Julie Josse, the missing link in statistics

At first glance, it might seem difficult to see a connection between the treatment of head injuries and statistical research. Since 2013, however, statistician Julie Josse has been working with intensive care physicians to improve the treatment of patients with multiple injuries.“From the moment firefighters arrive on the scene until the patient is admitted to the hospital, there is a significant loss of information and a risk of errors in decision-making. The goal of the TrauMatrixproject is to classify accident victims to better guide first responders,” explains the researcher.

Missing data

The statistical analysis of data collected by the thirty trauma centers involved in this project makes it possible to identify typical patient profiles based on medical information recorded by emergency responders (blood pressure, heart rate, etc.). Prior to the statistical analysis, Julie Josse also assists doctors with data collection:“For statistical work to be useful, you need to understand the field of application.” Her specialty—missing data in statistical analysis—is indispensable here, since information is often missing, either because it was overlooked in the heat of the moment or because the patient’s condition prevented certain tests from being performed.

Another key objective of this project is to evaluate the effectiveness of treatments for head injuries. With severe accidents, clinical studies are very difficult to conduct, primarily because obtaining patient consent is often impossible. Statistical analysis then serves as a means of evaluation, using the causal inference method.“This method allows us to determine, among different treatments, whether the observed differences are actually due to the treatments themselves and not to other variables. We must therefore disentangle the effects,” explains the scientist. She gives a telling example, that of hydroxychloroquine: “The causal inference method, for example, has shown that the positive effects observed in patients treated with hydroxychloroquine in the hospital were linked to the young age of the patients and not to the efficacy of the drug administered.”

Silicon Valley

Used in many fields, causal inference is all the rage. Julie Josse honed her skills as part of the Google Brain team in 2016: “A little bubble where researchers have every resource at their disposal to conduct fundamental research.”This isn’t the first time the young researcher has rubbed shoulders with the scientific elite. Before joining Inria in 2020, she spent 18 months at Stanford University between 2013 and 2016, then served as a professor at École Polytechnique from 2016 to 2020. Two very different experiences of scientific excellence:“the American friendliness of Silicon Valley versus the austerity of the French academic system.”

Julie Josse’s bachelor’s degree in applied mathematics from Brest did not necessarily set her on this prestigious path. As a mathematician, she actually preferred the precision of proofs to the uncertainties of statistics. A series of chance turns in her academic path eventually led her to a master’s degree in applied statistics and then a position as a statistical engineer at Agrocampus in 2007. There, her early work on the statistical analysis of sensory studies cemented her passion for the discipline. She embarked on a Ph.D. in the statistical management of missing data, which earned her the award for best thesis in 2010.

“The academic world of statistics is very open”

Her involvement in the statistical community around open access and knowledge sharing has opened doors for her.“The academic world of statistics is very open, as demonstrated, for example, by R Project, which allows researchers to make their statistical analysis programs available online for free,” explains the researcher, who was elected to the R Foundation following her contributions. As a doctoral student, she organized a gathering of this community in Rennes in 2009, under the skeptical gaze of some colleagues who couldn’t see how a doctoral student could bring together the global statistics research community.“Yet leading international statisticians came, particularly Trevo Hastie from Stanford University. That gathering earned me an invitation there!”

Her connections and the originality of her research on missing data don’t explain everything. Julie Josse’s career path is also impressive for her ability to consistently juggle two roles at once: engineer and doctoral student for three years, lecturer and researcher at Agrocampus with an additional part-time position in Silicon Valley for the next three years, and then professor at Polytechnique while also contributing to Google Brain. And during the lockdown related to the Covid pandemic, she teamed up with emergency room doctors to create an app that provides real-time information on hospital bed availability (the Icubam project).

With funding from Muse to build her team, Julie Josse is joining the Desbrest Institute of Epidemiology and Public Health (IDESP) to strengthen statistical research in health, particularly in the area of allergies. Her Inria-Inserm team, called PreMeDICaL (precision medicine by data integration and causal learning), aims to improve patient care by combining clinical expertise with the wealth of data from multiple sources.