Machine learning in neurology: what neurologists can learn from machines and vice versa
Artificial intelligence is increasingly becoming a part of everyday life. This raises the question whether clinical neurology can benefit from these novel methods to increase diagnostic accuracy. Several recent studies have used machine learning classifiers to predict whether subjects suffer from a neurological disorder. This article discusses whether these methods are ready to make their entrance into clinical practice. The underlying principles of classification will be explored, as well as the potential pitfalls. Strengths of machine learning methods are that they are unbiased and very sensitive to patterns emerging from small changes spread across a large number of variables. Potential pitfalls are that building reliable classifiers requires large amounts of well-selected data and extensive validation. Currently, machine learning classifiers offer neurologists a new diagnostic tool which can aid in the diagnosis of cases with a high degree of uncertainty.
KeywordsArtificial intelligence Machine learning Support vector machines Diagnostic accuracy Classification
The author thanks Bruno Bergmans, M.D., Ph.D., for useful comments on prior versions of this paper.
RB is a postdoctoral fellow of the Research Foundation Flanders (F.W.O.)
Compliance with ethical standards
Conflicts of interest
Dr. Bruffaerts reports no disclosures.
- 2.Rathore S, Habes M, Iftikhar MA et al (2017) A review on neuroimaging-based classification studies and associated feature extraction methods for Alzheimer’s disease and its prodromal stages. NeuroImage 155:530–548. https://doi.org/10.1016/j.neuroimage.2017.03.057 CrossRefPubMedPubMedCentralGoogle Scholar
- 6.Duncan J (2010) How intelligence happens. Yale University Press, New HavenGoogle Scholar