Random Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness

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.

Application of machine learning to construction injury prediction

“…To cite only a few examples, the RF algorithm has been used with success to predict patient risk for various diseases [43,46], identify central genes [23], develop automated stock trading strategies…”

Developing detailed age-specific thematic maps for coffee (Coffea arabica L.) in heterogeneous agricultural landscapes using random forests applied on Landsat 8 multispectral sensor

“…This way, the RF algorithm assesses the importance of each input variable to the outcome by comparing how much the OOB error increases when a variable is removed, while all others are left unchanged (Gislason et al 2004;Breiman & Cutler 2007;Adelabu et al 2013). In many applications that range from medical research, genetics, ecology and remote sensing, the RF algorithm has received attention because it has the ability to handle highly non-linear data, robustness to noise, tuning simplicity and is able to measure the importance of each feature in to model outcome (Caruana & Niculescu-Mizil 2006;Lu & Weng 2007;Rodriguez-Galiano et al 2012;Lebedev et al 2014). In other studies, the RF classifier was reported to be able to discriminate between degraded and healthy grassland, between young and mature sugarcane and between bare and coastal sand .…”

The Worldwide Alzheimer's Disease Neuroimaging Initiative: An update

“…A random forest model that combined cortical thickness and volumetric measures reached an AD/HC sensitivity/specificity of 89%/92%. When APOE-genotype and demographics information (age, sex, and education) were added, the model attained an MCI-to-AD-conversion sensitivity/specificity of 83%/81% [58]. In another study, multivariate ordinal regression applied to baseline structural MRI scans showed an accuracy of 82% in distinguishing HC-like (people with HC and stable MCI) from AD-like (MCI converters and those with AD), and an accuracy of 70% in predicting conversion at 12 months [59].…”