Study Identifies Key Factors Associated With Dementia Pathogenesis

Share this content:
Independent predictors of dementia have been identified that may potentially improve diagnosis and treatment.
Independent predictors of dementia have been identified that may potentially improve diagnosis and treatment.

VANCOUVER, British Columbia — Recent research has identified independent predictors of dementia to include age at diagnosis, transient ischemic attack and stroke status, and years of education, with vascular factors playing a greater role in disease pathogenesis than previously thought.

The findings were presented at the 2016 annual meeting of the American Academy of Neurology (AAN).

In the abstract, the researchers wrote that dementia encompasses a broad set of neurologic diseases, producing progressive declines in memory and/or thinking faculties, sometimes alongside personality and emotional disturbances.

“Worldwide, approximately 35.6 million people have dementia, and this number is only expected to grow due to an aging population,” they wrote. “Unfortunately, it is exceedingly difficult to predict who will develop dementia, let alone what type. This makes it difficult to mobilize various preventive strategies supported by mounting evidence.”

The aim of the study was to use a data-driven, machine-learning approach to develop better predictive algorithms for outcomes in dementia before and after diagnosis, and to highlight potentially underappreciated variables in its pathogenesis.

The study included data from the National Alzheimer's Coordinating Center Database, which totaled 7298 patients and featured demographic, genetic, and intermediate clinical information.

Primary outcomes were executive function, memory function, the Mini-Mental State Examination (MMSE), and Braak staging, all at death.

The researchers applied feature selection techniques such as principal component analysis, recursive feature selection, and extra trees classification to the dataset, and then applied 16 distinct machine-learning classifiers to the data in these 4 different forms, which included one unmodified.

Data revealed that age at diagnosis of cognitive decline, status of transient ischemic attack and stroke, and years of education were the most important independent variables.

In addition, researchers reported that at best, using unmodified data and a k-nearest neighbors classifier, accurate predictions of executive function were achieved 71.57% of the time, whereas memory function could be accurately predicted 63.73% of the time, MMSE results 62.7% of the time, and Braak stage 32.58% of the time.

“These results suggest that vascular factors may play a greater role in dementia pathogenesis than currently thought,” the researchers concluded. “Furthermore, using this method we were able to achieve prediction accuracies that compare favorably with the existing literature.”

Click here for more coverage from the 68th Annual Meeting of the American Academy of Neurology, April 15-21, 2016, in Vancouver, British Columbia, Canada.

Reference

Morgenstern J, Daley M, Hachinski V. Prognostic risk profiles for dementia: a machine learning approach. Presented at: The 68th Annual Meeting of the American Academy of Neurology; April 15-21, 2016; Vancouver, Canada. Abstract P1.091.

You must be a registered member of Neurology Advisor to post a comment.