Phase-change integrated photonics devices, evolving from the combinations of phase-change materials (PCMs) and silicon photonics, are subjecting to intensive research today. This is because phase-change integrated photonics devices inherit performances superiorities of both phase-change storage memory and silicon photonics in data transmission system. The programming mechanism of phase-change integrated photonic devices can be either optically driven switching or electrically driven switching. In comparison with its optical rival, electrically driven switching, usually realized by an integrated microheater, requires simpler programming scenario and less complex circuits and thus leads to a variety of microheater designs. To more comprehensively understand the physics behind different designs and to distinguish their performances variations, we first reviewed the operational principles of phase-change integrated photonics devices in association with common phase-change photonic materials. The current status of electrically driven switching-based devices is subsequently elaborated according to different microheaters, and their respective merits and drawbacks are interpreted. The technical challenges and future prospects of phase-change integrated photonics devices are eventually discussed.

Early prevention and detection of respiratory disease have attracted extensive attention due to the significant increase in people with respiratory issues. Restraining the spread and relieving the symptom of this disease is essential. However, the traditional auscultation technique demands a high-level medical skill, and computational respiratory sound analysis approaches have limits in constrained locations. A wearable auscultation device is required to real-time monitor respiratory system health and provides consumers with ease. In this work, we developed a Respiratory Sound Diagnosis Processor Unit (RSDPU) based on Long Short-Term Memory (LSTM). The experiments and analyses were conducted on feature extraction and abnormality diagnosis algorithm of respiratory sound, and Dynamic Normalization Mapping (DNM) was proposed to better utilize quantization bits and lessen overfitting. Furthermore, we developed the hardware implementation of RSDPU including a corrector to filter diagnosis noise. We presented the FPGA prototyping verification and layout of the RSDPU for power and area evaluation. Experimental results demonstrated that RSDPU achieved an abnormality diagnosis accuracy of 81.4 %, an area of 1.57 × 1.76 mm under the SMIC 130 nm process, and power consumption of 381.8 μW, which met the requirements of high accuracy, low power consumption, and small area.

Lung sounds, as an important physiological signal of human body, play a crucial role in the diagnosis and monitoring of respiratory diseases. In recent years, deep learning-based lung sound intelligent recognition technology has made significant progress in the field of medical auxiliary diagnosis, particularly with the deployment of deep learning models on edge devices to achieve local inference and low-latency diagnosis. This has become a key direction in the development of lung sound recognition systems. This paper systematically reviews the fundamental processes of lung sound recognition and analysis, focusing on the construction of lung sound classification models, model lightweight design, and the latest research progress in deploying these models on embedded systems, FPGA, and other hardware platforms. Additionally, the paper looks forward to the application prospects and development trends of this technology in the field of intelligent lung sound recognition. In particular, deployment pathways grounded in hardware–software co-design on embedded and edge platforms are poised to further advance health monitoring systems. This paper provides a comprehensive reference for understanding the key aspects of lung sound recognition technology, from algorithm research to hardware implementation.

Resistive Random-Access Memory (RRAM)-based Compute-in-Memory (CIM) architectures offer promising solutions for energy-efficient deep neural network (DNN) inference. However, conventional CIM accelerators suffer from high energy consumption due to frequent analog-to-digital (AD) and digital-to-analog (DA) signal conversions, especially when computing nonlinear activation functions (NAFs). This paper presents NeuAFG, a time-domain analog function generator designed to directly compute arbitrary NAFs in the analog domain. NeuAFG utilizes a one-hidden-layer ReLU neural network (ReLUNet) to approximate a variety of activation functions, incorporating two novel optimization algorithms: Low-Discrepancy Search (LDSearch) for efficient parameter initialization and KDE-FocusSampler for focused sampling of regions that are difficult to approximate. The hardware implementation integrates RRAM-based components and employs a Pulse-In-Pulse-Out method for robust time-domain computation. Experimental results demonstrate that NeuAFG reduces energy consumption in mainstream CIM accelerators such as ISAAC and RAELLA by 1.03× to 2.20×, with less than 1% accuracy loss in DNN inference tasks.