International Journal of Neural Systems最新文献

This paper proposes a generative model and transfer learning powered system for classification of Scanning Electron Microscope (SEM) images of defective nanofibers (D-NF) and nondefective nanofibers (ND-NF) produced by electrospinning (ES) process. Specifically, a conditional-Generative Adversarial Network (c-GAN) is developed to generate synthetic D-NF/ND-NF SEM images. A transfer learning-oriented strategy is also proposed. First, a Convolutional Neural Network (CNN) is pre-trained on real images. The transfer-learned CNN is trained on synthetic SEM images and validated on real ones, reporting accuracy rate up to 95.31%. The achieved encouraging results endorse the use of the proposed generative model in industrial applications as it could reduce the number of needed laboratory ES experiments that are costly and time consuming.

Epilepsy is a neurological disorder caused by brain dysfunction, which could cause uncontrolled behavior, loss of consciousness and other hazards. Electroencephalography (EEG) is an indispensable auxiliary tool for clinical diagnosis. Great progress has been made by current seizure identification methods. However, the performance of the methods on different patients varies a lot. In order to deal with this problem, we propose an automatic seizure identification method based on brain connectivity learning. The connectivity of different brain regions is modeled by a graph. Different from the manually defined graph structure, our method can extract the optimal graph structure and EEG features in an end-to-end manner. Combined with the popular graph attention neural network (GAT), this method achieves high performance and stability on different patients from the CHB-MIT dataset. The average values of accuracy, sensitivity, specificity, F1-score and AUC of the proposed model are 98.90%, 98.33%, 98.48%, 97.72% and 98.54%, respectively. The standard deviations of the above five indicators are 0.0049, 0.0125, 0.0116 and 0.0094, respectively. Compared with the existing seizure identification methods, the stability of the proposed model is improved by 78-95%.

Depression is a common mental disease that has a tendency to develop at a younger age. Early detection of depression with psychological intervention may effectively prevent youth suicide. The establishment of the computer-aided model may be efficient for early detection. However, the existing methods of automatic detection for depression mostly rely on unimodal data. Clinical research shows that patients with depression have specificity in speech, text, expression, and other modal data. Multimodal machine learning is emerging but not yet widely used for the detection of psychiatric disorders. The problem of existing multimodal detection models is that only global or local information is considered in feature fusion, which leads to the low accuracy of the depression detection model. Therefore, this study constructs an automatic detection model based on multimodal machine learning for adolescent depression. The proposed method first extracted four features from audio and text globally and locally; then construct a coarse-grained fusion model and fine-grained fusion model base on these four features; and fuse the coarse-grained and the fine-grained fusion model finally. Experiments on the real-world dataset demonstrate that the proposed method could improve the accuracy of depression detection automatically.

Background: Where self-report is unfeasible or observations are difficult, physiological estimates of pain are needed. Methods: Pain-data from 30 healthy adults were gathered to create a database of physiological pain responses. A model was then developed, to analyze pain-data and visualize the AI-estimated level of pain on a mobile app. Results: The initial low precision and F1-score of the pain classification algorithm were resolved by interpolating a percentage of similar data. Discussion: This system presents a novel approach to assess pain in noncommunicative people with the use of a sensor sock, AI predictor and mobile app. Performance analysis and the limitations of the AI algorithm are discussed.

Feature extraction is an essential procedure in the detection and recognition of epilepsy, especially for clinical applications. As a type of multichannel signal, the association between all of the channels in EEG samples can be further utilized. To implement the classification of epileptic seizures from the nonseizures in EEG samples, one graph convolutional neural network (GCNN)-based framework is proposed for capturing the spatial enhanced pattern of multichannel signals to characterize the behavior of EEG activity, which is capable of visualizing the salient regions in each sequence of EEG samples. Meanwhile, the presented GCNN could be exploited to discriminate normal, ictal and interictal EEGs as a novel classifier. To evaluate the proposed approach, comparison experiments were conducted between state-of-the-art techniques and ours. From the experimental results, we found that for ictal and interictal EEG signal discrimination, the presented approach can achieve a sensitivity of 98.33%, specificity of 99.19% and accuracy of 98.38%.

