Tags

Type your tag names separated by a space and hit enter

Random Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness.
Neuroimage Clin 2014; 6:115-25NC

Abstract

Computer-aided diagnosis of Alzheimer's disease (AD) is a rapidly developing field of neuroimaging with strong potential to be used in practice. In this context, assessment of models' robustness to noise and imaging protocol differences together with post-processing and tuning strategies are key tasks to be addressed in order to move towards successful clinical applications. In this study, we investigated the efficacy of Random Forest classifiers trained using different structural MRI measures, with and without neuroanatomical constraints in the detection and prediction of AD in terms of accuracy and between-cohort robustness. From The ADNI database, 185 AD, and 225 healthy controls (HC) were randomly split into training and testing datasets. 165 subjects with mild cognitive impairment (MCI) were distributed according to the month of conversion to dementia (4-year follow-up). Structural 1.5-T MRI-scans were processed using Freesurfer segmentation and cortical reconstruction. Using the resulting output, AD/HC classifiers were trained. Training included model tuning and performance assessment using out-of-bag estimation. Subsequently the classifiers were validated on the AD/HC test set and for the ability to predict MCI-to-AD conversion. Models' between-cohort robustness was additionally assessed using the AddNeuroMed dataset acquired with harmonized clinical and imaging protocols. In the ADNI set, the best AD/HC sensitivity/specificity (88.6%/92.0% - test set) was achieved by combining cortical thickness and volumetric measures. The Random Forest model resulted in significantly higher accuracy compared to the reference classifier (linear Support Vector Machine). The models trained using parcelled and high-dimensional (HD) input demonstrated equivalent performance, but the former was more effective in terms of computation/memory and time costs. The sensitivity/specificity for detecting MCI-to-AD conversion (but not AD/HC classification performance) was further improved from 79.5%/75%-83.3%/81.3% by a combination of morphometric measurements with ApoE-genotype and demographics (age, sex, education). When applied to the independent AddNeuroMed cohort, the best ADNI models produced equivalent performance without substantial accuracy drop, suggesting good robustness sufficient for future clinical implementation.

Authors+Show Affiliations

Centre for Age-Related Medicine, Stavanger University Hospital, Stavanger, Norway.Department of Neurobiology, Care Sciences and Society, Division of Neurogeriatrics, Alzheimer's Disease Research Centre, Karolinska Institute, Stockholm, Sweden.European Molecular Biology Laboratory, European Bioinformatics Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, United Kingdom.Department of Neurology, University Medical Center Ljubljana, Slovenia.Neuroinformatics and Image Analysis Laboratory, Department of Biomedicine, University of Bergen, Bergen, Norway ; Department of Radiology, Haukeland University Hospital, Bergen, Norway.Centre for Age-Related Medicine, Stavanger University Hospital, Stavanger, Norway ; Department of Neurobiology, Care Sciences and Society, Division of Neurogeriatrics, Alzheimer's Disease Research Centre, Karolinska Institute, Stockholm, Sweden.Department of Neurology, University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland.Department of Old Age Psychiatry and Psychotic Disorders, Medical University of Lodz, Lódz, Poland.Institute of Gerontology and Geriatrics, University of Perugia, Perugia, Italy.Aristotle University of Thessaloniki, Thessaloniki, Greece.GERONTOPOLE, UMR INSERM 1027, CHU, University of Toulouse, France.King's College London, Institute of Psychiatry, London, UK ; NIHR Biomedical Research Centre for Mental Health and Biomedical Research Unit for Dementia, London, UK.King's College London, Institute of Psychiatry, London, UK ; NIHR Biomedical Research Centre for Mental Health and Biomedical Research Unit for Dementia, London, UK.No affiliation info available

Pub Type(s)

Comparative Study
Journal Article
Randomized Controlled Trial
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

