Individual Prediction of Behavioral Variant Frontotemporal Dementia Development Using Multivariate Pattern Analysis of Magnetic Resonance Imaging Data

Paul Zhutovsky, Everard G. B. Vijverberg, Willem B. Bruin, Rajat M. Thomas, Mike P. Wattjes, Yolande A. L. Pijnenburg, Guido A. van Wingen, Annemiek Dols

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Patients with behavioral variant of frontotemporal dementia (bvFTD) initially may only show behavioral and/or cognitive symptoms that overlap with other neurological and psychiatric disorders. The diagnostic accuracy is dependent on progressive symptoms worsening and frontotemporal abnormalities on neuroimaging findings. Predictive biomarkers could facilitate the early detection of bvFTD. Objective: To determine the prognostic accuracy of clinical and structural MRI data using a support vector machine (SVM) classification to predict the 2-year clinical follow-up diagnosis in a group of patients presenting late-onset behavioral changes. Methods: Data from 73 patients were included and divided into probable/definite bvFTD (n=18), neurological (n=28), and psychiatric (n=27) groups based on 2-year follow-up diagnosis. Grey-matter volumes were extracted from baseline structural MRI scans. SVM classifiers were used to perform three binary classifications: bvFTD versus neurological and psychiatric, bvFTD versus neurological, and bvFTD versus psychiatric group(s), and one multi-class classification. Classification performance was determined for clinical and neuroimaging data separately and their combination using 5-fold cross-validation. Results: Accuracy of the binary classification tasks ranged from 72-82% (p<0.001) with adequate sensitivity (67-79%), specificity (77-88%), and area-under-the-receiver-operator-curve (0.80-0.9). Multi-class accuracy ranged between 55-59% (p<0.001). The combination of clinical and voxel-wise whole brain data showed the best performance overall. Conclusion: These results show the potential for automated early confirmation of diagnosis for bvFTD using machine learning analysis of clinical and neuroimaging data in a diverse and clinically relevant sample of patients.
Original languageEnglish
Pages (from-to)1229-1241
JournalJournal of Alzheimer's Disease
Volume68
Issue number3
DOIs
Publication statusPublished - 2019

Cite this

@article{b56724c097514cae8dc48011965dd861,
title = "Individual Prediction of Behavioral Variant Frontotemporal Dementia Development Using Multivariate Pattern Analysis of Magnetic Resonance Imaging Data",
abstract = "Patients with behavioral variant of frontotemporal dementia (bvFTD) initially may only show behavioral and/or cognitive symptoms that overlap with other neurological and psychiatric disorders. The diagnostic accuracy is dependent on progressive symptoms worsening and frontotemporal abnormalities on neuroimaging findings. Predictive biomarkers could facilitate the early detection of bvFTD. Objective: To determine the prognostic accuracy of clinical and structural MRI data using a support vector machine (SVM) classification to predict the 2-year clinical follow-up diagnosis in a group of patients presenting late-onset behavioral changes. Methods: Data from 73 patients were included and divided into probable/definite bvFTD (n=18), neurological (n=28), and psychiatric (n=27) groups based on 2-year follow-up diagnosis. Grey-matter volumes were extracted from baseline structural MRI scans. SVM classifiers were used to perform three binary classifications: bvFTD versus neurological and psychiatric, bvFTD versus neurological, and bvFTD versus psychiatric group(s), and one multi-class classification. Classification performance was determined for clinical and neuroimaging data separately and their combination using 5-fold cross-validation. Results: Accuracy of the binary classification tasks ranged from 72-82{\%} (p<0.001) with adequate sensitivity (67-79{\%}), specificity (77-88{\%}), and area-under-the-receiver-operator-curve (0.80-0.9). Multi-class accuracy ranged between 55-59{\%} (p<0.001). The combination of clinical and voxel-wise whole brain data showed the best performance overall. Conclusion: These results show the potential for automated early confirmation of diagnosis for bvFTD using machine learning analysis of clinical and neuroimaging data in a diverse and clinically relevant sample of patients.",
author = "Paul Zhutovsky and Vijverberg, {Everard G. B.} and Bruin, {Willem B.} and Thomas, {Rajat M.} and Wattjes, {Mike P.} and Pijnenburg, {Yolande A. L.} and {van Wingen}, {Guido A.} and Annemiek Dols",
year = "2019",
doi = "10.3233/JAD-181004",
language = "English",
volume = "68",
pages = "1229--1241",
journal = "Journal of Alzheimer's Disease",
issn = "1387-2877",
publisher = "IOS Press",
number = "3",

}

Individual Prediction of Behavioral Variant Frontotemporal Dementia Development Using Multivariate Pattern Analysis of Magnetic Resonance Imaging Data. / Zhutovsky, Paul; Vijverberg, Everard G. B.; Bruin, Willem B.; Thomas, Rajat M.; Wattjes, Mike P.; Pijnenburg, Yolande A. L.; van Wingen, Guido A.; Dols, Annemiek.

