From data science to improving remote NIV monitoring

Three questions for Dr Léo Grassion, pulmonologist at Bordeaux University Hospital

We are pleased to share with you an interview with Dr. Léo Grassion, member of the steering committee for the Renavo-tel registry, which is currently being set up under the auspices of the CNP de Pneumologie (French Federation of Pulmonology).
 Centralizing telemonitoring data and translating it into useful knowledge for practice: this is the challenge of this innovative project, which is open to all aspects of hospital and private pulmonology and will draw in particular on data from digital medical devices (DMDs) such as Vestalis.
 A concrete illustration of data science applied to improving the care of patients with chronic respiratory failure!

What is the purpose of creating the Renavo-tel register?

The aim of Renavo-tel is to centralize, for the first time at the national level, data on patients receiving non-invasive ventilation (NIV) or oxygen therapy (O2) who are being monitored remotely. The goal is to bring together two types of information that are currently not shared enough.
On the one hand, there is clinical data; and on the other hand, daily data from devices, which provide a high level of detail for monitoring. 
The registry is built by pulmonologists. When a patient is included, their clinical data is aggregated with telemonitoring data (progress, alerts, etc.). All of this is done with a view to gaining a better understanding of each patient’s specific needs and dynamics.

How can large volumes of data be transformed into useful indicators?

We will call on the expertise of data scientists. The aim is to train algorithms using this database with a view to applying them in everyday practice. The first step will be to understand which data is clinically relevant for the patient. Ultimately, the Holy Grail will be to understand how a patient’s condition is progressing and predict an event before it sends them to the emergency room.
Telemonitoring has changed the paradigm of follow-up care. Previously, patients would report symptoms to us retrospectively, and we would base our intervention on that information.
Now, telemonitoring provides subclinical data that we were not used to processing, and which could prove to be predictive of episodes.
This mass of data can be worrying, but it is precisely by aggregating it that relevant indicators can be identified. This is the whole point of this registry: extracting knowledge will enable tools to be developed and implemented in remote monitoring solutions, with a view to achieving better results.

What synergies do you anticipate with DMN publishers?

Collaboration with publishers is crucial in order to integrate data from DMNs into the registry, whether it concerns ventilator use, compliance, or leaks. Without DMNs, physicians do not have access to this information, which is essential for predicting treatment quality and compliance, among other things.
The next question is how the data will be used. Together with all the project partners, we hope that this registry will serve as a basis for numerous studies and analyses, leading to a better understanding of telemonitored patients. The central idea is that knowledge belongs to everyone. After the initial structuring stages (technical, scientific, etc.), which should be completed by the fall, we will specify the terms of collaboration with DMN publishers and other stakeholders, with the common goal of improving patient care.”

Philippe Salamitou, CEO of SRETT: “We are committed to facilitating the integration of data from Vestalis and its use within the Renavo-tel registry, because the contribution of data science and AI to improving remote monitoring care is an extremely promising area in which we want to maintain our involvement.”