This deliverable is the initial version of the Data Management Plan (DMP) of the LifeChamps project [1], in accordance to the regulations of the Pilot action on Open Access to Research Data of the Horizon 2020 programme (H2020) [2]. It contains provisional information about the data that will be produced and collected within the project, whether and how it will be made accessible for re-use and further exploitation, and how it will be curated and preserved. 

Based on the guidelines on FAIR (Findable, Accessible, Interoperable and Reusable) Data Management in Horizon 2020 and the General Data Protection Regulation (EU) 2016/679 (GDPR) [3], all relevant datasets to be produced within the lifespan of the project were recognised and an initial detailed list was produced accordingly.  

The project at its present stage is foreseen to develop a series of datasets related to issues ranging from user requirements related to Quality of Life (QOL) monitoring, and sensor data captured from patients. Specifically, datasets are planned to be collected in two ways: (i) the development data collection and (iii) the data that will be collected during the deployment of LifeChamps platform. The former dataset will help the development and improvement of the algorithms and systems of the LifeChamps, while the latter will constitute the actual datasets that are foreseen to be the main input of the LifeChamps platform.  

Given that the majority of the LifeChamps datasets involve data collection from human participants, the respective data produced either raw or processed, should be carefully handled under thorough consideration of ethical and privacy issues involved in such datasets. For all the identified LifeChamps datasets, specific parts that can be made publicly available have been identified in this first version of the project’s DMP.  

It should be noted that this is an initial version of the DMP of LifeChamps provided at month M06 of the project. As a consequence, the datasets described at this stage, represent an early reflection on the data that we foresee to be collected. During the evolution of the project, we expect that there will be some changes either to the content of the datasets or the information classification. Whenever important changes to the project occur due to inclusion of new datasets, changes in consortium policies or external factors, further updates to the DMP will be created. Regular check points on the status of the data will ensure that DMP is implemented as foreseen. Still, main principles (as described within this deliverable) is expected to remain intact until the end of the project, thus forming the main strategic axes of the overall Data Management Plan. 

The final version of the DMP will be delivered at month M36 of the LifeChamps project (month M36) as part of the final project report (Deliverable D1.6 “LifeChamps Project Report”).

[1] Grant Agreement Number 875329 — LifeChamps
[2] European Commission, “Open Access guidelines for H2020 projects”, [Online] Available: https://ec.europa.eu/research/participants/docs/h2020-funding-guide/cross-cutting-issues/open-access-data-management/open-access_en.htm
[3] General Data Protection Regulation (EU GDPR) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data and repealing Directive 95/46/EC. [Online] Available:  https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32016R0679&from=EN#d1e40-1-1

This deliverable represents the Interim Progress Report for the project, also including the Societal Impact Report. This document will be submitted to the Commission to review and evaluate the progress of work in the project. The report will reference all deliverables for the reporting period and will be delivered within sixty days after the reporting period.

The LifeChamps project aims to disrupt techniques for Big Data modelling, analysis, and aggregation under a novel context-aware data-intensive and large-scale analytics framework towards delivering multi-dimensional Quality of Life (QOL) solutions for different cancer life champions. In the development of any ICT-based application, specially targeted for healthcare purposes, it is paramount to have active involvement of all relevant stakeholders. 

This co-creation approach is implemented throughout the overall project, but especially in the Work Package in which this document is framed, WP2 – Requirements Analysis, Pilot Use Cases and Functional Architecture. In this framework, this deliverable, D2.4 LIFECHAMPS platform and reference architecture – initial version, describes the initial LIFECHAMPS system architecture, and the technical specifications for the individual LIFECHAMPS components, modules, and tools. 

In order to answer LifeChamps project objectives, it is needed to define different levels of the architecture, from high to low level, and from logical to physical views. A “4 + 1 architecture view model” has been chosen to provide standards and guidelines to each technical partner, and also to support an integrated approach, ensuring that all the pieces match together. The “4+ 1 architecture model” is based on the use of five simultaneous visualisations. The diverse views correspond to the logical view for the functionality, the development view for the codebase, the process view describing the communication between services, and the physical view with the implementation of its services. Plus, a scenario view to illustrate the architecture. 

It is also fundamental to detail each component of the architecture. LifeChamps architecture is composed at this moment by a set of 27 components and actors including patients, caregivers, and clinicians. 

