Mild cognitive impairment classification using combined structural and diffusion imaging biomarkers

Abstract

Alzheimer's disease is a multifactorial neurodegenerative disorder preceded by a prodromal stage called mild cognitive impairment (MCI). Early diagnosis of MCI is crucial for delaying the progression and optimizing the treatment. In this study we propose a random forest (RF) classifier to distinguish between MCI and healthy control subjects (HC), identifying the most relevant features computed from structural T1-weighted and diffusion-weighted magnetic resonance images (sMRI and DWI), combined with neuro-psychological scores. To train the RF we used a set of 60 subjects (HC = 30, MCI = 30) drawn from the Alzheimer's disease neuroimaging initiative database, while testing with unseen data was carried out on a 23-subjects Mexican cohort (HC = 12, MCI = 11). Features from hippocampus, thalamus and amygdala, for left and right hemispheres were fed to the RF, with the most relevant being previously selected by applying extra trees classifier and the mean decrease in impurity index. All the analyzed brain structures presented changes in sMRI and DWI features for MCI, but those computed from sMRI contribute the most to distinguish from HC. However, sMRI+DWI improves classification performance in training area under the receiver operating characteristic curve (AUROC = 93.5 ± 8%, accuracy = 88.8 ± 9%) and testing with unseen data (AUROC = 93.79%, accuracy = 91.3%), having a better performance when neuro-psychological scores were included. Compared to other classifiers the proposed RF provide the best performance for HC/MCI discrimination and the application of a feature selection step improves its performance. These findings imply that multimodal analysis gives better results than unimodal analysis and hence may be a useful tool to assist in early MCI diagnosis.

ECG Marker Evaluation for the Machine-Learning-Based Classification of Acute and Chronic Phases of Trypanosoma cruzi Infection in a Murine Model

“…The MDI indicates the degree of impurity (classification capability) for each variable according to the GINI index, which measures the probability of a particular variable being wrongly classified when it is randomly chosen. This algorithm has been selected because of its ability to assign the importance of biological features [ 11 , 23 , 24 ]. Furthermore, in a previous work in which various feature selection algorithms were compared, the MDI-based approach was one of the best performing [ 21 ].…”

Multimodal magnetic resonance imaging for Alzheimer's disease diagnosis using hybrid features extraction and ensemble support vector machines

“…Many different methods and algorithms were introduced to identify the AD micro macrostructural changes such as machine learning (ML) [8] and deep learning especially convolutional neural network (CNN). They capture low-to-high-level features from an input dataset without human intervention.…”

“…6 Furthermore, multiple studies have found that the putamen, thalamus, and other structures are all affected by AD. 7,8 On the other hand, diffusion tensor imaging (DTI) is a new MRI modality introduced to detect microstructural changes that are undetectable in anatomical scans. 9 Hence, it provides additional information to the sMRI.…”

“…However, it is associated with numerous neurodegenerative disorders 6 . Furthermore, multiple studies have found that the putamen, thalamus, and other structures are all affected by AD 7,8 …”

“…Jorge Perez‐Gonzalez 8 calculated the pertinent characteristics from sMRI and DTI and merged them with neuropsychological scores before inputting them to the RF classifier. Using characteristics from the hippocampus, thalamus, and amygdala in both the right and left hemispheres, they were able to identify between MCI and CN with an accuracy of 88.8%.…”