Skip to main content

Personalised Body Sensor Networks with Built-In Intelligence for Real-Time Risk Assessment and Coaching of Ageing workers, in all types of working and living environments

Periodic Reporting for period 1 - BIONIC (Personalised Body Sensor Networks with Built-In Intelligence for Real-Time Risk Assessment and Coaching of Ageing workers, in all types of working and living environments)

Reporting period: 2019-01-01 to 2020-06-30

Only half of European workers aged 55-64 are still in employment today - with strong gender and regional differences. To fight the consequences of demographic change, which puts considerable strain on social security systems, many EU countries have already raised the official retirement age and have often also restricted possibilities for early retirement.

This however can only be achieved if the work is made more sustainable for an ageing workforce, in terms of good working conditions, physical and mental well-being, and work–life balance.

BIONIC is a European research project aiming at developing a mobile, unobtrusive and privacy preserving platform for real-time ergonomic risk alerting and coaching, enabling the design of workplace interventions adapted to the needs and fitness levels of specific ageing workforce. Optimal engagement of the workers is supported by motivating, gamified education elements (Gamification) for prevention and self-management of musculoskeletal health in any working/living environment.

The general vision is that the BIONIC system can be integrated unobtrusively into different types of cloth and can be used by a worker throughout the working day.

The application aim is two-fold

1) to give health related feedback to the user
2) to obtain evidence-based data to medically sound workplace recommendation in any environment.

The BIONIC system will be adjustable to the fitness and age of the worker and should perform ergonomic monitoring and feedback to the user, e.g. for prevention exercises at home. The recorded data should be completely transparent for the user and should be useful as information source for strain estimation, e.g. by medical stuff.

The main objectives of the BIONIC project are:

Monitoring and prevention of work-related risk factors for the various types of musculoskeletal diseases (MSD) and related disorders.

Continuous medically sound persuasive and unobtrusive coaching, e.g. for men and women with chronic musculoskeletal diseases

The scientific and technical objectives to reach these goals are:

To create an innovative, holistic and unobtrusive body sensor network (BSN) that can be freely and loosely integrated into sportswear, workwear and Personal Protective Equipment

Allowing for completely mobile real-time and intelligent sensor data fusion and analysis

GDPR compliant user data management, analysis, presentation and protection

Kinematic and biomechanical models for age adapted ergonomic risk assessment and workplace analysis

Gamified user coaching and user interaction

Validation of the system in a real setting, i.e. at workplaces, home, outdoor activities
Work performed from the beginning of the work to the end of the period covered by the report and main results achieved so far:

The project can be roughly split into three different stages:

Year 1: System design phase

Year 2: System development and prototype implementation

Year 3: Pilots phase and system validation

In June 2020 half of the project was over, we are in the middle of the system development and prototype implementation phase. During the first year in the “system design phase” the main results can be summarized as follows:

With a user driven design approach, including questionnaires, and discussion with experts the user requirements, usage scenarios and medical requirements and trust concepts for medical wearables of the BIONIC framework have been defined. In parallel the state-of-the-art sensor hardware was reviewed, and a set of sensors was discussed and selected in with the consortium. Based on this information and feedback from the end-users a biomechanical model and ergonomic tools for this model (OWAS, ISO11226) were selected. Additionally, a procedure and a mobile application (app) for generating a 3D morphometry of the user was implemented. In parallel to these tasks the platform concept, including data types and categories and all the main components of the system have been defined. Once the user and medical requirements were defined the work on the gamification strategies for optimal user engagement started and was proposed at the end of year one. All previously mentioned achievements were analyzed and aligned with a Privacy and Data Protection Framework.

During year two, the “system development and prototype implementation” phase, different versions of prototypes were implemented to eliminate dependencies between components to obtain an early feedback from the partners. A heterogeneous, time synchronized network application layer software stack has been implemented and tested for the body sensor network. Possible hardware platforms and concepts for on body processing have been evaluated and a prototype running full body kinematics at 100Hz including carried load estimation from pressure insoles have been developed on an embedded application processor. Apart from OWAS and ISO 11226 an age adapted ergonomic risk model has been developed and the work on implementing it on the embedded application processor just started. The data platform including authorization and storage services has been implemented and tested. Work started in the pilot phase planning, which starts in the year three of the project. A detailed report on Data Protection Measures is being worked on right now that’s giving extensive analysis of the privacy and security measures for the system.
The BIONIC project focuses on technological and data protection challenges to obtain a mobile system that targets widespread usage of (long term) in field mobile ergonomic assessment with objective and reliable and trustable data management and measures as well as interactive feedback.

Expected advances beyond the state-of-the-art are:

Magnetometer-free and self-calibrating inertial body motion tracking, i.e. the sensor to body-segment mounting of inertial measurement units can be automatically estimated alongside the human pose tracking, thus allowing for minimal setup time and neglecting error prone calibration phases for inertial body motion capture, and without using magnetometers, i.e. applicable in any industrial surrounding.

Contributions to kinematic estimation from loosely coupled sensors, i.e. the body sensor network can be fully integrated into comfortable clothes and do not have to be tightly fixed on human bones and the obtain kinematic motion estimation generalizes to a wide variety of motions with calibrated kinematic uncertainty estimates.

Feasibility of long-term (one-day) completely mobile, online ergonomic assessment, based on human motion capture.

Modular multi-sensor network that can be easily synchronized and integrated into cloth.

Mobile consistent kinematic and load estimation from low-cost pressure in-soles and kinematic motion tracking

New approach to trustable monitoring system that is GDPR compliant

The expected impact of BIONIC target to ease the long-term applicability of ergonomic monitoring and assessment in-field for long-term mobile measures that support evidence based medical support and prevention educational with trustable data management. To ensure availability of the developed technology the scientific progress is frequently reported on international conferences, journals and fairs and the technology is will be available via SMEs.
Figure / Slide - BIONIC data flow
Figure / Slide - BIONIC main objectives
Figure / Slide - User owns his data
Figure / Slide - GDPR Compliance
“BIONIC,” an intelligent sensor network designed to reduce the physical demands at the workplace
Figure - Three phases of the BIONIC project
BIONIC – Loosely Coupled Senor Network integrated in the work wear
Figure / Slide - Secure wireless NFMI technology
Figure / Slide - Unobtrusive and loosely coupled Sensor Networks