data processing assistant for resting-state fmri advanced edition (dparsfa) toolbox  (MathWorks Inc)

Journal: Frontiers in Human Neuroscience

Article Title: Diagnostic model optimization method for ADHD based on brain network analysis of resting-state fMRI images and transfer learning neural network

doi: 10.3389/fnhum.2022.1005425

Figure Lengend Snippet: ADHD classification model based on TLNN. The model training process including: (1) loading the pre-trained model, the pre-trained parameters were transferred to the target domain (fMRI image); (2) the hyperparameters obtained from the natural images were fine-tuned; (3) the VGGNet or ResNet50 models are trained on the large dataset ImangeNet; (4) the weight parameters completed by training are transferred to the fMRI image classification task; (5) the middle and lower layers of the pre-trained model are used as the feature extractor of the target task; (6) the extracted features are nonlinear mapped through the fully connected layer; and (7) the final classification result is obtained. Conv means the number of convolution kernels. FCLs means fully connected layers.

Article Snippet: We ran the Data Processing Assistant for Resting-State fMRI (DPARSFA) on the platform MATLAB (R2016a) for data preprocessing: (1) ensure each point in the image comes from the actual signal at the same time by temporal layer correction; (2) through head movement realignment, subjects with more than 2 mm translation in the X-Y-Z axis or more than 2° rotation were excluded; (3) apply spatial normalization; and (4) conduct full-width-and-half-height Gaussian kernel smoothing on the images, with a kernel size of 8 × 8 × 8 mm, to reduce the impact of the noise and improve its signal-to-noise ratio (Chao-Gan and Yu-Feng, ; Yan et al., ; Sun et al., ).

Techniques: