Tutorials

The wearable systems presented include the Automatic Ingestion Monitor (AIM-2) for monitoring diet, nutrition, and ingestive behavior; the Screen Time Tracker (STT) for quantifying screen exposure in children aged 2–5 years; SmartStep, an insole-based sensor for monitoring physical activity and gait, and its successor, Smart LaceLock; the Personal Automatic Cigarette Tracker (PACT) for monitoring combustible cigarette consumption; and the Flexible Robust Instrumentation of Electronic Nicotine Delivery Systems (FRIENDS) for monitoring vaping behavior.

The tutorial will address sensor design considerations, embedded processing, data acquisition, and power management, as well as signal processing and ML/AI pipelines for behavior recognition and event detection. Lessons learned from real-world deployments will be discussed, with demonstrations of selected sensor systems.

The tutorial provides an overview of state-of-the-art hardware solutions for sign language data acquisition, including wearable sensors, vision-based systems, depth cameras, and inertial measurement units, with a focus on their metrological characteristics and limitations. From a methodological perspective, the tutorial reviews algorithmic approaches for sign language analysis, covering feature extraction, signal preprocessing, classification, and machine learning pipelines commonly adopted in AI-based recognition systems. Particular attention is devoted to the role of data quality, repeatability, and uncertainty, highlighting how metrological aspects directly influence the performance and reliability of AI models.

A hands-on session is included, where participants will work with pre-acquired datasets to explore supervised classification techniques using MATLAB’s Classification Learner toolbox. Through practical examples focused on the recognition of sign language alphabet letters, participants will gain insight into model training, validation, performance evaluation, and the critical impact of dataset quality before AI training.

Today the elderly are more than the young but the funds for the health system are decreasing. The medicine paradigm must be rethought. E-Health can be the solution to support for the delocalization of some medical services, The ability to combine the power of AI algorithms and data from different sensors and databases can greatly increase the reliability of the final choice of the right therapy. This is the new Medicine 4.0. The digitalization of the processes and the improvement of technology allow interfacing the human body with computers and Artificial Intelligence allows you to work with a large amount of data (big data) and identify unknown correlations between the parameters to allow a new diagnosis. Several new perspectives will be discussed in this presentation. We will investigate both new technologies showing wearable devices that can be used both to monitor patients at home (this topic was very important with the Covid 19) and Artificial Intelligence applied to medical image processing to perform remote diagnoses (once again used to distinguish pneumonia from lung problems due to Covid 19).

The opportunity of using this technological solution in daily life can have an enormous impact on many aspects of neuromuscular diseases, from the early diagnosis to the evaluation of intervention outcome, up to the monitoring of rehabilitation treatments. However, the use of these sensors requires basic knowledge on multi-body dynamics, human biomechanical models, and sensor-to-segment calibration. 

In the first part of this tutorial, an in-depth analysis of the altered kinematic patterns of some neuro-muscular pathologies is presented. An experimental-based computational model of the motor control function of the brain is explained. Finally, some real examples of kinematic analysis conducted on pathological populations, such as Parkinson’s, Multiple Sclerosis, Cerebral Palsy, or Stroke, are presented. 

In the second part, attendees will be guided through the creation of a biomechanical model in MATLAB, fed with anonymized IMU data of patients with neuromuscular disease, up to the extraction of the kinematic quantities of interest for the analysis of intervention outcome.