Predicting individualized clinical measures by a generalized prediction framework and multimodal fusion of MRI data

Abstract

Neuroimaging techniques have greatly enhanced the understanding of neurodiversity (human brain variation across individuals) in both health and disease. The ultimate goal of using brain imaging biomarkers is to perform individualized predictions. Here we proposed a generalized framework that can predict explicit values of the targeted measures by taking advantage of joint information from multiple modalities. This framework also enables whole brain voxel-wise searching by combining multivariate techniques such as ReliefF, clustering, correlation-based feature selection and multiple regression models, which is more flexible and can achieve better prediction performance than alternative atlas-based methods. For 50 healthy controls and 47 schizophrenia patients, three kinds of features derived from resting-state fMRI (fALFF), sMRI (gray matter) and DTI (fractional anisotropy) were extracted and fed into a regression model, achieving high prediction for both cognitive scores (MCCB composite r=0.7033, MCCB social cognition r=0.7084) and symptomatic scores (positive and negative syndrome scale [PANSS] positive r=0.7785, PANSS negative r=0.7804). Moreover, the brain areas likely responsible for cognitive deficits of schizophrenia, including middle temporal gyrus, dorsolateral prefrontal cortex, striatum, cuneus and cerebellum, were located with different weights, as well as regions predicting PANSS symptoms, including thalamus, striatum and inferior parietal lobule, pinpointing the potential neuromarkers. Finally, compared to a single modality, multimodal combination achieves higher prediction accuracy and enables individualized prediction on multiple clinical measures. There is more work to be done, but the current results highlight the potential utility of multimodal brain imaging biomarkers to eventually inform clinical decision-making.

Keywords: Individualized prediction; MATRICS Consensus Cognitive Battery (MCCB); MRI; Multimodal; Neuromarker; Schizophrenia.

Fig. 1

The working flowchart of the proposed prediction framework. Fig. 1 shows working flowchart of the proposed prediction model, in which preprocessing, feature selection (ReliefF), spatial clustering, feature subset selection and regression models are employed. Three MRI measures (fALFF, GM, FA) are combined together to realize the individualized prediction via linear or nonlinear regression analysis.

Fig. 2

Prediction results based on three multimodal features for MCCB MATRICS. Fig. 2: (A) prediction plot based on the proposed framework, cognitive scores (the MCCB composite) for 47 SZs and 50 HCs were plotted in the x-axis (HC: red dots, SZ: blue dots); the Y-axis represents the predicted values.(B) In the reliability test, 1000 bootstrap resamples were performed to estimate the distribution and 95% confidence intervals of correlation coefficients between cognitive score and neuroimaging.

Fig. 3

Identified brain regions that may serve as neuromarkers. Fig. 3 indicates the identified brain clusters of three MRI measures, with the cluster color reflecting its weight and sign, as well as the corresponding regression equation, in which the ultimate correlation with ground truth is 0.7033. In total, 15 clusters were identified, each of which was assigned a weigh that implied how much they contribute to the predicted measure (MCCB composite score). The brain views are left, top and front for each column, respectively.