Smart-healthcare System with Federated Learning

When:
15/03/2020 – 16/03/2020 all-day
2020-03-15T01:00:00+01:00
2020-03-16T01:00:00+01:00

Annonce en lien avec l’Action/le Réseau : MACLEANDoctorants

Laboratoire/Entreprise : GEMTEX, ENSAIT, France et University of Kent, Kent, UK
Durée : 36 mois
Contact : quoc-thong.nguyen@ensait.fr
Date limite de publication : 2020-03-15

Contexte :
World report on aging and health from the World Health Organization (WHO) in 2015 shows that the problem of global population aging is becoming more serious. The proportion of the population aged over 60 years old will increase from 12 % in 2015 to 22 % in 2050. With a twice growing speed, the number of elderly people aged 60 and over will reach 2 billion during the next 35 years. Increasing demand and costs for healthcare is a challenge because of the high populations and the difficulty to cover all patients by the available doctors. In this case, one possible solution is the incorporation of both wearable computing and the Internet of Things (IoT) technology into health. After an operation, patients usually go through the rehabilitation process where they follow a strict routine. All the physiological signals, as well as behaviors of the patient, are possible to be monitored with the help of smart garments. The system can be tuned to the requirement of the individual patient. The patient’s health status and behavior can be observed remotely by doctors.

The candidate will be working (50%) in Human-Centered Design (HCD) team, GEMTEX research laboratory, ENSAIT, France.
The candidate will also work at School of Computing; University of Kent, UK (50%).
PhD Diplomas are issued by both Ecole Central de Lille and the University of Kent.

Salary net: ~1450EUR, plus 600-800EUR during the period in the UK.

Sujet :
The aim of this project is to propose an AI and cloud-enabled smart healthcare system. In order to collect the patient’s health status data, we can use “intelligent garment” instead of wearable sensors. In the intelligent garment, body sensors are integrated with the textile garment, which shall take various factors into consideration, such as sensor type, strategic location for sensor placement, the layout of flexible electricity cable, weak signal acquisition equipment, low-power wireless communications, and user comfortableness. The pulse sensor, body temperature sensor, electrocardiography (ECG) sensor, myocardial sensor, blood oxygen sensor, electroencephalographic (EEG) sensor and batteries are all connected with flexible wires. In order to facilitate the washing of smart clothing, the non-waterproof components can be all removed by taking off the buttons of clothing. Users can remove these components before washing and then reinstall them to the garment by snap on the buttons back. We propose to develop a smart healthcare platform, which is composed of three key components: 1) federated learning models that are trained using data stored at multiple different homes of the patients without the data ever shared with a hospital or a tech company’s servers, 2) one or several computing devices that serve as the “edge” servers locally, and 3) the intelligent garment that can communicate with the edge device(s). The PhD student will extend the state-of-the-art in the area of Federated Learning, Deep Learning applications for Smart Healthcare System. An important part of his/her work will be devoted to publishing and presenting in peer-reviewed journals and at relevant conferences.

Profil du candidat :
1. Applicants should hold a master’s degree or equivalent in Computer Science, Automation or a closely related area.

2. a solid background in Computer Science/engineering.

3. ability to work on interdisciplinary research projects.

4. completed language proficiency (equivalent to IELTS 6.5 or higher) requirements.

Formation et compétences requises :
1. good at machine learning and statistics.
2. programming skills in Python.
3. at least one international publication.

The interested candidates please send your CV, motivation letter via e-mail under reference SHSFL to quoc-thong.nguyen@ensait.fr by 15/03/2020

Adresse d’emploi :
ENSAIT (École nationale supérieure des arts et industries textiles), Roubaix, France

University of Kent, UK

Document attaché : Smart-Healthcare-System-with-Federated-Learning.pdf