Alzheimer’s disease (AD) is typical dementia, which is progressive and irreversible. Usually, the clinical diagnosis of patients is at a later stage, so early diagnosis can control the patient’s condition in time. The doctor is usually diagnosing the patient’s condition by 3D brain magnetic resonance imaging (MRI). However, because the 3D MRI structures of adjacent stages are almost similar, the multi-class diagnosis of AD becomes difficult. Therefore, there is a need to enhance the ability to extract more discriminative features from 3DMRI, promoting more accurate diagnosis. In addition, not only the entire MRI changes but also local areas in the MRI. Therefore, it is necessary to pay attention to changes in the entire image and local areas and to fuse features of different scales. In this paper, we propose an innovative convolutional network architecture for feature extraction and feature fusion. It consists of three modules: 1) a network based on ResNet-10, 2) 3D asymmetric convolution block (ACB), 3) multi-scale channel attentional feature fusion (MS-CAFF) module. The proposed model has been tested on the ADNI dataset and achieved an accuracy of 88.33%, which is nearly 2% higher than the latest research.

Convolutional neural networks(CNNs) are often over-parameterized and cannot apply to existing resource-limited artificial intelligence(AI) devices. Some methods are proposed to model compress the CNNs, but these methods are data-driven and often unable when lacking data. To solve this problem, in this paper, we propose a data-free model compression and acceleration method based on generative adversarial networks and network pruning(named DFNP), which can train a compact neural network only needs a pre-trained neural network. The DFNP consists of the source network, generator, and target network. First, the generator will generate the pseudo data under the supervise of the source network. Then the target network will get by pruning the source network and use these generated data for training. And the source network will transfer knowledge to the target network to promote the target network to achieve a similar performance of the source network. When the VGGNet- 19 is select as the source network, the target network trained by DFNP contains only 25% parameters and 65% calculations of the source network. Still, it retains 99.4% accuracy on the CIFAR-10 dataset without any real data.

Knowledge distillation is an extensively researched model compression technology, which uses a large teacher network to transmit information to a small student network. The critical point of the knowledge distillation method to improve the performance of the student network is to find an effective method to extract the information from the feature. The attention mechanism is a widely used feature processing method to process features effectively and obtain more expressive information. In this paper, we propose to use the dual attention mechanism in knowledge distillation to improve the performance of student networks, which extracts information from the spatial and channel dimensions of the feature. The channel dimension attention is search 'what' channel is more meaningful, and the spatial dimension attention is determine 'where' part of the feature is more expressive in a feature map. We have conducted extensive experiments on different datasets, shown that by implementing a dual attention mechanism to extract more expressive information for knowledge transfer, the student network can achieve performance beyond the teacher network.

With the rapid development of deep learning, convolutional neural networks(CNNs) have achieved great success. But these high-capability CNNs often with a huge burden of computation and memory, which hinders these CNNs from applying to practical application. To solve this problem, in this paper, we proposed a method to train a compact model with high-capacity. The student network with fewer parameters and calculations will learning from the knowledge of the teacher network with more parameters and calculations. To promote the ability of the student network, the more expressive knowledge is extracted from the middle-layer feature of neural networks by attention mechanism, and the knowledge transforms more effective from the teacher network to the student network by the positive- unlabeled(PU) classifier. We validate our method in extensive experiments, showing that it can train the student network to achieve significant performance superior to the teacher network.

How to design an anthropomorphic hand with a few actuators to replicate the grasping functions of the human hand is still a challenging problem. This paper aims to develop a general theory for designing the anthropomorphic hand and endowing the designed hand with natural grasping functions. A grasping experimental paradigm was set up for analyzing the grasping mechanism of the human hand in daily living. The movement relationship among joints in a digit, among digits in the human hand, and the postural synergic characteristic of the fingers were studied during the grasping. The design principle of the anthropomorphic mechanical digit that can reproduce the digit grasping movement of the human hand was developed. The design theory of the kinematic transmission mechanism that can be embedded into the palm of the anthropomorphic hand to reproduce the postural synergic characteristic of the fingers by using a limited number of actuators is proposed. The design method of the anthropomorphic hand for replicating human grasping functions was formulated. Grasping experiments are given to verify the effectiveness of the proposed design method of the anthropomorphic hand.

How to design a robotic hand reflecting human hand motion information as much as possible is a constantly exploring problem. In this paper, we propose an approach to mechanical design of compliant underactuated finger for prosthetic hand based on the decomposition of human hand movements. Hand movements are decomposed into primary and secondary motion in PCA coordinate system. The primary motion is achieved in free motion via actuators, and the secondary motion is implemented with mechanical compliance matching statistics characteristic of human motion data. Although analysis and design of single finger is always throughout this paper, the same method can be generalized to the whole hand design and the parameters design of other mechanical configuration.