VALENCIENNES GENERAL HOSPITAL (VGH) is regularly quoted as one of the three most important general hospitals of France. VGH has a capacity of 1,850 beds, close to that of CHUV in Lausanne or HUG in Geneva. With capabilities in surgery, medicine, and obstetrics, this innovative facility serves as a reference hospital within the Hainaut-Cambrésis health territory.
VGH: THE CHALLENGE
Time series data to predict the future using machine learning
After implementing the CALYPS Saniia dashboard to optimize the organization and the planning of ambulatory operations in a Swiss hospital, our team realized that it could do more than just represent the past. We had data available in the form of time series, well suited to the use of Machine Learning (ML).
Emergencies are disrupting care planning and create uncertainty
In a discussion with a consortium of Parisian hospitals, we understood that uncertainty was at its height when arrivals from the emergencies had to be integrated into a planned flow. As emergencies generate around 30% of hospitalizations over a year, they will somehow disrupt the organization and care planning.
OUR SOLUTION: DATA SCIENCE
Health data preparation and cleaning for feature engineering
The implementation of an artificial intelligence solution based on Data Science has been done progressively at the CHUV, starting with the integration of health data. First: data preparation! At each iteration, objective was to evaluate the variables that impact predictions and take them into account in our predictive model. Only then did we integrate data from outside the hospital such as weather, sports venues, cultural events, holidays, or vacation schedules.
Evangelizing data science predictive outputs is key to foster adoption
We proceeded hand in hand with the medical staff who had to interpret and act on these predictions. Internal collaboration, evangelization, and training are three key elements of adherence to the system. It helps to build trust in the predictions and allaying any residual fear.
VALENCIENNES GENERAL HOSPITAL BENEFITS
During summer 2019, CALYPS started a real time experiment at Valenciennes General Hospital and, six weeks later, provided daily predictions of emergency arrivals with an accuracy of >90% over the period. See how CALYPS Saniia works in this news flash broadcasted on national TV channel France 2 (3 min.):
CALYPS Saniia revolutionises the way a service such as ours is approached and enables us to anticipate tension within the hospital. The results are impressive: they help my teams to better respond to the challenges posed by the constant increase in admissions.
CALYPS Saniia is part of an overall plan to find new organisational and budgetary room for manoeuvre using iA and data analysis from our HIS. This is the first outcome of our digital transformation policy with our Health Data Hub. Integrating CALYPS Saniia into our establishment was quick and easy.
Thanks to the know-how and professionalism of the Calyps team, the platform has enabled the identification of tangible ROI. Given the promising results, we wish to make CALYPS Saniia available to all major departments of our hospital.