A5082109564- Functional Brain Connectivity Studies
- Brain Tumor Detection and Classification
- Neural dynamics and brain function
- EEG and Brain-Computer Interfaces
- Emotion and Mood Recognition
- Domain Adaptation and Few-Shot Learning
- Educational Reforms and Innovations
- Neural Networks and Applications
- Medical Research and Treatments
- Advanced Neural Network Applications
- Neonatal and fetal brain pathology
Ocean University of China
2025
Shanghai University of Medicine and Health Sciences
2024
University of Shanghai for Science and Technology
2023-2024
Jiading District Central Hospital
2024
This study aims to enhance the accuracy of depression detection by leveraging representation learning from audio data. The data speech sets are sparse and costly annotate. Therefore, a self-supervised pre-training approach is employed improve performance, generalization capability, training efficiency downstream tasks. When processing unlabeled data, pre-trained representations based on may be interfered with noisy if there significant amount noise or errors present. Consequently, it...
As an important biomarker of neural aging, the brain age reflects integrity and health human brain. Accurate prediction could help to understand underlying mechanism aging. In this study, a cross-stratified ensemble learning algorithm with staking strategy was proposed obtain derived predicted difference (PAD) using T1-weighted magnetic resonance imaging (MRI) data. The approach characterized as by implementing two modules: one three base learners 3D-DenseNet, 3D-ResNeXt, 3D-Inception-v4;...
In neural network binarization, BinaryConnect (BC) and its variants are considered the standard. These methods apply sign function in their forward pass respective gradients backpropagated to update weights. However, derivative of is zero whenever defined, which consequently freezes training. Therefore, implementations BC (e.g., BNN) usually replace backward computation with identity or other approximate gradient alternatives. Although such practice works well empirically, it largely a...
Alzheimer's disease (AD) represents a prevalent, progressive neurodegenerative ailment marked by the gradual deterioration of memory and cognitive faculties. Resting-state functional magnetic resonance imaging (rs-fMRI) offers good specificity in AD reflecting early changes brain network. As result, combining popular deep learning methods with rs-fMRI network features has attracted wide attention. In our experiment, A cohort comprising 325 participants, sourced from Disease Neuroimaging...