• More accurate brain activation maps using Spatiotemporal models
  • Azam Saffar,1,* Yadollah Mehrabi,2
    1. Graduated, Department of Biostatistics, Faculty of Paramedical Sciences, Shahid Beheshti Medical University,
    2. Department of Epidemiology School of Public Health and Safety Shahid Beheshti University of Medical Sciences Tehran, Iran


  • Introduction: Precision and accuracy of the statistical analysis methods are essential for brain activation maps. Spatiotemporal correlation adjustment for considering this property that is embedded in fMRI data may increase their accuracy. The present study aimed to apply and assess the two fast spatiotemporal models to assess their accuracy.
  • Methods: We applied the spatiotemporal Gaussian process (STGP) and fast, fully Bayesian models as spatiotemporal classical and Bayesian, fast models for both simulated and experimental memory tfMRI data and compared the findings with General Linear Model (GLM). The models were fitted to the simulated data (1000 voxels,100 times points for 50 people) to assess their accuracy and precision. Functional and activation maps for all models were calculated in experimental data analysis.
  • Results: STGP and Bayesian models resulted in a higher Z-score in the whole brain, in the 1000 most activated voxels, and in the frontal lobe as the approved memory area. Based on the simulated data, these two models showed more accuracy and precision than the GLM models. However, their computational time was more than the GLM, as the price of model correction.
  • Conclusion: Spatiotemporal correlation consideration in the statistical models further improved the accuracy of models compared to the GLM model. This can result in more accurate activation maps.
  • Keywords: Brain mapping, fMRI data analysis, Accuracy Assessment, Spatiotemporal Correlation