BrainNetClass (current version: 1.1)

Toolbox for Brain Network Construction and Network-based Classification

BrainNetClass was developed by the IDEA (Image Display, Enhancement and Analysis) Lab at the University of North Carolina at Chapel Hill. It is a Matlab-based, open-coded, fully automated brain functional connectivity network-based disease classification toolbox with user-friendly GUI that automatically conducts functional network construction, network feature extraction and selection, parameter optimization, classification, model validation, and result demonstration. It was designed to help neuroscientists and doctors to easily and correctly work on building brain functional connectomics with state-of-the-art algorithms and conduct rigorous individualized disease diagnosis or other classification tasks even though they do not have abundant knowledge about machine learning.

BrainNetClass provides abundant algorithms for brain functional network construction, including those recently developed for high-order functional networks [1,4-6,8] and those utilizing sparse representation with advanced, biologically meaningful constraints for robust and consistent network construction [2,3,7]. It conducts standard yet rigorous network-based classification with choices for feature extraction, feature reduction, cross-validation, and performance evaluation. In addition to simple metrics (e.g., accuracy), BrainNetClass provides a battery of more comprehensive results (e.g., ROC curve, parameter sensitivity test, model robustness test, contributing discriminative features, and a full log of results report) for researchers to evaluate and interpret their models [9]. The classification model is saved for future use on newly acquired data to perform a quick diagnosis.

Download: (manual, exemplary data included)

More details? Please see the toolbox article:

To cite: Zhou, Z., Chen, X., Zhang, Y., Hu, D., Qiao, L., Yu, R., Yap, P.-T., Pan, G., Zhang, H., Shen, D., 2020, A Toolbox for Brain Network Construction and Classification (BrainNetClass). Human Brain Mapping.


This work was supported by the NIH grants (EB022880, AG041721, AG049371, and AG042599). If you feel it is helpful for your research, please cite the following papers.


[1] Chen, X., Zhang, H., Gao, Y., Wee, C.Y., Li, G., Shen, D., Alzheimer’s Disease Neuroimaging, I., 2016. High-order resting-state functional connectivity network for MCI classification. Hum Brain Mapp 37, 3282-3296.

[2] Qiao, L., Zhang, H., Kim, M., Teng, S., Zhang, L., Shen, D., 2016. Estimating functional brain networks by incorporating a modularity prior. NeuroImage 141, 399-407.

[3] Zhang, Y., Zhang, H., Chen, X., Liu, M., Zhu, X., Lee, S.-W., Shen, D., 2019. Strength and Similarity Guided Group-level Brain Functional Network Construction for MCI Diagnosis. Pattern Recognition, 88, 421-430.

[4] Zhang, H., Chen, X., Zhang, Y., Shen, D., 2017. Test-retest reliability of “high-order” functional connectivity in young healthy adults. Frontiers in Neuroscience, 11:439.

[5] Zhang, Y., Zhang, H., Chen, X., Lee, S.-W., Shen, D., 2017. Hybrid High-order Functional Connectivity Networks Using Resting-state Functional MRI for Mild Cognitive Impairment Diagnosis, Scientific Reports, 7: 6530.

[6] Chen, X., Zhang, H., Shen, D., 2017. Hierarchical High-Order Functional Connectivity Networks and Selective Feature Fusion for MCI Classification. Neuroinformatics, 15(3):271-284.

[7] Yu, R., Zhang, H., An, L., Chen, X., Wei, Z., Shen, D., 2017. Connectivity strength-weighted sparse group representation-based brain network construction for MCI classification. Human Brain Mapping, 38(5): 2370-2383.

[8] Zhang, H., Chen, X., Shi, F., Li, G., Kim, M., Giannakopoulos, P., Haller, S., Shen, D., 2016. Topographic Information based High-Order Functional Connectivity and its Application in Abnormality Detection for Mild Cognitive Impairment, Journal of Alzheimer’s Disease, 54(3): 1095-1112.

[9] Zhou, Z., Chen, X., Zhang, Y., Hu, D., Qiao, L., Yu, R., Yap, P.-T., Pan, G., Zhang, H., Shen, D., 2020, A Toolbox for Brain Network Construction and Classification (BrainNetClass). Human Brain Mapping. DOI: 10.1002/hbm.24979.