A practical problem in supervised deep learning for medical image segmentation is the lack of labeled data which is expensive and time-consuming to acquire. In contrast, there is a considerable amount of unlabeled data available in the clinic. To make better use of the unlabeled data and improve the generalization on limited labeled data, in this paper, a novel semi-supervised segmentation method via multi-task curriculum learning is presented. Here, curriculum learning means that when training the network, simpler knowledge is preferentially learned to assist the learning of more difficult knowledge. Concretely, our framework consists of a main segmentation task and two auxiliary tasks, i.e. the feature regression task and target detection task. The two auxiliary tasks predict some relatively simpler image-level attributes and bounding boxes as the pseudo labels for the main segmentation task, enforcing the pixel-level segmentation result to match the distribution of these pseudo labels. In addition, to solve the problem of class imbalance in the images, a bounding-box-based attention (BBA) module is embedded, enabling the segmentation network to concern more about the target region rather than the background. Furthermore, to alleviate the adverse effects caused by the possible deviation of pseudo labels, error tolerance mechanisms are also adopted in the auxiliary tasks, including inequality constraint and bounding-box amplification. Our method is validated on ACDC2017 and PROMISE12 datasets. Experimental results demonstrate that compared with the full supervision method and state-of-the-art semi-supervised methods, our method yields a much better segmentation performance on a small labeled dataset. Code is available at https://github.com/DeepMedLab/MTCL.

Identifying brain abnormalities in autism spectrum disorder (ASD) is critical for early diagnosis and intervention. To explore brain differences in ASD and typical development (TD) individuals by detecting structural features using T1-weighted magnetic resonance imaging (MRI), we developed a deep learning-based approach, three-dimensional (3D)-ResNet with inception (I-ResNet), to identify participants with ASD and TD and propose a gradient-based backtracking method to pinpoint image areas that I-ResNet uses more heavily for classification. The proposed method was implemented in a preschool dataset with 110 participants and a public autism brain imaging data exchange (ABIDE) dataset with 1099 participants. An extra epilepsy dataset with 200 participants with clear degeneration in the parahippocampal area was applied as a verification and an extension. Among the datasets, we detected nine brain areas that differed significantly between ASD and TD. From the ROC in PASD and ABIDE, the sensitivity was 0.88 and 0.86, specificity was 0.75 and 0.62, and area under the curve was 0.787 and 0.856. In a word, I-ResNet with gradient-based backtracking could identify brain differences between ASD and TD. This study provides an alternative computer-aided technique for helping physicians to diagnose and screen children with an potential risk of ASD with deep learning model.

Hippocampal pyramidal cells and interneurons play a key role in spatial navigation. In goal-directed behavior associated with rewards, the spatial firing pattern of pyramidal cells is modulated by the animal's moving direction toward a reward, with a dependence on auditory, olfactory, and somatosensory stimuli for head orientation. Additionally, interneurons in the CA1 region of the hippocampus monosynaptically connected to CA1 pyramidal cells are modulated by a complex set of interacting brain regions related to reward and recall. The computational method of reinforcement learning (RL) has been widely used to investigate spatial navigation, which in turn has been increasingly used to study rodent learning associated with the reward. The rewards in RL are used for discovering a desired behavior through the integration of two streams of neural activity: trial-and-error interactions with the external environment to achieve a goal, and the intrinsic motivation primarily driven by brain reward system to accelerate learning. Recognizing the potential benefit of the neural representation of this reward design for novel RL architectures, we propose a RL algorithm based on [Formula: see text]-learning with a perspective on biomimetics (neuro-inspired RL) to decode rodent movement trajectories. The reward function, inspired by the neuronal information processing uncovered in the hippocampus, combines the preferred direction of pyramidal cell firing as the extrinsic reward signal with the coupling between pyramidal cell-interneuron pairs as the intrinsic reward signal. Our experimental results demonstrate that the neuro-inspired RL, with a combined use of extrinsic and intrinsic rewards, outperforms other spatial decoding algorithms, including RL methods that use a single reward function. The new RL algorithm could help accelerate learning convergence rates and improve the prediction accuracy for moving trajectories.