Blog

Functions-as-a-Service to improve the Healthentia eHealth Platform

Innovation Sprint is a Belgian SME that offers technology solutions to the eHealth and Life Sciences business sectors, focusing on Big Data Analytics & Artificial Intelligence on Real-World Data (RWD). Innovation Sprint has a multidisciplinary team with expertise in IT solutions, data science, eHealth, and clinical research, with more than 20 years of experience in healthcare and a strong research background.

Innovation Sprint’s main product – Healthentia – is a CE-marked Class I Medical Device that captures clinical outcomes from mobile, medical and IoT devices, using a patient-centric mobile application and offering AI-driven smart services, such as biomarker discovery, patient phenotyping, in silico trials, and virtual coaching. Healthentia is being used in many of the company’s research projects, but also in numerous studies with hospitals and pharmaceutical companies.


In PHYSICS, Innovation Sprint is providing the eHealth Use Case, through its Healthentia platform, . How can we make sure that the increasing number of AI-based smart services that are offered through the platform scale with the growing number of users across different studies? How can we remain agile in our development and deployment processes? How can we ensure that all data processing operations conform with strict security, GDPR and medical guidelines? These are some of the challenges that Innovation Sprint is facing, as an innovative eHealth startup in its growth phase.

From a bird’s eye view, the eHealth Use Case in PHYSICS can be described using the image below. Users of the Healthentia platform will be collecting Real-World Data (e.g. activity tracking data, self-reported events and questionnaire results). The data will be used by three ML components: Digital Composite Biomarker discovery, involves training a machine learning model to predict relevant health outcomes. Explainable AI will derive the most important aspects leading to the predictions. Finally, study participants will be clustered based on a Digital Phenotyping approach. The outcomes of all the ML components are used to power various Smart Services, such as Risk Stratification, Virtual Coaching and In Silico Trials.


It is clear that the eHealth Use Case is handling senstive data of end-users. In the previous blog post, we have read about security, the inherent benefits of going serverless, and the remaining security considerations that will need to be explored during the execution of the PHYSICS Project. Issues related to privacy, handling of personal data, and other (medical) regulations will need to be explored in a similar way.


Follow ISPRINT on Twitter and on LinkedIn to see the latest updates!




You might be interested in …