The Variability Insights Project

Dear Colleagues,

We are pleased to announce the launch of the Variability Insights Project (VIP). VIP is a crowd-contributed data-mining platform for neuroimaging-based subject classification.

What can you obtain?

Users on this platform can obtain a full view of inter-individual variability and suggestions regarding potential subgroups from the functional and/or structural maps they upload. These pieces of information help users to oversee the patterns of individual variability in their data, better interpret the variability in neuroimaging data, and discover neuroimaging markers for classifying the subjects. All kinds of metrics computed from structural and/or functional images can be uploaded, such as ICA component maps, activation maps, functional connectivity maps, reho maps, alff maps, grey matter maps, and so on. Users are encouraged to share structural and functional maps in their study, as well as basic demographical data. Upon accumulating a large amount of structural, functional, and behavioral data, VIP aims to apply data-driven and machine learning approach to associate features from the above modalities. With these efforts, we attempt to make neuroimaging an effective approach for discovering novel knowledge, instead of merely examine hypotheses proposed from behavioral differences. Meanwhile, VIP provides customized data-mining service for users.


 

Why do we launch VIP?

Linking minds and behavior to brain mechanisms is a long-lasting endeavor in our progress to understand ourselves. Neuroimaging techniques have helped to advance our understanding, but their contributions are confined by limited data and the traditional way of inference. Due to limitations in data acquisition and data-mining capability, most neuroimaging studies examine hypotheses proposed from behavioral differences. However, we are still unable to propose hypotheses based on variability in neuroimaging data per se. In other words, up-to-date, neuroimaging method only serves to examine hypotheses from other field, but has not been able to discover knowledge based neural mechanisms.

The intention of VIP is to promote the advantage of neuroimaging data, such as rich information and strong physiological relevance, making it an effective knowledge-discovering approach, and to propose novel brain-behavior associations based on variability in neuroimaging data. Specifically, we start from analyzing individual variability in neuroimaging data, detecting features that stably reflect subject groupings, and then establish brain-behavior associations. Compared to behavior-brain associations proposed from behavioral differences, the brain-behavior associations have much more solid biological basis and therefore may have larger specificity and validity. This endeavor requires large amount of neuroimaging data and proper data-mining methods. We have established a “discover-validate” paradigm for data-mining and have examined its effectiveness in a series of real practice. Therefore, we launch this VIP platform to serve users by analyzing variability in their neuroimaging data while accumulating data for “knowledge-discover” in neuroimaging data.

We hope you enjoy the VIP service and share your results. Due to temporary limitation on data-upload capacity, please contact Dr. Zhi Yang (yangz -at- psych.ac.cn) for instructions to upload your data and obtain the results.

Best Regards,

Zhi Yang Ph.D.

Associate Professor

Institute of Psychology, Chinese Academy of Sciences

16 Lincui Road, Chaoyang District, Beijing 100101, China

 

Top Ten VIP References

[2016] Yang Z, Zuo XN*, McMahon KL, Craddock RC, Kelly C, de Zubicaray GI, Hickie I, Bandettini PA, Castellanos FX, Milham MP*, Wright MJ (2016). Genetic and Environmental Contributions to Functional Connectivity Architecture of the Human Brain. Cereb Cortex. Published Online, doi: 10.1093/cercor/bhw027.

[2015] Yang Z*, Qiu J, Wang P, Liu R, Zuo X* (2016). Brain Structure-Function Associations Identified in Large-Scale Neuroimaging Data. Brain Structure & Function. Published Online, doi:10.1007/s00429-015-1177-6.

[2015] Xu T, Yang Z (co-first author), Jiang L, Xing XX, Zuo XN (2015). A Connectome Computation System for discovery science of brain. Science Bulletin 60, 86-95.

[2015] 杨志*, 左西年* (2015). 神经影像大数据与心脑关联: 方法学框架与应用. 科学通报60, 966-975.

[2014] Yang Z*, Huang Z, Gonzalez-Castillo J, Dai R, Northoff G, Bandettini P (2014). Using fMRI to decode true thoughts independent of intention to conceal. NeuroImage 99, 80-92.

[2014] Yang Z*, Chang C, Xu T, Jiang L, Handwerker D, Castellanos F, Milham M, Bandettini P, Zuo X* (2014). Connectivity trajectory across lifespan differentiates the precuneus from the default network. NeuroImage 89, 45-56.

[2014] Yang Z, Xu Y*, Xu T, Hoy C, Handwerker D, Chen G, Northoff G, Zuo X*, Bandettini P (2014). Brain network informed subject community detection in early-onset schizophrenia. Scientific Reports 4, 5549.

[2012] Yang Z*, Zuo X, Wang P, Li Z, Laconte S, Bandettini PA, Hu X (2012). Generalized RAICAR: Discover homogeneous subject (sub)groups by reproducibility of their intrinsic connectivity networks.  NeuroImage 63, 403-414.

[2012] Xu G, Jiang Y, Ma L, Yang Z*, Weng X* (2012). Similar spatial patterns of neural coding of category selectivity in FFA and VWFA under different attention conditions.  Neuropsychologia 50, 862-868.

[2008] Yang Z, LaConte S, Weng X, Hu X* (2008). Ranking and averaging independent component analysis by reproducibility (RAICAR).  Human Brain Mapping 29, 711-725.