How the Dedalus Big Data and AI Platform Underpins the PERSIST Cancer Survivor Programme

by José Luis Bravo, Rafael Pérez

How the Dedalus Big Data and AI Platform Underpins the PERSIST Cancer Survivor Programme

A consortium of 13 partners across Europe came together in January 2020 to develop an innovative approach to improve the quality of life for cancer survivors. PERSIST, which is funded by the European Commission’s Horizon 2020 research and innovation programme, is a support system for patients and clinicians designed to improve the coordination of care across the entire patient journey.

Top of mind with the project was to develop a truly patient-centric perspective by capturing their experiences to improve the current and future care of cancer survivors and to enable better coordination of care across the healthcare continuum.

Built using artificial intelligence and big data technologies, PERSIST provides clinical decision support based on anonymised health data from thousands of patients, while also using mHealth applications to collect wellbeing data from participating patients. Integral to both of these capabilities is a big data platform that integrates these two systems, enabling decisions to be made in real-time.

To make all this possible, each member of the consortium plays a key role – from the clinical partners bringing cancer survivors on board to share their data for this transformative project, to technology partners developing mobile applications and chatbots, to those developing clinical applications for decision support, to the overarching platform that connects all these interfaces.

That is where Dedalus and the DC4H platform – which is the next step in the evolution of Open Health Connect – is so crucial to the project. DC4H was built to enable users to integrate and ingest data across disparate applications, to link datasets, to gain evidence-based insights about populations, and to support clinical workflows across organisations.

PERSIST places significant emphasis on patient-centricity, which is a priority across all Dedalus solutions and projects. At a more clinical level, a bracelet worn by the patient tracks observations such as heart rate and blood pressure, as well as lifestyle data, for example exercise, and sends this data back to the platform.

However, PERSIST is equally focused on the patient’s subjective experience, including the emotional journey they have experienced as cancer survivors. Currently this data is being gathered from audio or video recordings on the patient’s mobile application and information is automatically sent to and stored on the data platform. Technology developers within the consortium are working on chatbot features that will also capture patient-experience data.

Safe and secure – and patient friendly

Where patient data is involved, safety and security, as well as accessibility, are of paramount importance. Dedalus has therefore built strong measures into the platform to prevent any unauthorised access to the data, and to ensure the patient has full control over his or her data, in line with General Data Protection Regulation requirements. The patient data is further controlled by limiting its access to doctors in the clinical study, who can only leverage the data of their assigned patients – in other words, patients from their hospital who have given their consent to that specific healthcare organisation.

Steps to better protect patient data include constant control management mechanisms in the platform, which allows the patient to provide consent for the clinical study to gather and process their data. If the patient decides to withdraw from the clinical study, they can choose whether to donate the data already captured or remove it, in which case all data from that patient is removed from the platform.

The project also ensures patients’ identities are protected. Each patient is given a device ID and a user ID that are not linked to the data inside the platform. No personal data is stored on the platform and the video testimonials are encrypted, with data from those recordings being automatically processed within the platform to gather subjective markers, such as mood detection from patients voices or facial expressions.

In addition to data being gathered from patients, the big data platform also stores a vast amount of retrospective data from patients outside the clinical study. To safeguard this data, the platform anonymises the data and adds an extra security layer to block the data extraction from the platform. Researchers need to be connected to a special VPN that allows them to access the platform and apply automated learning processes based on that data.

Another important consideration with the data has been developing security measures to ensure no patient information is lost. Patients who agree to participate could help to improve the experience for other cancer patients in future; however, sharing their experience can be extremely painful and emotional. It is vital, therefore, that patients are never asked to repeat any measures so they don’t have to keep reliving their cancer journey.

Every PERSIST partner is committed to supporting cancer survivors as they share their data and experiences. Dedalus has committed to ensuring that information is safely gathered and stored, according to GDPR requirements and in line with the wishes of the patient. Ensuring that patient trust will further help to drive engagement, which ultimately will improve the quality of life of all cancer survivors.

About José Luis Bravo, Healthcare IT Project Manager:

José Luis is Computer Engineer and certified as PMP© and ITIL©. He supports hospitals’ digital transformation and innovation on change in Healthcare Organizations, offering Dedalus experience and technologies to deliver patient-centric care.

About Rafael Pérez, Senior Architect Lead, DC4H:

Rafael is a Telecommunications Engineer specialized in Telematics. In Dedalus, he is currently working as a Lead architect for Care Plans in the scope of the DC4H platform and the Technical Lead for the EU funded Persist project.

Please open in latest version of Chrome, Firefox, Safari browser for best experience or update your browser.

Update Browser