Functional Connectivity Combined With a Machine Learning Algorithm Can Classify High-Risk First-Degree Relatives of Patients With Schizophrenia and Identify Correlates of Cognitive Impairments

Abstract

Schizophrenia (SCZ) is an inherited disease, with the familial risk being among the most important factors when evaluating an individual’s risk for SCZ. However, robust imaging biomarkers for the disease that can be used for diagnosis and determination of the prognosis are lacking. Here, we explore the potential of functional connectivity (FC) for use as a biomarker for the early detection of high-risk first-degree relatives (FDRs). Thirty-eight first-episode SCZ patients, 38 healthy controls (HCs), and 33 FDRs were scanned using resting-state functional magnetic resonance imaging. The subjects’ brains were parcellated into 200 regions using the Craddock atlas, and the FC between each pair of regions was used as a classification feature. Multivariate pattern analysis using leave-one-out cross-validation achieved a correct classification rate of 88.15% [sensitivity 84.06%, specificity 92.18%, and area under the receiver operating characteristic curve (AUC) 0.93] for differentiating SCZ patients from HCs. FC located within the default mode, frontal-parietal, auditory, and sensorimotor networks contributed mostly to the accurate classification. The FC patterns of each FDR were input into each classification model as test data to obtain a corresponding prediction label (a total of 76 individual classification scores), and the averaged individual classification score was then used as a robust measure to characterize whether each FDR showed an SCZ-type or HC-type FC pattern. A significant negative correlation was found between the average classification scores of the FDRs and their semantic fluency scores. These findings suggest that FC combined with a machine learning algorithm could help to predict whether FDRs are likely to show an SCZ-specific or HC-specific FC pattern.

Classifying Vulnerability to Sleep Deprivation Using Resting-State Functional MRI Graph Theory Metrics

“…These procedures were processed in the “Prepare feature set” program. In the second step, Leave-one-out cross-validation (LOOCV) was used to evaluate the performance of the classifier ( Liu et al, 2020 ). In LOOCV, data from one subject was used as test data and the classifier is trained on the remaining dataset.…”

“…The translational applicability of such data to clinical practice should be based on inferences at the individual rather than group level. With recent advancements in the field of machine learning, such as the support vector machine (SVM) model, a multivariate pattern recognition machine learning (ML) technique especially well-suited for discriminating high-dimensional rsFC fMRI data, measurements derived from fMRI combined with artificial intelligence algorithms have led to improvements in diagnoses, classification, and treatment outcome prediction for a range of situations ( Zhao et al, 2018 ; Liu et al, 2020 ). Furthermore, multivariate machine learning techniques are more sensitive to differences that are subtle and spatially distributed because they consider inter-regional correlations, which might be undetectable using group comparisons ( Liu et al, 2020 ).…”

“…With recent advancements in the field of machine learning, such as the support vector machine (SVM) model, a multivariate pattern recognition machine learning (ML) technique especially well-suited for discriminating high-dimensional rsFC fMRI data, measurements derived from fMRI combined with artificial intelligence algorithms have led to improvements in diagnoses, classification, and treatment outcome prediction for a range of situations ( Zhao et al, 2018 ; Liu et al, 2020 ). Furthermore, multivariate machine learning techniques are more sensitive to differences that are subtle and spatially distributed because they consider inter-regional correlations, which might be undetectable using group comparisons ( Liu et al, 2020 ). Because SD is associated with widespread changes in functional networks, graph-based measurements of network organization, such as degree centrality (DC) ( Wang et al, 2011 ), might have potential in predicting vulnerability to SD-induced deficits in function.…”

Predicting Attentional Vulnerability to Sleep Deprivation: A Multivariate Pattern Analysis of DTI Data

“…Unlike traditional group-level univariate analysis, cutting-edge machine learning techniques can detect differences that are subtle and spatially distributed and have been indicated to improve diagnoses, classification, and treatment-outcome prediction in a range of situations. 13 , 14 However, to our best knowledge, studies using DTI and machine learning strategies for SD vulnerability prediction are rare. In the current study, we applied a linear support vector machine learning approach to investigate whether the WM diffusion metrics can predict vulnerability to SD.…”

Graph-Theory-Based Degree Centrality Combined with Machine Learning Algorithms Can Predict Response to Treatment with Antipsychotic Medications in Patients with First-Episode Schizophrenia

“…With the recent advancements in the field of machine learning, measurements derived from rs-fMRI combined with artificial intelligence algorithms have led to the improvements in the diagnosis, classification, and treatment outcome prediction for a range of diseases, in particular, for schizophrenia [ 10 ]. Our pervious study has indicated that functional connectivity can be a sensitive marker to differentiate SCZ from healthy controls (HCs) [ 11 ]; baseline spontaneous regional activities were also found to be predictive of early response to treatment for SCZ [ 12 ]. Because SCZ is associated with widespread changes in functional networks, graph-based measurements of network organization, such as degree centrality, might have potential in predicting treatment effects.…”