Dual-Model Radiomic Biomarkers Predict Development of Mild Cognitive Impairment Progression to Alzheimer’s Disease

Abstract

Predicting progression of mild cognitive impairment (MCI) to Alzheimer’s disease (AD) is clinically important. In this study, we propose a dual-model radiomic analysis with multivariate Cox proportional hazards regression models to investigate promising risk factors associated with MCI conversion to AD. T1 structural magnetic resonance imaging (MRI) and 18F-Fluorodeoxyglucose (FDG) positron emission tomography (PET) data, from the AD Neuroimaging Initiative database, were collected from 131 patients with MCI who converted to AD within 3 years and 132 patients with MCI without conversion within 3 years. These subjects were randomly partition into 70% training dataset and 30% test dataset with multiple times. We fused MRI and PET images by wavelet method. In a subset of subjects, a group comparison was performed using a two-sample t-test to determine regions of interest (ROIs) associated with MCI conversion. 172 radiomic features from ROIs for each individual were established using a published radiomics tool. Finally, L1-penalized Cox model was constructed and Harrell’s C index (C-index) was used to evaluate prediction accuracy of the model. To evaluate the efficacy of our proposed method, we used a same analysis framework to evaluate MRI and PET data separately. We constructed prognostic Cox models with: clinical data, MRI images, PET images, fused MRI/PET images, and clinical variables and fused MRI/PET images in combination. The experimental results showed that captured ROIs significantly associated with conversion to AD, such as gray matter atrophy in the bilateral hippocampus and hypometabolism in the temporoparietal cortex. Imaging model (MRI/PET/fused) provided significant enhancement in prediction of conversion compared to clinical models, especially the fused-modality Cox model. Moreover, the combination of fused-modality imaging and clinical variables resulted in the greatest accuracy of prediction. The average C-index for the clinical/MRI/PET/fused/combined model in the test dataset was 0.69, 0.73, 0.73 and 0.75, and 0.78, respectively. These results suggested that a combination of radiomic analysis and Cox model analyses could be used successfully in survival analysis and may be powerful tools for personalized precision medicine patients with potential to undergo conversion from MCI to AD.

The Power of EEG to Predict Conversion from Mild Cognitive Impairment and Subjective Cognitive Decline to Dementia

“…Recently, one study using a machine learning program to evaluate structural brain changes using MRI as predictors of conversion reported very high AUC, and another study using voxelbased FDGPET also reported very good re sults [19,21]. Best results are though reported from stud ies using various biomarkers in combination [24,25]. However, to design a study to examine the discriminant power of a combination of biomarkers and a cognitive test is difficult due to a circularity problem.…”

“…The use of various MRI and PET methods and cerebrospinal fluid (CSF) examination, measuring concentrations of amyloidβ, phosphorylated and total tau proteins, as well as other proteins, including inflam matory markers, have shown to be helpful in predicting patients with MCI and SCD who will convert to demen tia, especially to AD [16][17][18][19][20][21][22]. A recent Cochrane review, however, does not support the use of βamyloid PET as biomarker for AD, but studies examining the combina tion of cognitive tests with CSF proteins, MRI, and FDGPET have shown excellent results in identifying people with MCI that would convert to dementia [23][24][25]. However, the use of biomarkers are time and per sonnel consuming, costly, and some of them are inva sive, and, consequently, not ideal for use in daily clinical practice, except in subspecialized settings.…”

Parkinson's Disease Diagnosis Using Neostriatum Radiomic Features Based on T2-Weighted Magnetic Resonance Imaging

“…However, given its power in capturing the microstructural changes in tissues and its correlation with clinical endpoints (17) and age progression (18,19), the use of radiomics is expected to increase in neurodegenerative disorders (20). Presently, radiomics has been applied to the diagnosis of neurodegenerative diseases, including Alzheimer's disease (AD), amyotrophic lateral sclerosis, and Machado-Joseph disease with conventional MRI (21)(22)(23)(24)(25), which have similar pathological changes with PD, such as atrophy, abnormal proteins, or iron deposition in many brain regions. In a longitudinal study, radiomics successfully detected microstructural changes in invisible normal-appearing white matter on conventional T2 fluid-attenuated inversionrecovery (FLAIR) images (18).…”

Big Imaging Data: Klinische Bildanalyse mit Radiomics und Deep Learning

“…Sowohl von Zhou et al [75] als auch von Spasov et al [76] wurde untersucht, ob mithilfe von Radiomics vorhergesagt werden kann, ob leichte kognitive Störungen sich zu Alzheimer entwickeln. Obwohl beide mit demselben Patientenkollektiv arbeiten, der Datenbank der Alzheimer's Disease Neuroimaging Initiative (http://adni.loni.usc.edu/), unterscheiden sie sich in der Auswahl des Kollektivs und den eingesetzten Methoden.…”

“…Obwohl beide mit demselben Patientenkollektiv arbeiten, der Datenbank der Alzheimer's Disease Neuroimaging Initiative (http://adni.loni.usc.edu/), unterscheiden sie sich in der Auswahl des Kollektivs und den eingesetzten Methoden. In Zhou et al [75] wurde ein klassischer Ansatz verfolgt, bei dem aus MRT und 18 F-FDG-PET-Daten eine Radiomics Signatur erstellt wurde. MRT und PET-Scans wurden dabei nicht separat betrachtet, sondern mittels Wavelet-Transformation fusioniert.…”