic brain injury (TBI) and air pollution as factors of risk.
Although these factors have been identified by several systematic reviews, it is important to consider the relationship between these risk factors throughout an individual’s lifespan. Machine learning has the ability to use longitudinal data to understand disease trajectories and heterogeneity.
An example of a machine learning model used for dementia is the Cardiovascular Risk Factors, Aging, and Dementia (CAIDE) model, which uses patient data and midlife risk factors to predict dementia risk in 20 years with moderate precision (area under the curve). (AUC), 0.77; 95% CI, 0.71-0.83).
A recent study showed that using a variety of machine learning algorithms is superior to using common dementia risk models alone, such as the CAIDE model, in predicting 2-year dementia risk ; However, the clinical utility of these models is questioned as they are unlikely to contribute to the prevention of dementia progression since neurodegeneration will have already occurred.
The use of multimodal data in the CAIDE model more closely reflects clinical practice, while the use of a variety of machine learning models can increase sensitivity and specificity, potentially at the expense of clinical translatability.
Some major current challenges in using machine learning to predict dementia risk include: risk These factors can exert effects through a variety of different biological pathways, which can make the identification of mechanistic targets difficult.
Limitations of the study include the inability to establish causality using machine learning models, biased machine learning models due to lack of inclusion of minority populations, and difficulties in understanding or interpret the models.
“Multidisciplinary collaboration is needed to harness the potential of ML (machine learning), maximize the use of available resources and access to data, and improve traditional approaches to advance dementia prevention research,” the researchers wrote.
Disclosure: Some study authors have declared affiliations with biotechnology, pharmaceutical, and/or device companies. Please see the original reference for a complete list of author disclosures.
This article was originally published on Neurology Advisor