How these components provide concrete functionalities to end-users as a system is specified in the Logical View of the architecture, whereas how components communicate with each other is presented in the different diagrams of the Process View. The development and the physical views, as well as the scenario view will be detailed in the final version of the LifeChamps architecture by month 18 of the project. Overall, LifeChamps architecture should be considered as a live system, which evolves during the project’s development lifecycle.

The LifeChamps project aims to harness techniques for Big Data modelling, analysis, and aggregation under a novel context-aware data-intensive and large-scale analytics framework towards delivering multi-dimensional Quality of Life solutions for different cancer life champions. Hence, focusing on frailty in geriatric oncology, the LifeChamps concept is designed to answer the targeted population’s particular needs, concerning high quality and independent living, where integration with a personalised care pathway for long-term adjustment and health care management is essential. 

To meet end-users’ needs and relevant stakeholders’ expectations a co-creation approach was followed by conducting several studies within the context of the analysis, collection, and extraction of LifeChamps end-users’ requirements and established their views, preferences, and expectations from the developing LifeChamps System. These outcomes were the entry point for the design and implementation of the LifeChamps Platform and reference architecture.  

Comprehensive, the LifeChamps system is composed by its actors, the LifeChamps Platform, which includes the end-user applications, the computation, the analysis, and other required services; the Edge, encompassing everything deployed at patient’s homes; and the Data Warehouse, that acts as the secure storage and integrator of data from multiple heterogeneous sources. 

The present work describes not only the final LifeChamps system components but also its structure and reference architecture to support the system development and integration, from high to low level, and from Logical to Physical views through the 4+1 architectural model. This model consists of 4 views – Logical, Process, Development, and Physical – plus one extra view illustrating a small set of use cases or scenarios. This final version comprises the Development, the Physical, and the Scenarios views and complements the initial document D2.4. 

However, the LifeChamps system and architecture are not closed entities but ongoing activities. Consequently, the current work also includes a revision of the whole system, as well as the LifeChamps components. It incorporates the introduced updates and changes resulting from the work carried out since the initial version. This process might continue until the end of the project, supporting and incorporating its necessities. Therefore, this “final” version should be understood as the picture of the system and architecture at month 18 rather than a final and closed version.

The LifeChamps project targets recovering older (pre-frail and frail) cancer patients, but also caregivers and multidisciplinary health professionals with a comprehensive solution capable of offering tools and mechanisms to promote patients‘ empowerment and improved quality of life via timely and more accurate clinical decision support at the point of care.

In recent years, HPC platforms have contributed significantly to the development of digital health technologies [1]. With the advantage of faster, more powerful computing systems, computer science has taken a great leap forward over the past two decades, which has facilitated big data processing in the medical science field.

Recognition of rare and chronic diseases and health patterns may help patients receive faster and better medical services. For example, specific biomarker data from RNA/DNA analysis allows clinicians to predict many disorders early [2], before the conditions become acute. However, the large amount of data required to be processed requires equally scalable HPC platforms to be able to analyse it in a reasonable timeframe. These biomarkers and metabolic data analysis also provide valuable hints for oncological studies and predict many genetic disorders [3][4]. Moreover, it helps to assist doctors in evidence-based diagnosis, which saves time compared to manual data processing and lowers overall costs to medical facilities. HPC platforms can process the ensemble of data sources from clinical trials and patients’ medical history, which helps to verify the preliminary results [3][4].

Artificial Intelligence can be used to build extensive systems for most of the commonly used software packages in the areas mentioned above [5], which makes it able to drive innovation in the combination between High-Performance Computers and AI/ML, in addition to the integration with the current workloads and the required technology segments. LifeChamps is a structured solution based on a scalable predictive engine that attempts to determine which features affect the oncological patients’ quality of life the most, during and after the treatment.

In this context, the scope of this deliverable is to give the reader instructions on how to access and leverage the production-ready LifeChamps HPC platform, which has been described in detail in “D3.1-Container infrastructure and libraries for Big Data analytics on HPC”. The platform is fed with data from different sources as these are defined as input devices/methods, as part of the LifeChamps Edge platform. The deliverable is a demonstrator based on the pre-recorded demo “LifeChamps HPC Workshop” (https://youtu.be/0d9Qvbu8w-Y). It provides an overview of the current state of the HPC platform, including the user interface (UI), the custom HPC LifeChamps platform, and instructions (an appendix of a developer guide) on how a developer/data scientist working on the project would use the platform to develop data models.

