Classification of First-Episode Schizophrenia Using Multimodal Brain Features: A Combined Structural and Diffusion Imaging Study

Abstract

Recent neuroanatomical pattern recognition studies have shown some promises for developing an objective neuroimaging-based classification related to schizophrenia. This study explored the feasibility of reliably identifying schizophrenia using single and multimodal multivariate neuroimaging features. Multiple brain measures including regional gray matter (GM) volume, cortical thickness, gyrification, fractional anisotropy (FA), and mean diffusivity (MD) were extracted using fully automated procedures. We used Gradient Boosting Decision Tree to identify the most frequently selected features of each set of neuroanatomical metric and fused multimodal measures. The current classification model was trained and validated based on 98 patients with first-episode schizophrenia (FES) and 106 matched healthy controls (HCs). The classification model was trained and tested in an independent dataset of 54 patients with FES and 48 HCs using imaging data acquired on a different magnetic resonance imaging scanner. Using the most frequently selected features from fused structural and diffusion tensor imaging metrics, a classification accuracy of 75.05% was achieved, which was higher than accuracy derived from a single imaging metric. Most prominent discriminative features included cortical thickness of left transverse temporal gyrus and right parahippocampal gyrus, the FA of left corticospinal tract and right external capsule. In the independent cohort, average accuracy was 76.54%, derived from combined features selected from cortical thickness, gyrification, FA, and MD. These features characterized by GM abnormalities and white matter disruptions have discriminative power with respect to the underlying pathological changes in the brain of individuals having schizophrenia. Our results further highlight the potential advantage of multimodal data fusion for identifying schizophrenia.

A longitudinal neurite and free water imaging study in patients with a schizophrenia spectrum disorder

“…Posterior limb white matter integrity reductions, measured as fractional anisotropy, were previously reported in a small group of patients with early-stage schizophrenia [47]. Further underscoring the role of this region in the illness, a machine learning based pattern recognition study discovered that fractional anisotropy of the corticospinal tract was one of the most prominent discriminant features separating HC from first-episode schizophrenia patients [48]. Interestingly, the posterior limb of the internal capsule was one of the few white matter regions where no fractional anisotropy abnormalities were detected in chronic schizophrenia patients in a large individual data meta-analysis in conducted by the ENIGMA DTI working group [1].…”

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

“…Cutting-edge machine learning methods have been applied in structural and functional neuroimaging studies and have revealed that multivariate patterns of brain change are sensitive enough to classify individual SCZ patients (Liu et al, 2015;Yuan et al, 2018b). Combining cortical thickness, gyrification of gray matter, and fractional anisotropy and mean diffusivity of white matter, Liang et al (2019) used a gradient boosting decision tree to identify SCZ patients, reaching an average accuracy of 76.54%. Using global and nodal network properties derived from a graph theory analysis, Jo et al (2020) revealed that functional network properties had a high discriminatory ability for classifying SCZ patients and HCs.…”

sefOri: selecting the best-engineered sequence features to predict DNA replication origins

“…Classification algorithms were used to separate the positive samples from the negative ones in each of the four datasets {D(Sc), D(Sp), D(Kl), D(Pp)}. This study utilized five classification algorithms to evaluate their capabilities of predicting DNA replication origins, i.e., support vector machine (SVM) (Weston, et al, 2001), random forest (RF) (Jang, et al, 2018;Li, et al, 2018), multinomial naïve Bayes (MNB) classifier (Pan, et al, 2018), gradient boosting decision tree (GBDT) (Liang, et al, 2018;Wang, et al, 2019), and back propagation neural network (BPNN) (Rumelhart, et al, 1986).…”