Lung sounds remain vital in clinical diagnosis as they reveal associations with pulmonary pathologies. With COVID-19 spreading across the world, it has become more pressing for medical professionals to better leverage artificial intelligence for faster and more accurate lung auscultation. This research aims to propose a feature engineering process that extracts the dedicated features for the depthwise separable convolution neural network (DS-CNN) to classify lung sounds accurately and efficiently. We extracted a total of three features for the shrunk DS-CNN model: the short-time Fourier-transformed (STFT) feature, the Mel-frequency cepstrum coefficient (MFCC) feature, and the fused features of these two. We observed that while DS-CNN models trained on either the STFT or the MFCC feature achieved an accuracy of 82.27% and 73.02%, respectively, fusing both features led to a higher accuracy of 85.74%. In addition, our method achieved 16 times higher inference speed on an edge device and only 0.45% less accuracy than RespireNet. This finding indicates that the fusion of the STFT and MFCC features and DS-CNN would be a model design for lightweight edge devices to achieve accurate AI-aided detection of lung diseases.

No abstract

The dynamic behavior of a monolithic dual-wavelength distributed feedback laser was fully investigated and mapped. The combination of different driving currents for master and slave lasers can generate a wide range of different operational modes, from single mode, period 1 to chaos. Both the optical and microwave spectrum were recorded and analyzed. The detected single mode signal can continuously cover from 15GHz to 50GHz, limited by photodetector bandwidth. The measured optical four-wave-mixing pattern indicates that a 70GHz signal can be generated by this device. By applying rate equation analysis, the important laser parameters can be extracted from the spectrum. The extracted relaxation resonant frequency is found to be 8.96GHz. With the full operational map at hand, the suitable current combination can be applied to the device for proper applications.

No abstract

We report a terahertz quantum-cascade vertical-external-cavity surface-emitting laser (QC-VECSEL) emitting around 1.9 THz with up to 10% continuous fractional frequency tuning of a single laser mode. The device shows lasing operation in pulsed mode up to 102 K in a high-quality beam, with the maximum output power of 37 mW and slope efficiency of 295 mW/A at 77 K. Challenges for up-scaling the operating wavelength in QC metasurface VECSELs are identified.