… In general, most GCN studies focused on extracting useful features from brain connectivity data to do classification or to conduct associative analysis. Inferring the relationship between …
… predictivemodeling (38) using leave-one-subject-out cross-validation. In this pipeline, feature-selection and model-… -based predictivemodeling as our prediction approach because a …
… of the FC matrix as features for prediction, namely, each subject has a feature vector in the dimension of (… To confirm that our predictivemodels captured FC variations specific to reading …
X Ma, L Gao, K Tan - Bioinformatics, 2014 - academic.oup.com
… More importantly, we found that M-module–based features achieve significantly higher accuracy in predicting cancer stages (Supplementary Data). This result emphasizes the …
X Yang, Z Xu, Y Xi, J Sun, P Liu, P Li, J Jia, H Yin… - Psychiatry Research …, 2020 - Elsevier
… We then constructed a linear regression predictivemodel with the 18 LASSO features and the PANSS_total scores as the independent and dependent variables, respectively. To …
… lesion-deficit predictionmodels for each measure. For some measures, the FC prediction was far … To further investigate shared features between models, we used multitask learning. All …
C Liu, J Fan, B Bailey, RA Müller… - International Journal of …, 2023 - Wiley Online Library
… Results from our comparative study suggest that functional connectivity (FC) measured with DTW may … of machine learning models and feature engineering methods. As we investigated …
… By only doing matrix completion on selected features instead of all features, the computational cost is greatly reduced and the focus benefits predictivemodeling. RMC is applied on the …
OP Jones, NL Voets, JE Adcock, R Stacey, S Jbabdi - NeuroImage: Clinical, 2017 - Elsevier
… We trained a predictivemodel on pairs of resting-state and … models successfully learned variations in both patient and control responses from the individual resting-connectivityfeatures. …