Analysis of MRI Data in Diagnostic Neuroradiology

Magnetic resonance imaging (MRI) is a noninvasive imaging tool for neuro-radiological diagnosis. Numerous concepts of automated MRI analysis andthe use of machine learning have been proposed to assist diagnosis and prog-nosis. While these academic innovations have proven effective in principlewithin controlled environments, their application to clinical practice hasfaced unmet requirements, such as the ability to perform reliably across aheterogeneous population, to work robustly in the presence of comorbidi-ties, and to be invariant to scanner hardware and image quality. The lack ofrealistic confidence bounds and the inability to handle missing data have alsoreduced the application of most of these methods outside of academic stud-ies. Mastering the complex challenges in the diagnostic process may helpresearchers discover novel biological constructs in multimodal data and im-prove stratification for clinical trials, paving the way for precision medicine.This review presents the state of the art of computerized brain MRI analysisfor diagnostic purposes. We critically evaluate the current clinical usefulnessof the methods and highlight challenges and future perspectives of the field.