European Commission logo
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS

Big Data Models and Intelligent tools for Quality of Life monitoring and participatory empowerment of head and neck cancer survivors

Periodic Reporting for period 2 - BD4QoL (Big Data Models and Intelligent tools for Quality of Life monitoring and participatory empowerment of head and neck cancer survivors)

Período documentado: 2021-07-01 hasta 2022-12-31

Head and Neck Cancer (HNC) imposes an extremely high socioeconomic burden on patients during and after cancer, including costs from treatment-induced morbidities, loss of workforce participation and short-term disability. These factors can have a profound negative influence on emotional well-being, interpersonal relationships, loss of work, stigmatization, and social integration, especially for women. Early recognition of clinical and potentially actionable significant change is difficult for expert health care providers, the more for patients. Person-centred care has proven to improve HNC survivors’ QoL and survival, contributing to lowering the socioeconomic burden of the disease on patients, healthcare systems and the society.
BD4QoL addresses survivors’ and physicians’ needs and overcomes the cultural, psychological organizational and technological barriers to systematic and coordinated monitoring of HNC survivors’ QoL. BD4QoL takes advantage of mobile technologies, e-coaching tools (chatbot) and AI algorithms that allow unobtrusive and continuous collection and monitoring of QoL determinants and the delineation of personalized QoL trajectories. We hypothesize that, by implementing the project’s tools in practice, we may easily add efficiency to current monitoring of treated HNC patients and successfully organize improvements of healthcare activities to timely intercept clinically relevant turning points that affects QoL, which are nowadays unknown or poorly appreciated.
The main outcomes are a set of tools integrated into a platform for HNC survivors' monitoring consisting of:
1. a mobile app that collects indicators of physical activity, sleep, social activities,
2. a chatbot for patients' support and empowerment in self-management of their QoL daily issues and treatment squelae, tailored and physicians' driven information
3. a web-based tool for point of care physicians allowing individual survivors' monitoring and early identification of QoL deterioration from the analysis of the indicators collected from patients, linked to a data management tool (REDCap)
4. a web app that allows the collection of individual QoL questionnaires
5. advanced models for personalized QoL prediction that will allow profiling of survivors and tailoring the follow-up strategies and post-treatment support.
The developed tools are being experienced in a randomized controlled trial (RCT) involving 420 HNC survivors in Italy and UK.
The following results have been achieved:
• The QoL indicators to be collected have been defined (physical activity, sleep, social activity, emotional status) in line with survivorship guidelines, semantically mapped into behaviors modifications and verified against the collection capabilities of the mobile phones embedded sensors and functions.
• The mobile apps for QoL monitoring that include an automatic monitoring of accrual and data collection for continuous verification of the system functioning and of users' compliance, the patient's support interactive e-coach Bidi chatbot and the QoL measurement tools have been developed, tested, released and are being used in the RCT
• The implemented BD4QoL data infrastructure is operational, checked and qualified for cybersecurity, privacy and disaster recovery. The data management system is established; the required security and privacy protection features are in place, relying on an established legal and ethical framework
• The BD4QoL data collection and PoC tool are operational an used in the RCT
• The prediction algorithms from historical studies datasets have been realized and results are ready for publication
• Data sharing agreements were signed by all Consortium partners, ethical approvals for the RCT were achieved by all recruiting hospitals
• The RCT enrolment started in 2021; more than 90 patients were recruited as of year-end 2022
• Users' support and maintenance of the BD4QoL platform was established, participants portal was published to support end users
• Dissemination started through the web, social media and specific communications and publications for the scientific community and the general public, the project was presented at congresses
• Collaboration is ongoing with two clustering initiatives (CS-AIW Cluster started by FAITH project and Horizon Results Booster led by CAPABLE project)
• Project results identification and market analysis was performed; IPRs are being defined by contributing partners
• Continuous project monitoring, risks management and results quality assurance is in place through plenary meetings in person and online biweekly, thematic meetings conducted almost on a weekly basis to address task-specific issues and take decisions. EU funding was correctly managed and distributed. Periodic reporting (on a 6 months basis) was collected to check the partners' involvement and results and to early manage deviations.
The main innovations expected from BD4QoL are:
• Continuous and unobtrusive post-treatment monitoring of QoL for HNC survivors, detection of related risks and enactment of timely self-empowerment intervention based on AI-powered chatbot
• Improvement in the personalization of QoL monitoring, by leveraging multiple conventional (e.g. electronic health records) and unconventional (e.g. mHealth) data sources, categorized in a holistic ontology
• Investigation of behavioral markers that allow the extraction of relevant features, longitudinally and unobtrusively, that could prospectively substitute more obtrusive methods like conventional questionnaires
• Inclusion of affective features, as detectable through AI-based technologies (e.g. IBM Watson)
• Investigation of AI and ML models that can predict, rather than just measure, QoL trajectories
• Improve physicians' support at the point of care, by integrating the above-mentioned innovations within consistent workflow apps, for improved decision making
These innovations will lead to the following impacts:
• Increase the latitude of data relating to the QoL trajectories of HCN survivors, and describe them in an appropriate comprehensive ontology that will facilitate their use, reuse and sharing, along FAIR data management principles
• Increase scientific knowledge on how behavioral and affective features, when added to conventional clinical data impact modelling of QoL trajectories, assess their predictive potential and ability to elucidate causative mechanisms
• Better HNC survivors’ follow up workflows, supported by chatbot self-empowerment support, alerting, data visualization and patient-physician communication
• Conduction of HTA, usability and acceptability studies to provide evidence-based support for better decision and policy making regarding cancer survivorship management
• Better QoL for HNC survivors, from the combination and application of the above innovations and impacts
• Increased knowledge on QoL determinants and socioeconomic impacts for different HNC survivors’ subgroups.
• Assessment of m-health technologies potential for health and QoL monitoring. of the limitations of current regulatory frameworks and of the technical barriers imposed by mobile devices providers.
The continuity and unobtrusiveness of HNC survivors monitoring, envisaged by the BD4QoL platform, assumes an increased interest in the aftermath of the Covid-19 pandemic which entailed a reduction of the face-to-face follow up visits and an additional requirement for remote monitoring of survivors.
BiDi chatbot for patients' empowerment and support
Project mission
BD4QoL solutions for Quality of life monitoring after head and neck cancer
BD4QoL secure cloud data infrastructure
BD4QoL main objectives
Mobile apps for QoL indicators collection
BD4QoL concept schema
BD4QoL functional architecture