The Health Data Hub (HDH) announced this week that the CNIL had authorized four new projects. Deep.Piste, DeepSarc, HUGO SHARE and Rexetris thus join the first three projects that had received CNIL approval at the end of October, CoviSAS, supported by the MIAI artificial intelligence chair at Grenoble-Alpes University and the Semeia company, Frog Covid, supported by the Clinityx design office specialized in data collection solutions and algorithms and the INSERM research unit Cardiovascular MArkers in Stressed COndiTions (MASCOT) and CoData (ICANS and Quantmetry).
Less than a year after its creation, the Health Data Hub reported that it has reached a major milestone with the authorization by the CNIL of four new projects. This authorization is essential since it conditions effective access to the data needed to conduct these projects. Using non-nominative health data, these projects aim to improve the quality of care in different disciplines (heart failure, transplants, cancer, etc.). For example, this will involve comparing treatment for rare cancers, particularly sarcoma, or developing a tool for automatic detection of tumors on mammograms. To do this, non-nominative data will be analysed in a secure manner thanks to the Health Data Hub’s technological platform, while respecting patients’ rights.
Deep.Piste: improving breast cancer screening thanks to artificial intelligence
Nearly 12,000 women die each year from breast cancer. Using artificial intelligence and Health Insurance data on care consumption, the Deep.Piste project will develop an automatic analysis of mammograms and refine the understanding of risk factors. This will make it possible to identify cases that could benefit from a lighter screening system (by reducing, for example, the frequency of mammograms to be performed for certain patients as part of the screening process) or by strengthening the screening process for others.
The Deep.Piste project is carried out by the company Epiconcept in collaboration with the regional cancer screening coordination centre (CRCDC) of Occitania.
DeepSarc: personalized medicine for sarcoma patients
Sarcomas are rare and multiform tumours affecting soft tissue, bone and cartilage. Nearly 4,000 people each year are diagnosed with them, with a poor prognosis for survival. 40 years of clinical research have led to the majority of treatment recommendations. However, due to the too small cohort size, linked to the rarity of these cancers, some issues could not be addressed by this approach.
The DeepSarc project, led by the Léon Bérard Center in Lyon in partnership with the Bergonié Institute, the Bordeaux Cancer Control Center and the Gustave Roussy Institute, offers a complementary approach by analyzing cohort data combined with data on healthcare consumption from the Assurance Maladie, in order to identify the most appropriate treatments for each patient profile, thus improving the chances of survival.
HUGO SHARE: limiting drug interactions and therapeutic disruptions in at-risk hospitalized patients
During hospitalizations, elderly patients with chronic diseases already treated in the city very often receive additional treatment. These patients may therefore be exposed to a significant risk of drug interactions or, on the contrary, suffer from the interruption of their treatment, which may lead to more or less serious health consequences. Every year in France, the misuse of medicines is responsible for 10,000 deaths, of which patients over 65 years of age are the most affected.
The Hugo Share project led by the University Hospitals of the Grand Ouest (HUGO) aims, by cross-referencing the databases of 6 hospitals with the data on the consumption of care in town from the Assurance Maladie, to better understand and prevent these interactions in order to protect the most fragile patients.
Rexetris, towards greater personalization of immunological treatments for kidney transplant patients
Information on the relationship between patients’ drug exposures and long-term effects is currently very incomplete, both in general and in organ transplantation. In the context of life-long treatment, knowledge of these relationships would make it possible to optimise therapeutic strategies, doses and formulas of these drugs.
Led by the Limoges University Hospital, with the support of Inserm and Optim’Care, the Rexetris project is studying the relationship between exposure to immunosuppressive drugs and the long-term outcome of the kidney transplant patient and the graft.
CoviSAS: a project serving a representative at-risk population
Patients with Obstructive Sleep Apnea Syndrome (OSA), due to repeated oxygen deprivation, often develop associated diseases that may make them vulnerable to VID19 (obesity, diabetes, high blood pressure, cardiovascular disease).
The CoviSAS project, led by the MIAI artificial intelligence chair at the University of Grenoble-Alpes and Semeia, a provider of software solutions using artificial intelligence, aims to find out the prevalence of severe forms of COVID-19 in these patients, and to identify the combinations of diseases associated with OSA (co-morbidities) leading to a higher rate of resuscitation stays or death. As Pierre Hornus, CEO of Semeia, says, “With the CoviSAS project, we are seeking to identify who, among people suffering from sleep apnea, is at risk of developing a severe form of COVID-19 in order to promote better prevention”.
In order to reconstruct care pathways, the project will use health insurance data, in particular data on hospitalizations and drug consumption. The results will thus contribute to improving knowledge of the epidemic and defining prevention and early care strategies for patients affected by these co-morbidities.
Frog Covid: the ambition of a broad identification of risk factors and a forecast of needs for people undergoing intensive care.
The Frog Covid study is also looking at recurrent associations of other diseases in patients with severe (hospitalization) or very severe (intensive care admission) forms of VIDOC-19. Through this study, the Clinityx data collection and algorithmic solutions research department and the INSERM Cardiovascular MArkers in Stressed COndiTions (MASCOT) research unit are seeking to identify factors predictive of the risk of developing severe to very severe forms of COVID-19, in order to define profiles of patients particularly at risk.
The project also aims to gain a better understanding of the paths taken by patients who have been in intensive care, according to their length of stay and their associated diseases. As Nicolas Glatt, CEO of Cliniyx, reminds us, the aim is therefore to “understand the medical history of multipathological patients who have been hospitalized or in intensive care, in order to better anticipate care needs”.
For each characterized patient profile, the project will make it possible to better predict the patient’s medical needs, his or her care after hospitalization, and the consequences of resuscitation on his or her quality of life (employment, social life, etc.). To do this, the project will rely both on non-nominative data from the Health Insurance and hospital stays of these patients and, for comparison, on follow-up data from patients treated by intensive care units for influenza or viral pneumopathy in 2017 and 2018.
CoData: Studying the impact of the pandemic on the management of breast cancer patients
Women with breast cancer had their management changed during the first confinement. The Institute of Cancerology Strasbourg Europe (ICANS) launched the CoData project aimed at using health data to analyze the impact of these changes and to develop tools to help institutions manage the return to normality. The statistical analyses will be partly carried out by Quantmetry.
The non-nominative data mobilized will provide a complete overview of the care pathways of these patients during the first wave of the epidemic: ICANS care data include diagnoses, information on surgeries and therapies; and Medicare data include drug prescriptions.
Translated from Zoom sur les quatre nouveaux projets du Health Data Hub