This report concerns the Deliverable 3.5 “Health Data Intelligence Cloud Service” generated in the context of WP3 “Big Data Platform and HPC Infrastructure”, and more specifically, within the Task 3.5 “Secure Data Warehouse Management” of the LifeChamps project. It focuses on reporting the Secure Data Warehouse (DW) of the LifeChamps platform as well as on providing detailed information regarding its functionality as well as its interaction with other LifeChamps components.

As the name suggests, the DW component of the LifeChamps platform will implement a DW management distributed infrastructure with, relational database components for structured data like meta-data, and NoSQL data management components for raw, unstructured data. Furthermore, in conjunction with T3.3, data sharing, and protection methods were developed to allow the efficient exchange of data within the scope of LifeChamps. The output of DW will feed the LifeChamps dashboard being developed within Task 4.3- Scalable Analytics Engine, and the Task 5.4 – LifeChamps Web-Based Analytical Dashboard for Healthcare Professionals and enables the data storage from multiple heterogeneous sources as well as data retrieval algorithms for gathering relevant data for specific events, across multiple indexes in the Open Distro for Elasticsearch (ODFE).

Overall, this deliverable presents an in-depth description of the functionality of the DW component of LifeChamps. The topics that are discussed within the deliverable include:

a) the reference architecture of DW, b) the overall functionality of the DW, c) an elaboration of the main modules composing the DW component, d) the interactions of the DW component with other LifeChamps components (e.g., the Message Bus), e) the security mechanisms implemented by the DW, and f) the deployment and usage of the DW component along the visualization and reporting of the results to the end user.

Additional information about the integration of DW into the LifeChamps platform have been already reported within D6.2 “LifeChamps integrated platform – initial version” and will be further analysed in the Deliverables D6.3 “Platform integration and functional testing results” and D6.4 “LifeChamps integrated platform – final version”.

This deliverable presents in detail the functionality of the Edge Analytics Engine (ΕΑΕ) and the Cloud to Edge control plane of the LifeChamps platform. The ΕΑΕ is mainly responsible for the initial ingestion and preprocessing of data coming from different sensors. It is implemented as a stand-alone application running on the edge devices installed in patients’ homes, which in this case are Raspberry Pis. Through the various interfaces with the available sensors, the data are collected, processed and pushed to the cloud for further processing and storage.

The advanced computational potential of the edge devices allows us to perform analytical tasks on the edge, which would have to take place on the cloud a few years ago. The edge analytics engine can support both elementary processing, complex event processing and more advanced edge analytics.

The report describes the architecture used to implement the edge analytics engine and its relation to other components. It also describes the tasks that take place, including data ingestion, data preprocessing, and calculations of useful metrics and insights, and data transmission to the cloud. Finally, it includes information about the cloud to edge control plane, the orchestration and provisioning of the software running on the edge. It covers Tasks 4.4 and 4.5 in the DoA, while it is accompanied by full implementation of the described material.

This deliverable reports the results of Task 5.1 Quality of Life Monitoring and Modelling and Task 5.2 Frailty Domain Knowledge Models. The initial set of digital biomarkers to serve as clinical assessment proxies, as well as the developed frailty domain knowledge model based on experts’ knowledge will be presented.

Task 5.1 relied on the usage, initial analysis and modelling of a pre-existing dataset of AUTH, that included sensor monitoring of older adults in tandem with clinical assessments. Correlation between statistical features of sensor data and clinical assessment endpoints was attempted. The main outcome of the task was the identification of a preliminary valid list of mobility digital biomarkers, including activity levels, number of steps and walking speed, and daily functioning (TV watching patterns) for the assessment of several domains of QoL (Quality of Life), such as cognitive, mental and functional status of older adults.

Task 5.2 relied on the collaboration of health care professionals, to contribute with their expertise and experience in the clinical setting, with the technical team from AUTH and coordination of HULAFE. The output of this partnership is the Frailty Fuzzy Cognitive Map (FCM), which is one of the major preliminary outputs that will be included in the LifeChamps dashboard for clinicians.

Together, both tasks represent a relevant part of this project in which it has been analysed (both in the literature and in the clinical setting) how the dimensions of quality of life and frailty are identified and interact in older adult cancer patients.

