Radiological classification of dementia from anatomical MRI assisted by machine learning-derived maps.

BACKGROUND AND PURPOSE Many artificial intelligence tools are currently being developed to assist diagnosis of dementia from magnetic resonance imaging (MRI). However, these tools have so far been difficult to integrate in the clinical routine workflow. In this work, we propose a new simple way to use them and assess their utility for improving diagnostic accuracy. MATERIALS AND METHODS We studied 34 patients with early-onset Alzheimer's disease (EOAD), 49 with late-onset AD (LOAD), 39 with frontotemporal dementia (FTD) and 24 with depression from the pre-existing cohort CLIN-AD. Support vector machine (SVM) automatic classifiers using 3D T1 MRI were trained to distinguish: LOAD vs Depression, FTD vs LOAD, EOAD vs Depression, EOAD vs FTD. We extracted SVM weight maps, which are tridimensional representations of discriminant atrophy patterns used by the classifier to take its decisions and we printed posters of these maps. Four radiologists (2 senior neuroradiologists and 2 unspecialized junior radiologists) performed a visual classification of the 4 diagnostic pairs using 3D T1 MRI. Classifications were performed twice: first with standard radiological reading and then using SVM weight maps as a guide. RESULTS Diagnostic performance was significantly improved by the use of the weight maps for the two junior radiologists in the case of FTD vs EOAD. Improvement was over 10 points of diagnostic accuracy. CONCLUSION This tool can improve the diagnostic accuracy of junior radiologists and could be integrated in the clinical routine workflow.

[1]  P. Battista,et al.  Frontiers for the Early Diagnosis of AD by Means of MRI Brain Imaging and Support Vector Machines. , 2016, Current Alzheimer research.

[2]  N. Schuff,et al.  Different regional patterns of cortical thinning in Alzheimer's disease and frontotemporal dementia. , 2006, Brain : a journal of neurology.

[3]  P. Scheltens,et al.  Research criteria for the diagnosis of Alzheimer's disease: revising the NINCDS–ADRDA criteria , 2007, The Lancet Neurology.

[4]  G. Binetti,et al.  A brief neuropsychological assessment for the differential diagnosis between frontotemporal dementia and Alzheimer's disease , 2001, European journal of neurology.

[5]  W. Moon,et al.  Structural MR Imaging in the Diagnosis of Alzheimer's Disease and Other Neurodegenerative Dementia: Current Imaging Approach and Future Perspectives , 2016, Korean journal of radiology.

[6]  Vladimir Vapnik,et al.  An overview of statistical learning theory , 1999, IEEE Trans. Neural Networks.

[7]  Juha Koikkalainen,et al.  Differential diagnosis of neurodegenerative diseases using structural MRI data , 2016, NeuroImage: Clinical.

[8]  Nick C Fox,et al.  Accuracy of dementia diagnosis—a direct comparison between radiologists and a computerized method , 2008, Brain : a journal of neurology.

[9]  W. K. Simmons,et al.  Circular analysis in systems neuroscience: the dangers of double dipping , 2009, Nature Neuroscience.

[10]  M. Bozzali,et al.  Patterns of Cerebellar Gray Matter Atrophy Across Alzheimer’s Disease Progression , 2018, Front. Cell. Neurosci..

[11]  Emma J. Burton,et al.  A comprehensive study of gray matter loss in patients with Alzheimer’s disease using optimized voxel-based morphometry , 2003, NeuroImage.

[12]  X. Wu,et al.  Individual patient diagnosis of AD and FTD via high-dimensional pattern classification of MRI , 2008, NeuroImage.

[13]  N. Malykhin,et al.  Hippocampal neuroplasticity in major depressive disorder , 2015, Neuroscience.

[14]  P. Scheltens,et al.  Atrophy of medial temporal lobes on MRI in “probable” Alzheimer's disease and normal ageing: diagnostic value and neuropsychological correlates , 2012, Journal of Neurology, Neurosurgery & Psychiatry.

