Sarah Zohar, Inserm research director, and Adrien Coulet, Inria research officer, both members of the HeKA project-team ("Discovery of biomedical knowledge guided by health models and data"), are taking over this space dedicated to the e-health of tomorrow. They explain their team's vocation and its multidisciplinary organization, which is emblematic of both Inria's ambition in e-health and the dynamics of PariSanté Campus.
How the HeKA team has been created?
Adrien Coulet: Inria and Inserm launched a call for expressions of interest in 2019 to create joint research teams. We then took this opportunity to consolidate our research project. Previously, when I was a lecturer at the University of Lorraine and a member of the Orpailleur project-team, I conducted work in biomedical data mining with Anita Burgun's team as part of the ANR PractikPharma project. This is how we started our collaboration. The shared desire of Inria and Inserm to support cutting-edge research at the interface of digital technology and medicine allows us to go even further with HeKA.
Sarah Zohar: Yes, HeKA was born of a great opportunity. I was previously deputy director of Anita Burgun's team at the Cordeliers Research Center (Inserm, University of Paris). Encouraged by various contacts at Inria, Inserm, and the Prairie Institute, we seized the momentum of the call for expressions of interest. For a year we worked remotely because of health restrictions, in working groups, to consolidate our project around three major research areas: patient representation, modeling of the patient care trajectory, and methods for future clinical trials. We turned in our project in May 2020 and it was accepted.
What is the focus of your research team?
S. Z.: We develop methods, models and tools that have a common goal: to improve health and patient care. What motivates our work is to answer questions such as: How can we help with diagnosis and prognosis? How to better take into account temporality in the multi-source and multi-scale modeling of health care data? How to give access to innovative molecules as early as possible? To do this, we are working on innovative designs and methods to enrich clinical trials with healthcare data and scientific knowledge.
Our research is very applied. Our specificity is to be a multidisciplinary team that brings together experts in medical informatics, biostatistics, applied mathematics, clinicians (including pediatricians) and pharmacists. Half of the permanent staff are hospital and university researchers associated with the AP-HP. Some of them are in charge of hospital data warehouses and are therefore very familiar with their content and use.
We are fortunate to be able to work with experts in rare diseases and pediatrics from the Necker Hospital and cancer experts from the Georges Pompidou Hospital. Thanks to these interactions we can produce models that are clinically interpretable. Because if the result is not clinically interpretable, it will not be beneficial to practitioners or to patients.
A. C.: We aim to produce tools for caregivers but also for patients. There is a big movement to get patients more involved. Patients are entitled to access their medical records. We can imagine projects where in addition to accessing their data, patients can have access to the decision options that doctors have for them, and in some cases participate in that decision making. This is called shared medical decision making.
S. Z.: Yes, we also aim for therapeutic education, with shared decision tools that can be used jointly by patients and their practitioners to provide them with relevant scientific information. To do this, we work with human and social science teams, including ECEVE (Clinical Epidemiology and Economic Evaluation applied to Vulnerable Populations) at Robert Debré Hospital.
What does your participation in PariSanté Campus mean to you?
S. Z.: PariSanté Campus is a great opportunity! It's a very exciting and challenging project that offers a unity of place, focused on digital health. It's probably the opportunity for a new type of research for us because we will be less scattered.
In particular, we will share half of the floor where we will be located with startups. We will be able to exchange easily. It will be easier for the team members to realize that their skills can be put to good use.
I am a great believer in the dynamics of location. I have initiated many collaborations at the coffee machine! We're going to come across a lot of expertise, entrepreneurs, patient associations... Little by little, common projects will emerge.
A. C.: Our university hospital colleagues will join us several days a week at PariSanté Campus. They are enthusiastic about the idea of having a place dedicated to their research activity and to the team because they are very involved, both in research and in supervising theses.
The site will offer us interesting interactions. PariSanté Campus will host companies. Many of us have relationships with industrialists, whether they are colleagues involved in startups or in the framework of CIFRE theses.
We will also be working alongside national health data management organizations such as the Health Data Hub. In addition to our projects that use hospital data, we have had and are submitting projects with national medico-administrative data (health insurance and others) that are accessible via the Health Data Hub.
What are your ambitions for the next 5 or 10 years?
A. C.: We use health data to develop methods and tools for clinical decision support. I hope that in 5 years I will have participated in the development of a tool that is interesting enough to be tested in real conditions in a clinical trial, and then used by doctors.
We are taking advantage of a rather unprecedented context for researchers. I measure our luck because our discipline is supported by a real institutional will. Our supervisors are convinced of the importance of our research and are more inclined to give us the means to recruit doctoral students and engineers.
S. Z.: At the team level, I hope to keep the same dynamic that PariSanté Campus has instilled. Since our creation, we have had different projects focusing on different diseases. We are able to respond together to calls for projects in a transversal manner on multiple indications and diseases by proposing approaches based on artificial intelligence, statistics and applied mathematics. We need to keep it up, to live up to our ambitions!
The lines are moving. Interdisciplinarity is very good for setting up projects. But on an individual level it is more complicated. It is sometimes complex to be in the right evaluation committee! I hope that in five years' time, there will be more Inria and Inserm researchers in these areas.