The complexity of raw functional magnetic resonance imaging (fMRI) data with artifacts leads to significant challenges in multioperations with these data. FMRI data analysis is extensively used in neuroimaging fields, but the tools for processing fMRI data are lacking. A novel APP DESIGNER conversion, preprocessing, and postprocessing of fMRI (CPREPP fMRI) tool is proposed and developed in this work. This toolbox is intended for pipeline fMRI data analysis, including full analysis of fMRI data, starting from DICOM conversion, then checking the quality of data at each step, and ending in postprocessing analysis. The CPREPP fMRI tool includes 12 conversions of scientific processes that reflect all conversion possibilities among them. In addition, specific preprocessing order steps are proposed on the basis of data acquisition mode (interleaved and sequential modes). A severe and crucial comparison between statistical parametric and nonparametric mapping approaches of second-level analysis is presented in the same tool. The CPREPP fMRI tool can provide reports to exclude subjects with the extreme movement of the head during the scan, and a range of fMRI images are generated to verify the normalization effect easily. Real fMRI data are used in this work to prepare fMRI data tests. The experiment stimuli are chewing and biting, and the data are acquired from the National Magnetic Resonance Research (UMRAM) Center in Ankara, Turkey. A free dataset is used to compare the methods for postprocessing fMRI tests. INDEX TERMSAnalyze data (img/hdr), DICOM, fMRI, NIFTI, parametric and nonparametric approaches.

The RGNG can detect the active zones in the brain, analyze brain function, and determine the optimal number of underlying clusters in fMRI datasets. This algorithm can define the positions of the center of an output cluster corresponding to the minimal MDL value.

In this work, a novel conversion and visualization fMRI (VCfMRI) toolbox is proposed. The VCfMRI tool is enabled to read, write 3-D volume data (.dcm, .nii, .img, hdr and .mat format) as well as multi conversion operations between them are performed in the same package. In the current work, real fMRI data are used and all data are acquired by MRI scanner type Siemens/3T in National Magnetic Resonance Research Center (UMRAM)-Bilkent University. About 62 analyses functions have been implemented and incorporated in analysis about 7 GUI tools for multiple conversions of fMRI modalities, reading/writing and viewing in all fMRI data formats, visualizing 3-dimensional (sagittal, coronal and horizontal slices) statistical and non-statistical neuroimaging, thresholding and overlaying viewing. The presented package is a simple tool to address several issues that related to complexity in visualizing and conversion between multi-formats of fMRI data. This work enables the user to visualize and deals with fMRI data in an easy way especially for physicians, healthcare specialists and researchers whose faced challenges about how handling with these type of data.

Clustering algorithms are used to group data depending on a distance. Best clustering analysis should be resisting the presence of outliers, less sensitive to initialization as well as the input sequence ordering. This article compares the performance among three of prototype‐based unsupervised clustering algorithms: Neural Gas (NG), Growing Neural Gas (GNG) and Robust Growing Neural Gas (RGNG). Based on NG and GNG, there are different clustering algorithms proposed and suggested in different literatures. So, in this work a comparison between the two basic clustering algorithms NG and GNG have presented using the performance evaluation of these techniques, in contrast to the RGNG which was proposed within the GNG. Another comparison due to the MDL criterion between RGNG that used MDL value as the clustering validity index, versus GNG and NG combined with MDL. Statistical estimations are applied to explain the meaning of the output results when these algorithms fed to the synthetic 2D dataset. Moreover, a simple software package is designed and implemented as an automatic clustering model for any dataset to use as a part of the neural network course. NG, GNG and RGNG algorithms are performed in the same package using a MATLAB‐based Graphical User Interface (GUI) tool. This visual tool lets the students/ researchers visualize the desired results using plots also clicking a few buttons.

Background: Cluster analysis is a robust tool for exploring the underlining structures in data and grouping them with similar objects. In the researches of Functional Magnetic Resonance Imaging (fMRI), clustering approaches attempt to classify voxels depending on their time-course signals into a similar hemodynamic response over time. Objective: In this work, a novel unsupervised learning approach is proposed that relies on using Enhanced Neural Gas (ENG) algorithm in fMRI data for comparison with Neural Gas (NG) method, which has yet to be utilized for that aim. The ENG algorithm depends on the network structure of the NG and concentrates on an efficacious prototype-based clustering approach. Methods: The comparison outcomes on real auditory fMRI data show that ENG outperforms the NG and statistical parametric mapping (SPM) methods due to its insensitivity to the ordering of input data sequence, various initializations for selecting a set of neurons, and the existence of extreme values (outliers). The findings also prove its capability to discover the exact and real values of a cluster number effectively. Results: Four validation indices are applied to evaluate the performance of the proposed ENG method with fMRI and compare it with a clustering approach (NG algorithm) and model-based data analysis (SPM). These validation indices include the Jaccard Coefficient (JC), Receiver Operating Characteristic (ROC), Minimum Description Length (MDL) value, and Minimum Square Error (MSE). Conclusion: The ENG technique can tackle all shortcomings of NG application with fMRI data, identify the active area of the human brain effectively, and determine the locations of the cluster center based on the MDL value during the process of network learning.