Artificial intelligence (AI) powered Internet of things (IoT) enabled medical devices can play a vital role in data-driven predictive health intelligence networks. Modern wearable devices are capable of continuously sensing single and multichannel signals. Furthermore, they are able to process, analyze, and store. In addition, they can transmit the sensed or processed multimodal data to the edge computing or cloud computing server for further analysis and diagnosis of various types of communicable and non-communicable diseases.

However, there are many design, development, and deployment challenges that need to be addressed to reduce false alarms of unsupervised health monitoring systems under different types of noises and artifacts. These false alarms are unavoidable in the case of wearable devices and unsupervised health monitoring systems. Contextual information fusion for improving the accuracy of the diagnostic system and reducing energy consumption to improve battery life by exploring data acquisition, lightweight signal processing, and machine learning techniques play primary roles in deciding the future of long-term battery-powered IoT monitoring devices and unsupervised monitoring devices.

The aim of this Special Issue is to bring together original research articles and review articles highlighting recent research advances in low-power biomedical circuits in sensory circuits and systems. Submissions discussing energy-efficient signal processing and machine learning architectures for non-invasive contact sensing, processing and analyzing biosignals such as electrocardiogram (ECG), phonocardiogram (PCG), photoplethysmogram (PPG), ballistocardiogram (BCG), seismocardiogram (SCG) are also welcome. Research can also include respiratory signals and other non-biomedical signals which can provide relevant information (e.g., physical activity, personal health habits) for developing more accurate and reliable on-device, edge- and cloud-based intelligent diagnostic systems.

Potential topics include but are not limited to the following:

• Low-power sensory circuits and systems
• Low-power biomedical circuits and systems
• Energy-efficient biosignal assessment systems
• Effective and efficient biosignal denoising techniques
• Low-power very large-scale integration (VLSI) architectures for spectrograms
• Image extraction with deep neural network (DNN) low-power
• VLSI architectures for machine learning
• Classifier contextual information extraction and fusion for improving accuracy and reliability of diagnostic systems
• Efficient signal and machine learning techniques to address the challenges of vital sign extraction under adverse recording conditions
• Energy harvesting circuits and systems for wearable devices