The following sections of this deliverable will provide an identification of the most relevant approaches to identify and measure frailty and QoL in the literature, followed by the methodology applied in each of the tasks and the results obtained in each of them. Therefore, after the general introduction, objectives and timelines, all the work related to Task 5.1 will be presented first, and then all the development related to Task 5.2. The deliverable will conclude summarizing the results obtained in both tasks.

This deliverable, framed into the task T5.5 “LifeChamps AI-based Interactive End User Application for QOL Assessment”, provides an overview of the functionalities embedded into the LifeChamps mHealth application for QoL monitoring, as part of the whole mHealth solution that will be assessed during the LifeChamps PUCs. In addition, and according to the “Demonstration” nature of this report, it provides a link to the actual LifeChamps mobile application file for Android smartphones (APK file [Android Package Kit]) so it can be downloaded, installed, and tested by the reviewer.

This report presents theLifeChamps Platform Integration Plan and Development Environment setup. The first part of the document focuses on the LifeChamps Continuous Integration/Continuous Deployment (CI/CD) platform that will host all the development, testing and integration activities. It outlines all the technical documentation of all the software components comprising the CI/CD platform that adopt the Agile methodology and the DevOps culture to support both the development and testing activities. Moreover, we present the LifeChamps development & testing environment, consisting of both hardware and software resources that were setup to provide a secure environment for the development, testing and deployment of the project’s artifacts, completely separated from the pilot’s environment. The LifeChamps development & testing environment is based on Kubernetes platform. Additionally, we present the CI/CD workflows that were adopted by the development teams of the project to deploy, test, and integrate their software components. On the second part of the document, the integration plan that will guide the integration, testing and deployment activities is presented. We document in detail the integration points amongst the components of the LifeChamps platform. Following the outlined integration plan, WP6 with the forthcoming deliverables (D6.2 and D6.3) will deliver the LifeChamps Platform in two major releases, an initial one on M22 and a final one on M32.

This document presents the initial version of the LifeChamps integrated platform. Its objective is to unify the software components and services of the platform. The first part of the document focuses on the development and integration status of the components and services of the platform, their developed functionalities and the integration and testing scenarios that will be executed to achieve seamless integration among the LifeChamps components and services (section 3).

Then the second part outlines the LifeChamps Integration and Testing methodology that will be followed throughout the project (sections 0 and 5). All LifeChamps software components and services will be continuously integrated, deployed and tested in the development environment before being further deployed in the envisioned pilot cases. Moreover, we present the Continuous Integration/Continuous Deployment (CI/CD) workflows that will be adapted by the development teams to automatically deploy their applications in the development and production environments. The main goal is to create a first integrated prototype of the platform that will feed the WP7 pilot cases. The initial version of the platform will be tested by the end users and get their initial feedback.

Finally, a common testing template that will be used by all LifeChamps software components to describe their test cases is presented. A part of detailed testing scenarios is presented in section 8. A full list of overall testing activities and scenarios will be described and delivered in D6.3 Platform integration and functional testing results on M34. A revised integration plan is also presented.

This deliverable meets the Ethics requirement H – Requirement No. 1 as described in the Grant Agreement. It presents an overview of the procedures and criteria that will be used to identify and recruit research participants. It also confirms that recruitment procedures within the LifeChamps project comply with medical research ethics and GDPR.

This deliverable meets the Ethics requirement D9.3 PODP (Protection of Personal Data) – Requirements No. 4 as described in the Grant Agreement. It presents an overview of data security, data protection, ethics risks and data transfers. It also confirms that data handling within the LifeChamps project complies with national and EU legislation.  

When dealing with both project and personal data, Deliverable 9.3 should be read in conjunction with Deliverable 9.1 – Recruitment of research participants and with Deliverable 1.3 – Data management plan.

This deliverable examines the issue of incidental findings during research carried out under the LifeChamps project and how the partners will address it. The obligation to address the possibility of discovering incidental findings and describe in advance the procedure that will be followed is an ethical requirement and a formal obligation identified by the European Commission for all research involving human participants. To this end, this deliverable first identifies what constitutes an incidental finding in research, analyses potential risks involved, and then reports the mitigating measures to such risks in form of an incidental findings policy of the LifeChamps project. The report concludes that, as unlikely as it is to incur incidental findings during the LifeChamps project, partners are ready to implement the policy which will mitigate ethical risks that may arise.