Classification and transfer learning of sleep spindles based on convolutional neural networks.

Background: Sleep plays a critical role in human physiological and psychological health, and electroencephalography (EEG), an effective sleep-monitoring method, is of great importance in revealing sleep characteristics and aiding the diagnosis of sleep disorders. Sleep spindles, which are a typical phenomenon in EEG, hold importance in sleep science. Methods: This paper proposes a novel convolutional neural network (CNN) model to classify sleep spindles. Transfer learning is employed to apply the model trained on the sleep spindles of healthy subjects to those of subjects with insomnia for classification. To analyze the effect of transfer learning, we discuss the classification results of both partially and fully transferred convolutional layers. Results: The classification accuracy for the healthy and insomnia subjects' spindles were 93.68% and 92.77%, respectively. During transfer learning, when transferring all convolutional layers, the classification accuracy for the insomnia subjects' spindles was 91.41% and transferring only the first four convolutional layers achieved a classification result of 92.80%. The experimental results demonstrate that the proposed CNN model can effectively classify sleep spindles. Furthermore, the features learned from the data of the normal subjects can be effectively applied to the data for subjects with insomnia, yielding desirable outcomes. Discussion: These outcomes underscore the efficacy of both the collected dataset and the proposed CNN model. The proposed model exhibits potential as a rapid and effective means to diagnose and treat sleep disorders, thereby improving the speed and quality of patient care.

Clinical course and treatment efficacy of COVID-19 near Hubei Province, China: A multicentre, retrospective study.

Currently, COVID-19 has been reported in nearly all countries globally. To date, little is known about the viral shedding duration, clinical course and treatment efficacy of COVID-19 near Hubei Province, China. This multicentre, retrospective study was performed in 12 hospitals in Henan and Shaanxi Provinces from 20 January to 8 February 2020. Clinical outcomes were followed up until 26 March 2020. The viral shedding duration, full clinical course and treatment efficacy were analysed in different subgroups of patients. A total of 149 COVID-19 patients were enrolled. The median age was 42 years, and 61.1% (91) were males. Of them, 133 (89.3%) had fever, 131 of 144 (91%) had pneumonia, 27 (18.1%) required intensive care unit (ICU) management, 3 (2%) were pregnant, and 3 (2%) died. Two premature newborns were negative for SARS-CoV-2. In total, the median SARS-CoV-2 shedding period and clinical course were 12 (IQR: 9-17; mean: 13.4, 95% CI: 12.5, 14.2) and 20 (IQR: 16-24; mean: 21.2, 95% CI: 20.1, 22.3) days, respectively, and ICU patients had longer median viral shedding periods (21 [17-24] versus 11 [9-15]) and clinical courses (30 [22-33] vs. 19 [15.8-22]) than non-ICU patients (both p < .0001). SARS-CoV-2 clearances occurred at least 2 days before fatality in 3 non-survivors. Current treatment with any anti-viral agent or combination did not present the benefit of shortening viral shedding period and clinical course (all p > .05) in real-life settings. In conclusion, the viral shedding duration and clinical course in Henan and Shaanxi Provinces were shorter than those in Hubei Province, and current anti-viral therapies were ineffective for shortening viral shedding duration and clinical course in real-world settings. These findings expand our knowledge of the SARS-CoV-2 infection and may be helpful for management of the epidemic outbreak of COVID-19 worldwide. Further studies concerning effective anti-viral agents and vaccines are urgently needed.

[Effect of ultrasound microbubble carrying herpes simplex virus thymidine kinase on hepatocellular carcinoma in mice].

OBJECTIVE: To observe the effect of ultrasound microbubble carrying herpes simplex virus thymidine kinase hepatocellular carcinoma in mice. METHODS: Kunming mice were inoculated subcutaneously with H22 tumor cells. 40 male mice bearing subcutaneous hepatoma were randomized into 4 groups: PBS (group A), HSV1-TK (group B), HSV1-TK (group C), and microbubble carrying HSV1-TK (group D) were injected into the tail vein every 3 days. Mice in group C and D were exposed to ultrasound. The expression of TK protein was detected by western blot. Ganciclovir (GCV) was intraperitoneally injected at a dose of 100 mg x kg (-1) x d(-1) in group B, group C and group D. The tumor size was measured every 2 days. RESULTS: TK gene could be injected precisely into hepatocellular carcinoma with ultrasound monitor, and the expression of TK protein was found in all 4 groups. Expression in group D was higher than others (P < 0.05). The rate of tumor growth inhibition were 0 in group A, 3.90%+/-1.80% in group B, 22.70%+/-2.86% in group C, 41.25%+/-3.20% in group D (group B vs group C, P < 0.05; group D vs group C, P < 0.05; group D vs group B, P < 0.05). CONCLUSION: Ultrasound microbubble not only improve target gene therapy, but also enhance transfection efficiency.