[15]  N. Schuff,et al.  MRI signatures of brain macrostructural atrophy and microstructural degradation in frontotemporal lobar degeneration subtypes. , 2012, Journal of Alzheimer's disease : JAD.

[16]  M. Jorge Cardoso,et al.  Patterns of progressive atrophy vary with age in Alzheimer's disease patients , 2018, Neurobiology of Aging.

[17]  Jesse A. Brown,et al.  Machine learning based hierarchical classification of frontotemporal dementia and Alzheimer's disease , 2019, NeuroImage: Clinical.

[18]  Harald Hampel,et al.  Free and Cued Selective Reminding Test – accuracy for the differential diagnosis of Alzheimer's and neurodegenerative diseases: A large-scale biomarker-characterized monocenter cohort study (ClinAD) , 2017, Alzheimer's & Dementia.

[19]  Sirio Cocozza,et al.  Cerebellum and neurodegenerative diseases: Beyond conventional magnetic resonance imaging , 2017, World journal of radiology.

[20]  Rogier A. Feis,et al.  Single Subject Classification of Alzheimer's Disease and Behavioral Variant Frontotemporal Dementia Using Anatomical, Diffusion Tensor, and Resting-State Functional Magnetic Resonance Imaging. , 2018, Journal of Alzheimer's disease : JAD.

[21]  H. Akiyama,et al.  Clinical features of the behavioural variant of frontotemporal dementia that are useful for predicting underlying pathological subtypes of frontotemporal lobar degeneration , 2018, Psychogeriatrics : the official journal of the Japanese Psychogeriatric Society.

[22]  Rebecca M. E. Steketee,et al.  Multiparametric computer-aided differential diagnosis of Alzheimer’s disease and frontotemporal dementia using structural and advanced MRI , 2016, European Radiology.

[23]  Osamu Abe,et al.  Voxel-based morphometric brain comparison between healthy subjects and major depressive disorder patients in Japanese with the s/s genotype of 5-HTTLPR , 2017, Scientific Reports.

[24]  Jeroen van der Grond,et al.  Alzheimer Disease and Behavioral Variant Frontotemporal Dementia: Automatic Classification Based on Cortical Atrophy for Single-Subject Diagnosis. , 2016, Radiology.

[25]  Karl J. Friston,et al.  Voxel-Based Morphometry—The Methods , 2000, NeuroImage.

[26]  John Ashburner,et al.  A fast diffeomorphic image registration algorithm , 2007, NeuroImage.

[27]  Juha Koikkalainen,et al.  Five-class differential diagnostics of neurodegenerative diseases using random undersampling boosting , 2017, NeuroImage: Clinical.

[28]  Nick C Fox,et al.  Automatic classification of MR scans in Alzheimer's disease. , 2008, Brain : a journal of neurology.

[29]  Janet B W Williams,et al.  Diagnostic and Statistical Manual of Mental Disorders , 2013 .

[30]  Nick C Fox,et al.  Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. , 2011, Brain : a journal of neurology.

[31]  Nick C Fox,et al.  Rates of global and regional cerebral atrophy in AD and frontotemporal dementia , 2001, Neurology.

[32]  Jing Yang,et al.  Voxelwise meta-analysis of gray matter anomalies in Alzheimer's disease and mild cognitive impairment using anatomic likelihood estimation , 2012, Journal of the Neurological Sciences.

[33]  J. Hodges,et al.  Cerebellar atrophy and its contribution to cognition in frontotemporal dementias , 2018, Annals of neurology.

[34]  Marie Chupin,et al.  Automatic classi fi cation of patients with Alzheimer ' s disease from structural MRI : A comparison of ten methods using the ADNI database , 2010 .

[35]  Kiralee M. Hayashi,et al.  The topography of grey matter involvement in early and late onset Alzheimer's disease. , 2007, Brain : a journal of neurology.

[36]  Pierrick Coupé,et al.  Accuracy of MRI Classification Algorithms in a Tertiary Memory Center Clinical Routine Cohort , 2020, Journal of Alzheimer's disease : JAD.