Diagnostics 2024 : Special Issue: Human Health, Well-Being and Activity Recognition with Wearable Sensors

Dear Colleagues,

Small, lightweight, cheap sensors and processors are embedded in various smart devices, such as smartwatches, phones, glasses and even clothes. They produce large amounts of data about us, the environment and their usage patterns. These data can be used to infer information about human health, well-being and activities, enabling a rich set of applications for assisting users.

Such mobile/wearable artificial intelligence applications utilize machine learning techniques for data processing. Besides traditional machine learning approaches, the use of deep learning, federated learning and split learning are also emerging in processing wearable sensor data. On the other hand, wearable sensing devices are still resource-constrained devices in terms of computation, storage and battery. Hence, they are mostly used to collect data and then the data is transferred to the Cloud where both training and inference are performed. However, recent efforts have been made to simplify and run machine learning and even deep learning algorithms on these devices, thanks to the introduction of dedicated hardware accelerators, the latest neural network (NN) hardware, multi-core processors and larger memory becoming more common in the devices. Distributed machine learning approaches such as federated learning take advantage of retaining data, which are often personal, on the device by enabling only the sharing of the model parameters with a server.

In this Special Issue, we focus on the most popular application areas of wearable sensing: health, well-being and activity recognition, and seek research contributions that enable the use of recent machine learning approaches, some of which are named above.

Topics of interest include, but are not limited to, the following application fields:

Sensor-based machine learning;
Wearable sensors, motion sensors, force/pressure sensors and EMG sensors for human well-being, health and activities;
Machine learning and deep learning for wearable data analysis;
Attention models for wearable data analysis;
Health and activity monitoring in the working environment;
Mhealth and/or eHealth solutions using wearable sensors;
Behavior recognition;
Time series analysis on wearable sensor data;
Model adaptation and compression for resource constrained wearables;
Edge intelligence and edge computing for wearable data analysis;
On-device machine learning with resource constrained wearables;
Federated Learning for wearable data analysis.