Alzheimer’s disease (AD) is an irreversible progressive neurodegenerative disease. Intervention in the early stage of AD is a new path for AD treatment that is being explored. The behavioral and pathological effects of anodal transcranial direct current stimulation (AtDCS) at the early stage of AD in the mouse model, amyloid precursor protein (APP)/presenilin-1 (PS1) transgenic mice, were investigated based on our previous studies. Thirty-three 6-month-old male APP/PS1 mice were randomly divided into the model group (AD group), model + sham stimulation group (ADST group) and stimulation group (ADT group). Eleven 6-month-old male C57 wild-type mice were randomly selected as a control group (CTL group). The ADT group received 10 AtDCS sessions. The Morris water maze (MWM) task and novel object recognition (NOR) task were used to test mouse memory. Nissl staining, Western blot (WB), immunohistochemistry and immunofluorescence staining of β-amyloid (Aβ42), glial fibrillary acidic protein (GFAP) and NF200 were conducted for pathological analysis. The ADT group and the CTL group had a shorter escape latency and more platform-region crossings than the AD group and ADST group in the MWM. There was no significant difference in the discrimination index among the groups in the NOR task. Pathological analysis showed visible differences between the AD group and ADT group. This study revealed that early-stage APP/PS1 transgenic mice did not show recognition memory impairment. AtDCS effectively improved spatial learning and memory in the early-stage APP/PS1 transgenic mouse model of AD, alleviating Aβ burden and having a protective effect on neurons. AtDCS could improve AD-related symptoms by activating many glial cells to promote the degradation and clearance of Aβ or directly affecting production and degradation of Aβ to reduce glial activation. AtDCS is an effective means of early intervention in the early stage of AD.
BACKGROUND: Sleep staging is an important part of sleep research. Traditional automatic sleep staging based on machine learning requires extensive feature extraction and selection. OBJECTIVE: This paper proposed a deep learning algorithm without feature extraction based on one-dimensional convolutional neural network and long short-term memory. METHODS: The algorithm can automatically divide sleep into 5 phases including awake period, non-rapid eye movement sleep period (N1 ∼ N3) and rapid eye movement using the electroencephalogram signals. The raw signal was processed by the wavelet transform. Then, the processed signal was directly input into the deep learning algorithm to obtain the staging result. RESULTS: The accuracy of staging is 93.47% using the Fpz-Cz electroencephalogram signal. When using the Fpz-Cz and electroencephalogram signal, the algorithm can obtain the highest accuracy of 94.15%. CONCLUSION: These results show that this algorithm is suitable for different physiological signals and can realize end-to-end automatic sleep staging without any manual feature extraction.
The real-time sleep staging algorithm that can perform inference on mobile devices without burden is a prerequisite for closed-loop sleep modulation. However, current deep learning sleep staging models have poor real-time efficiency and redundant parameters. We propose a lightweight and high-performance sleep staging model named Micro SleepNet, which takes a 30-s electroencephalography (EEG) epoch as input, without relying on contextual signals. The model features a one-dimensional group convolution with a kernel size of 1 × 3 and an Efficient Channel and Spatial Attention (ECSA) module for feature extraction and adaptive recalibration. Moreover, the model efficiently performs feature fusion using dilated convolution module and replaces the conventional fully connected layer with Global Average Pooling (GAP). These design choices significantly reduce the total number of model parameters to 48,226, with only approximately 48.95 Million Floating-point Operations per Second (MFLOPs) computation. The proposed model is conducted subject-independent cross-validation on three publicly available datasets, achieving an overall accuracy of up to 83.3%, and the Cohen Kappa is 0.77. Additionally, we introduce Class Activation Mapping (CAM) to visualize the model’s attention to EEG waveforms, which demonstrate the model’s ability to accurately capture feature waveforms of EEG at different sleep stages. This provides a strong interpretability foundation for practical applications. Furthermore, the Micro SleepNet model occupies approximately 100 KB of memory on the Android smartphone and takes only 2.8 ms to infer one EEG epoch, meeting the real-time requirements of sleep staging tasks on mobile devices. Consequently, our proposed model has the potential to serve as a foundation for accurate closed-loop sleep modulation.
Objective: Sensory feedback of upper-limb prostheses is widely desired and studied. As important components of proprioception, position, and movement feedback help users to control prostheses better. Among various feedback methods, electrotactile stimulation is a potential method for coding proprioceptive information of a prosthesis. This study was motivated by the need for proprioception information for a prosthetic wrist. The flexion-extension (FE) position and movement information of the prosthetic wrist are transmitted back to the human body through multichannel electrotactile stimulation.Approach: We developed an electrotactile scheme to encode the FE position and movement of the prosthetic wrist and designed an integrated experimental platform. A preliminary experiment on the sensory threshold and discomfort threshold was performed. Then, two proprioceptive feedback experiments were performed: a position sense experiment (Exp 1) and a movement sense experiment (Exp 2). Each experiment included a learning session and a test session. The success rate (SR) and discrimination reaction time (DRT) were analyzed to evaluate the recognition effect. The acceptance of the electrotactile scheme was evaluated by a questionnaire.Main results: Our results showed that the average position SRs of five able-bodied subjects, amputee 1, and amputee 2 were 83.78, 97.78, and 84.44%, respectively. The average movement SR, and the direction and range SR of wrist movement in five able-bodied subjects were 76.25, 96.67%, respectively. Amputee 1 and amputee 2 had movement SRs of 87.78 and 90.00% and direction and range SRs of 64.58 and 77.08%, respectively. The average DRT of five able-bodied subjects was less than 1.5 s and that of amputees was less than 3.5 s.Conclusion: The results indicate that after a short period of learning, the subjects can sense the position and movement of wrist FE. The proposed substitutive scheme has the potential for amputees to sense a prosthetic wrist, thus enhancing the human-machine interaction.
Stroke leads to serious long-term disability. Electrical epidural cortical stimulation has made significant improvements in stroke rehabilitation therapy. We developed a preliminary wireless implantable passive interface, which consists of a stimulating surface electrode, receiving coil, and single flexible passive demodulated circuit printed by flexible printed circuit (FPC) technique and output pulse voltage stimulus by inductively coupling an external circuit. The wireless implantable board was implanted in cats’ unilateral epidural space for electrical stimulation of the primary visual cortex (V1) while the evoked responses were recorded on the contralateral V1 using a needle electrode. The wireless implantable board output stable monophasic voltage stimuli. The amplitude of the monophasic voltage output could be adjusted by controlling the voltage of the transmitter circuit within a range of 5–20 V. In acute experiment, cortico-cortical evoked potential (CCEP) response was recorded on the contralateral V1. The amplitude of N2 in CCEP was modulated by adjusting the stimulation intensity of the wireless interface. These results demonstrated that a wireless interface based on a microcoil array can offer a valuable tool for researchers to explore electrical stimulation in research and the dura mater-electrode interface can effectively transmit electrical stimulation.