Language

eng

PubMed ID

25379423

Citation

Lebedev, A V., et al. "Random Forest Ensembles for Detection and Prediction of Alzheimer's Disease With a Good Between-cohort Robustness." NeuroImage. Clinical, vol. 6, 2014, pp. 115-25.
Lebedev AV, Westman E, Van Westen GJ, et al. Random Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness. Neuroimage Clin. 2014;6:115-25.
Lebedev, A. V., Westman, E., Van Westen, G. J., Kramberger, M. G., Lundervold, A., Aarsland, D., ... Simmons, A. (2014). Random Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness. NeuroImage. Clinical, 6, pp. 115-25. doi:10.1016/j.nicl.2014.08.023.
Lebedev AV, et al. Random Forest Ensembles for Detection and Prediction of Alzheimer's Disease With a Good Between-cohort Robustness. Neuroimage Clin. 2014;6:115-25. PubMed PMID: 25379423.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Random Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness. AU - Lebedev,A V, AU - Westman,E, AU - Van Westen,G J P, AU - Kramberger,M G, AU - Lundervold,A, AU - Aarsland,D, AU - Soininen,H, AU - Kłoszewska,I, AU - Mecocci,P, AU - Tsolaki,M, AU - Vellas,B, AU - Lovestone,S, AU - Simmons,A, AU - ,, Y1 - 2014/08/28/ PY - 2014/02/02/received PY - 2014/06/06/revised PY - 2014/08/26/accepted PY - 2014/11/8/entrez PY - 2014/11/8/pubmed PY - 2015/7/21/medline KW - ADNI KW - AddNeuroMed KW - Alzheimer's disease KW - Computer-aided diagnosis KW - Mild cognitive impairment KW - Multi-center study KW - Random Forest KW - Structural MRI SP - 115 EP - 25 JF - NeuroImage. Clinical JO - Neuroimage Clin VL - 6 N2 - Computer-aided diagnosis of Alzheimer's disease (AD) is a rapidly developing field of neuroimaging with strong potential to be used in practice. In this context, assessment of models' robustness to noise and imaging protocol differences together with post-processing and tuning strategies are key tasks to be addressed in order to move towards successful clinical applications. In this study, we investigated the efficacy of Random Forest classifiers trained using different structural MRI measures, with and without neuroanatomical constraints in the detection and prediction of AD in terms of accuracy and between-cohort robustness. From The ADNI database, 185 AD, and 225 healthy controls (HC) were randomly split into training and testing datasets. 165 subjects with mild cognitive impairment (MCI) were distributed according to the month of conversion to dementia (4-year follow-up). Structural 1.5-T MRI-scans were processed using Freesurfer segmentation and cortical reconstruction. Using the resulting output, AD/HC classifiers were trained. Training included model tuning and performance assessment using out-of-bag estimation. Subsequently the classifiers were validated on the AD/HC test set and for the ability to predict MCI-to-AD conversion. Models' between-cohort robustness was additionally assessed using the AddNeuroMed dataset acquired with harmonized clinical and imaging protocols. In the ADNI set, the best AD/HC sensitivity/specificity (88.6%/92.0% - test set) was achieved by combining cortical thickness and volumetric measures. The Random Forest model resulted in significantly higher accuracy compared to the reference classifier (linear Support Vector Machine). The models trained using parcelled and high-dimensional (HD) input demonstrated equivalent performance, but the former was more effective in terms of computation/memory and time costs. The sensitivity/specificity for detecting MCI-to-AD conversion (but not AD/HC classification performance) was further improved from 79.5%/75%-83.3%/81.3% by a combination of morphometric measurements with ApoE-genotype and demographics (age, sex, education). When applied to the independent AddNeuroMed cohort, the best ADNI models produced equivalent performance without substantial accuracy drop, suggesting good robustness sufficient for future clinical implementation. SN - 2213-1582 UR - https://www.unboundmedicine.com/medline/citation/25379423/Random_Forest_ensembles_for_detection_and_prediction_of_Alzheimer's_disease_with_a_good_between_cohort_robustness_ L2 - https://linkinghub.elsevier.com/retrieve/pii/S2213-1582(14)00132-6 DB - PRIME DP - Unbound Medicine ER -