RESUMO
OBJECTIVE: To compare the accuracy and generalizability of an automated deep neural network and the Philip Sleepware G3™ Somnolyzer system (Somnolyzer) for sleep stage scoring using American Academy of Sleep Medicine (AASM) guidelines. METHODS: Sleep recordings from 104 participants were analyzed by a convolutional neural network (CNN), the Somnolyzer and skillful technicians. Evaluation metrics were derived for different combinations of sleep stages. A further comparison between the Somnolyzer and the CNN model using a single-channel signal as input was also performed. Sleep recordings from 263 participants with a lower prevalence of OSA served as a cross-validation dataset to validate the generalizability of the CNN model. RESULTS: The overall agreement between automated and manual scoring for sleep staging in 104 participants outperformed that of the Somnolyzer according to various metrics (accuracy: 81.81 % vs. 77.07 %; F1: 76.36 % vs. 73.80 %; Cohen's kappa: 0.7403 vs. 0.6848). The results showed that the left electrooculography (EOG) single-channel model had minor advantages over the Somnolyzer. In terms of consistency with manual sleep staging, the CNN model demonstrated superior performance in identifying more pronounced sleep transitions, particularly in the N2 stage and sleep latency metrics. Conversely, the Somnolyzer showed enhanced proficiency in the analysis of REM stages, notably in measuring REM latency. The accuracy in the cross-validation set of 263 participants was also above 80 %. CONCLUSIONS: The CNN-based automated deep neural network outperformed the Somnolyzer and is sufficiently accurate for sleep study analyses using the AASM classification criteria.
RESUMO
Metamaterial perfect absorbers (MPAs) are attractive platforms for the unique manipulation of electromagnetic waves from near-field to far-field. Narrow-band MPAs are particularly intriguing for their potential applications as thermal emitters or biosensors. In this work, we proposed ultra-narrow-band MPAs based on surface lattice resonance (SLR) modes of WS2 nanodisk arrays on gold films. The SLR modes stem from the coupling between the magnetic dipole modes of individual nanodisks and the Rayleigh anomaly of the array giving rise to high quality-factor resonances. With proper design of the nanodisk array, an ultra-narrow-band of 15 nm is achieved in the near infrared wavelength range. The underneath gold film provides the loss channel converting the incident light within the narrow band into heat in the gold film, effectively creating a perfect absorber. Systematic numerical simulations were performed to investigate the effects of the geometrical parameters on their optical properties, demonstrating the great tunability of this type of MPAs as well as their potential for engineering light-matter interactions.
RESUMO
Plasmonic nanoantennas can significantly enhance the light-matter interactions at the nanoscale, and as a result have been used in a variety of applications such as sensing molecular vibrations in the infrared range. Indium-tin-oxide (ITO) shows metallic behavior in the infrared range, and can be used for alternative plasmonic materials. In this work, we numerically studied the optical properties of hexagonal ITO nanodisk and nanohole arrays in the mid-infrared. Field enhancement up to 10 times is observed in the simulated ITO nanostructures. Furthermore, we demonstrated the sensing of the surface phonon polariton from a 2-nm thick SiO2 layer under the ITO disk arrays. Such periodic arrays can be readily fabricated by colloidal lithography and dry etching techniques; thus, the results shown here can help design efficient ITO nanostructures for plasmonic infrared applications.