In: Journal of Alzheimer's Disease, Vol. 68, No. 3, 2019, p. 1229-1241.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Individual Prediction of Behavioral Variant Frontotemporal Dementia Development Using Multivariate Pattern Analysis of Magnetic Resonance Imaging Data

AU - Zhutovsky, Paul

AU - Vijverberg, Everard G. B.

AU - Bruin, Willem B.

AU - Thomas, Rajat M.

AU - Wattjes, Mike P.

AU - Pijnenburg, Yolande A. L.

AU - van Wingen, Guido A.

AU - Dols, Annemiek

PY - 2019

Y1 - 2019

N2 - Patients with behavioral variant of frontotemporal dementia (bvFTD) initially may only show behavioral and/or cognitive symptoms that overlap with other neurological and psychiatric disorders. The diagnostic accuracy is dependent on progressive symptoms worsening and frontotemporal abnormalities on neuroimaging findings. Predictive biomarkers could facilitate the early detection of bvFTD. Objective: To determine the prognostic accuracy of clinical and structural MRI data using a support vector machine (SVM) classification to predict the 2-year clinical follow-up diagnosis in a group of patients presenting late-onset behavioral changes. Methods: Data from 73 patients were included and divided into probable/definite bvFTD (n=18), neurological (n=28), and psychiatric (n=27) groups based on 2-year follow-up diagnosis. Grey-matter volumes were extracted from baseline structural MRI scans. SVM classifiers were used to perform three binary classifications: bvFTD versus neurological and psychiatric, bvFTD versus neurological, and bvFTD versus psychiatric group(s), and one multi-class classification. Classification performance was determined for clinical and neuroimaging data separately and their combination using 5-fold cross-validation. Results: Accuracy of the binary classification tasks ranged from 72-82% (p<0.001) with adequate sensitivity (67-79%), specificity (77-88%), and area-under-the-receiver-operator-curve (0.80-0.9). Multi-class accuracy ranged between 55-59% (p<0.001). The combination of clinical and voxel-wise whole brain data showed the best performance overall. Conclusion: These results show the potential for automated early confirmation of diagnosis for bvFTD using machine learning analysis of clinical and neuroimaging data in a diverse and clinically relevant sample of patients.

AB - Patients with behavioral variant of frontotemporal dementia (bvFTD) initially may only show behavioral and/or cognitive symptoms that overlap with other neurological and psychiatric disorders. The diagnostic accuracy is dependent on progressive symptoms worsening and frontotemporal abnormalities on neuroimaging findings. Predictive biomarkers could facilitate the early detection of bvFTD. Objective: To determine the prognostic accuracy of clinical and structural MRI data using a support vector machine (SVM) classification to predict the 2-year clinical follow-up diagnosis in a group of patients presenting late-onset behavioral changes. Methods: Data from 73 patients were included and divided into probable/definite bvFTD (n=18), neurological (n=28), and psychiatric (n=27) groups based on 2-year follow-up diagnosis. Grey-matter volumes were extracted from baseline structural MRI scans. SVM classifiers were used to perform three binary classifications: bvFTD versus neurological and psychiatric, bvFTD versus neurological, and bvFTD versus psychiatric group(s), and one multi-class classification. Classification performance was determined for clinical and neuroimaging data separately and their combination using 5-fold cross-validation. Results: Accuracy of the binary classification tasks ranged from 72-82% (p<0.001) with adequate sensitivity (67-79%), specificity (77-88%), and area-under-the-receiver-operator-curve (0.80-0.9). Multi-class accuracy ranged between 55-59% (p<0.001). The combination of clinical and voxel-wise whole brain data showed the best performance overall. Conclusion: These results show the potential for automated early confirmation of diagnosis for bvFTD using machine learning analysis of clinical and neuroimaging data in a diverse and clinically relevant sample of patients.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85064399861&origin=inward

UR - https://www.ncbi.nlm.nih.gov/pubmed/30909224

U2 - 10.3233/JAD-181004

DO - 10.3233/JAD-181004

M3 - Article

VL - 68

SP - 1229

EP - 1241

JO - Journal of Alzheimer's Disease

JF - Journal of Alzheimer's Disease

SN - 1387-2877

IS - 3

ER -