Predicting future clinical changes of MCI patients using longitudinal and multimodal biomarkers.

Daoqiang Zhang,Dinggang Shen,Kewei Chen

doi:10.1371/journal.pone.0033182

Abstract

Accurate prediction of clinical changes of mild cognitive impairment (MCI) patients, including both qualitative change (i.e., conversion to Alzheimer's disease (AD)) and quantitative change (i.e., cognitive scores) at future time points, is important for early diagnosis of AD and for monitoring the disease progression. In this paper, we propose to predict future clinical changes of MCI patients by using both baseline and longitudinal multimodality data. To do this, we first develop a longitudinal feature selection method to jointly select brain regions across multiple time points for each modality. Specifically, for each time point, we train a sparse linear regression model by using the imaging data and the corresponding clinical scores, with an extra ‘group regularization’ to group the weights corresponding to the same brain region across multiple time points together and to allow for selection of brain regions based on the strength of multiple time points jointly. Then, to further reflect the longitudinal changes on the selected brain regions, we extract a set of longitudinal features from the original baseline and longitudinal data. Finally, we combine all features on the selected brain regions, from different modalities, for prediction by using our previously proposed multi-kernel SVM. We validate our method on 88 ADNI MCI subjects, with both MRI and FDG-PET data and the corresponding clinical scores (i.e., MMSE and ADAS-Cog) at 5 different time points. We first predict the clinical scores (MMSE and ADAS-Cog) at 24-month by using the multimodality data at previous time points, and then predict the conversion of MCI to AD by using the multimodality data at time points which are at least 6-month ahead of the conversion. The results on both sets of experiments show that our proposed method can achieve better performance in predicting future clinical changes of MCI patients than the conventional methods.

Highlights

Alzheimer’s disease (AD), the most common form of dementia, is a progressive age-related neurodegenerative disease usually diagnosed in people over 65 years of age
Over the past several years, many high-dimensional pattern classification methods have been developed for classification of AD and Mild cognitive impairment (MCI) based on different modalities of biomarkers, e.g., the structural brain atrophy measured by magnetic resonance imaging (MRI) [2,3,4], the metabolic brain alterations measured by fluorodeoxyglucose positron emission tomography (FDG-PET) [5,6], and the pathological amyloid depositions measured through cerebrospinal fluid (CSF) [3,7,8,9], etc
In the CONCAT method, for each subject, all data from different modalities and different time points are first concatenated into a long vector, and a standard feature selection method based on Lasso [37] is performed, followed by using the standard support vector machines (SVM) for final regression

Summary

Introduction

Alzheimer’s disease (AD), the most common form of dementia, is a progressive age-related neurodegenerative disease usually diagnosed in people over 65 years of age. It’s important to predict whether a certain MCI subject will convert into AD at future time points or not. This is a qualitative prediction, which can be solved through classification between MCI-C and MCI-NC. Because AD is a progressive neurodegenerative disease, there exist continuous changes between the measured clinical scores, e.g., Mini Mental State Examination (MMSE) and Alzheimer’s Disease Assessment Scale - Cognitive Subscale (ADAS-Cog), at follow-up time points. It’s important to predict the future clinical scores based on the data at previous time points, which is especially helpful for monitoring disease progression. Different from predicting MCI conversion, predicting future clinical scores requires a quantitative prediction, which needs to be solved by learning a regression model, instead of a classification model

Methods

Results

Discussion

Conclusion