Skip to main content
European Commission logo print header

HEalth related Activity Recognition system based on IoT – an interdisciplinary training program for young researchers

Periodic Reporting for period 2 - HEART (HEalth related Activity Recognition system based on IoT – an interdisciplinary training program for young researchers)

Reporting period: 2019-09-01 to 2021-08-31

The context for the Heart project is related to the increasing use of wearable sensors to remotely monitor human activities and vital parameters. The topic has become truly crucial in the management of the Covid-19 pandemic worldwide, where remote patients (treated at home or living in not easy accessible areas) needed monitoring services. The potential innovation of the Internet of Things (IoT) in personal health coaching programs is particularly useful to prevent chronic diseases that are booming in context characterized by aging societies (as in Europe and China). Indeed, those solutions rise a number of technical, legal, cultural and socio-economic challenges, which must be solved to enable a commercial solution.
The IoT can be a key growth business area for EU firms, which need competencies and skills for the exploitation of their innovation potential abroad. China is a key destination market, but products and solutions have to be adapted to the needs, behaviors and habits of Chinese consumer and to the specificity of the market. It is also to be considered that Chinese firms have achieved a high level of innovation in the use of digital technology to serve the healthcare industry, as shown in the management of the Pandemic. The impact of HEART can be found at industry and academic level, as well as in terms of quality of ESRs’ attitude and skills.
Heart aimed to:
• allow the formation of a team of new researchers, with a completely new profile, based on a strong interdisciplinary attitude, technical skills, creativity and entrepreneurial allure;
• release a health integrated activity recognition platform able to detect activities from heterogeneous data, using scalable algorithms, while safeguarding the privacy of the persons;
• provide EU firms with for new profiles of researchers having the needed skills in the IoT domain, thus increasing their international competitiveness.
The work carried has been in line with the Annex 1 to the Grant Agreement. The progress achieved in the training of ESRs and in the research activity is fully satisfactory. All deliverables were submitted on time and activities have been developed according to the initial plan, except for the mode of delivery of training and final conference which were held on line due to the restrictions of the Pandemic.
In WP1, all ESRs and supervisors have worked together to conceptualize the research activity and agreed on a common research approach. To this purpose they have sketched the state-of-the-art of the challenges the HEART project focuses on thanks to a literature review. They have identified the main research gaps and each ESR has defined the research hypothesis according to a genuine interdisciplinary approach.

In WP2, ESRs used existing data base as well as on the analysis of secondary level information, both in Europe and in China. ESRs worked together with an interdisciplinary approach, supported by specific and constant training. At the same time, they used the tools of analysis that were more appropriate for their specific discipline. Thanks to this double approach, they contributed to write the report on first data analysis and formulation of results on single ESRs projects, leading later to the release of the “alpha version” of the Health integrated activity recognition platform and subsequent prototype.

In WP3, all scientific activities were developed, including consultation with stakeholders and implementation of foresight scenario. ESRs have worked together to understand how major current trends are supposed to affect the healthcare industry as well as other major sectors where the use of personalized solutions driven by IoT is more likely to boom. Multiple mid-term scenarios were drafted and discussed with relevant stakeholders.

WP 4 relates to the development of interdisciplinarity skills. All the interdisciplinary workshops were conducted, thus contributing to the development of high interdisciplinary skills among ESRs.

WP 5 is related to all training activities, aiming to transfer ESRs those skills that are necessary to conduct the foreseen research activities, along with transferable skills. Career Development Plans were assessed periodically and their final versions were also released. The secondment program of ESRs have been duly completed and despite the pandemic all the ESRs spent at least 50% of their time in a not academic institution.
Participation to local training activities, in house training initiatives and external events was finalized according to what indicated in Annex I, as well as the participation to conferences by ESRs. All network-wide training activities foreseen in the project were implemented, even if, starting from February 2020, in remote modality.

WP 6 is related to dissemination, communication & public engagement, as well as to exploitation. All deliverables were submitted on time and activities carried out properly. Almost target values of KPI have been reached or even exceeded, moreover additional activities not originally foreseen in the Annex I have been conducted (such as additional videos of ESRs). All initiatives were conducted under the umbrella of the rules of IPR management procedures.

WP 7 was focused on management. All activities carried out under this WP (including management, coordination, reporting, monitoring, follow-up, evaluation, project bodies meetings) have been duly carried out, without any deviation and in a very satisfactory way.
The HEART goal was to train 6 ESRs to PhD level by means of collaborative research projects aiming to the development of a health integrated activity recognition prototype platform and its exploitation at European and Chinese level.
Key progress beyond the state of the art (SOTA) and expected results are described below.
Innovation on activity recognition algorithms and their scalability. In Heart project, a combination of active and transfer learning was approached, in order to reduce the amount of required labelled data. ESRs and supervisors worked towards the generalization of combined learning methods that are able to train models with a hierarchical structure (e.g. conditional random fields), which is typically present in complex activities (composed of a collection of primitive activities).
Innovation on privacy and security. In the domains of smart homes and healthcare, the life of people is improved using IoT devices that gather personal and environmental data, which is analyzed for offering new insights into users’ lifestyle.
The platform and specific solutions described in D2.4 can be considered an advancement of the SOTA as they are focused on systems with privacy embedded early on in the development lifecycle, coherently integrating health-specific privacy, security and legal requirements and enhancing the privacy by design paradigm.
Innovation on consumer needs and market penetration in China. Innovative remote monitoring systems must be conceived starting from a specific and precise understanding of customer needs, habits and acceptability of TRL both in Europe and in China. Heart has advanced the SOTA by defining appropriate Chinese consumer perspective and market analysis dedicated to the Chinese IoT and wearable sector.
Partnering and cooperation opportunities with Chinese firms were provided, also thanks to foresight scenarios for the implementation and market development of IoT solutions both in the healthcare and other related industries.