Epidemics are unpredictable, often deadly. Wherever they appear on the planet, we are all potentially concerned. With the globalization of air transport, a microbe can travel around the world in less than 24 hours.
My thesis focuses on building an automated infrastructure. Indeed, it focuses on a strong involvement of modern automation and artificial intelligence techniques for the management of medical data such as epidemic data, ensuring an efficient way to collect a large amount of information in a short time.
We managed to develop an Ansible Playbook responsible for the implementation of PostgreSQL databases, as well as a dedicated Playbook for the aggregation of data stored on several servers. In addition, we have addressed the security aspect of our infrastructure with the application of several techniques that ensure that database servers are sealed against various possible attacks.
As proof of concept, we decided to exploit the capacity of this infrastructure in data management, in particular the ability to collect and consolidate data from several servers. To then use the data collected to train predictive models. The data used to form the models are real medical data consisting of infant cry samples, of which there are two classes of samples: samples of healthy cases, and samples of cases with pathologies.
The models developed are based on the technique (Deep Feed Forward Neural Network), with the data we will collect using our infrastructure, we will be able to train the models to predict the classes of our samples, while aiming for the best possible results.
| Date | 9 Jul 2023 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Chakib Tadj (Supervisor) |
|---|
Sari, Y. (Author),
Tadj (Supervisor),
9 Jul 2023Student thesis: Master's thesis › Master in Engineering: Engineering