IEM Seminar Series: Translating Raw Real-World Sensor Data into Clinically Meaningful Digital Health Measures
Abstract
The ability to monitor movement and activity in everyday environments using wearable sensors has opened new avenues for assessing health status in clinical and aging populations. However, extracting meaningful, validated clinical metrics from raw sensor data remains a complex, multidisciplinary challenge.
In this talk, I will discuss the differences between consumer-grade and research-grade wearables, and the technical and clinical hurdles in developing validated algorithms to analyze sensor data from individuals with severe motor impairments (e.g., Parkinson’s disease, multiple sclerosis, hip fracture, cerebral palsy).
I will further discuss how advanced signal processing techniques can extract rich, multidimensional features from real-world behavioral data, with a particular focus on the concept of ‘complexity’. Initially developed in the analysis of physiological signals, complexity metrics are increasingly recognized as comprehensive indicators of functional health. I will illustrate how these concepts can be extended to movement and activity patterns, and how reductions in complexity are observed in clinical conditions such as chronic pain, frailty, and fear of falling in older adults.
Finally, current limitations and clinical requirements as well as the potential of AI in this context will be discussed, highlighting future research directions at the intersection of movement science, clinical relevance, and data-driven precision health.
Bio
Anisoara Ionescu, PhD, research focuses on algorithm development and data-driven methods to extract clinically meaningful metrics from inertial and physiological signals, aimed at improving clinical assessment, particularly in neurological and motor-impaired populations. She has contributed to numerous interdisciplinary national and EU projects and played a key role in the validation of digital mobility outcomes in initiatives like Mobilise-D (EU, Innovative Medicine Initiative -IMI). She was a co-recipient of the 2015 Leenaards Prize for translational research.
The ability to monitor movement and activity in everyday environments using wearable sensors has opened new avenues for assessing health status in clinical and aging populations. However, extracting meaningful, validated clinical metrics from raw sensor data remains a complex, multidisciplinary challenge.
In this talk, I will discuss the differences between consumer-grade and research-grade wearables, and the technical and clinical hurdles in developing validated algorithms to analyze sensor data from individuals with severe motor impairments (e.g., Parkinson’s disease, multiple sclerosis, hip fracture, cerebral palsy).
I will further discuss how advanced signal processing techniques can extract rich, multidimensional features from real-world behavioral data, with a particular focus on the concept of ‘complexity’. Initially developed in the analysis of physiological signals, complexity metrics are increasingly recognized as comprehensive indicators of functional health. I will illustrate how these concepts can be extended to movement and activity patterns, and how reductions in complexity are observed in clinical conditions such as chronic pain, frailty, and fear of falling in older adults.
Finally, current limitations and clinical requirements as well as the potential of AI in this context will be discussed, highlighting future research directions at the intersection of movement science, clinical relevance, and data-driven precision health.
Bio
Anisoara Ionescu, PhD, research focuses on algorithm development and data-driven methods to extract clinically meaningful metrics from inertial and physiological signals, aimed at improving clinical assessment, particularly in neurological and motor-impaired populations. She has contributed to numerous interdisciplinary national and EU projects and played a key role in the validation of digital mobility outcomes in initiatives like Mobilise-D (EU, Innovative Medicine Initiative -IMI). She was a co-recipient of the 2015 Leenaards Prize for translational research.
Practical information
- General public
- Free
Contact
- Prof. Jean-Philippe Thiran, IEM Director