Machine learning in neurodegenerative disease research: a growing trend

In a recent review posted to the medRxiv* preprint server, scientists explore the utility of machine learning methods in the sphere of neurodegenerative disease diagnosis, prognosis, and treatment effect prediction.

Study: Using machine learning methods in neurodegenerative disease research: A scoping review. Image Credit: sfam_photo / Shutterstock.com

*Vital notice: medRxiv publishes preliminary scientific reports that are usually not peer-reviewed and, due to this fact, mustn’t be considered conclusive, guide clinical practice/health-related behavior, or treated as established information.

Background

Neurodegenerative diseases are detrimental age-related pathological conditions related to progressive deterioration of the neuronal network within the central and peripheral nervous systems. Consequently, all neurodegenerative diseases are related to progressively disabling symptoms that ultimately lead to finish lack of autonomy and death.

Probably the most common neurogenerative diseases include Alzheimer’s disease, Parkinson’s disease, multiple sclerosis, amyotrophic lateral sclerosis, and Huntington’s disease.

In the USA, each Alzheimer’s disease and Parkinson’s disease are essentially the most common neurodegenerative diseases. Current estimates indicate that as much as 6.2 million persons are living with Alzheimer’s disease within the U.S., whereas Parkinson’s disease currently affects about a million Americans. As life expectancy increases in many countries throughout the world, researchers predict that the prevalence of those neurodegenerative diseases can even rise.

To enhance the management of those incurable diseases, it’s important to know disease pathogenesis, develop accurate diagnostic and prognostic tools, and discover targeted therapies. Using machine learning methods is increasing in the sphere of neurodegenerative disease research for rapidly and accurately analyzing disease-related data, which is important for supporting diagnostic and therapeutic innovations.

In the present scoping review, scientists explore the utility of machine learning methods within the study of the five most prevalent neurodegenerative diseases, including Alzheimer’s disease, multiple sclerosis, amyotrophic lateral sclerosis, Parkinson’s disease, and Huntington’s disease.

Study design

Various scientific databases were searched to discover studies that utilized machine learning methods for the diagnosis, prognosis, and treatment prediction of 5 neurodegenerative diseases. All studies published between January 2016 and December 2020 were included within the evaluation.

A complete of 4,471 studies were screened, 1,485 of which were ultimately included in the ultimate evaluation. The data extracted from each study included kind of neurodegenerative disease, publication yr, sample size, machine learning algorithm data type, primary clinical goal, and machine learning method type. Each qualitative and quantitative analyses of the study results were conducted.   

The growing use of machine learning methods

A gradual increase in the usage of machine learning methods in neurodegenerative diseases was observed over time. More specifically, the variety of studies using these methods increased from 172 in 2016 to 490 in 2020, thus reflecting a 185% increase within the incorporation of this technology. Alzheimer’s disease and Parkinson’s disease were essentially the most studied neurodegenerative diseases using machine learning methods.

In the chosen studies, imaging was essentially the most commonly analyzed data type, followed by functional, clinical, biospecimen, genetic, electrophysiological, and molecular analyses. Imaging and functional data were essentially the most commonly used data types in Alzheimer’s disease and Parkinson’s disease, respectively. About 68% of imaging data was related to Alzheimer’s disease and 76% of functional data was related to Parkinson’s disease.

Regarding primary clinical goals, machine learning methods were most steadily used for disease diagnosis, followed by disease prognosis and prediction of treatment effects. Imaging data remained essentially the most commonly used data type for disease diagnosis and prognosis. For the prediction of treatment effect, functional data were essentially the most commonly used data type.

A complete of two,734 sorts of machine learning methods were utilized in the chosen studies. Amongst these methods, support vector machine, random forest, and convolutional neural network were most steadily noted. As well as, 322 unique methods were identified within the review.  

Significance

The present scoping review indicates a rise in the applying of machine learning methods in neurodegenerative disease research. The scientists explain that the recognition of those methods is increasing to enhance the clinical course of those detrimental diseases.

Although certain treatments are currently available to alleviate among the physical and mental symptoms related to neurodegenerative diseases, there stays a scarcity of therapies able to slowing the progression of neuronal death. Thus, there stays an urgent need to extend the applying of machine learning methods to discover prognostic biomarkers and discover novel therapeutics for the treatment of neurodegenerative diseases.

*Vital notice: medRxiv publishes preliminary scientific reports that are usually not peer-reviewed and, due to this fact, mustn’t be considered conclusive, guide clinical practice/health-related behavior, or treated as established information.