[agents] postdoc and engineer positions at LITIS lab, Rouen, France

[Laurent Vercouter]

The LITIS lab at INSA Rouen Normandy, France is recruiting:

- 1 post-doctoral fellow for a period of 18 months
- 1 engineer for a period of 12 months

as part of the SICoPaD project (Interactive and Personalized Monitoring of Home Hospitalized Patients) funded by the Normandy region.

The 2 positions are to be filled from January 2019 (date may be postponed depending on the availability of the successful candidate).

Context of the project:

The multiplication and diversification of connected devices that can be deployed at a reasonable cost in domestic environments opens up many perspectives on the monitoring of daily activities for assistance or supervision. In a medical context, the emergence of these Internet of Things has generated a strong interest in real-time patient monitoring, patient care, collection and exchange of information between patients and carers.... The SICoPaD project is interested in the data analysis and processing provided by connected objects deployed in the patients home. In this project, we will focus more specifically on oral chemotherapy, a therapeutic modality that is becoming increasingly common in the management of cancer patients. The patient becomes responsible for ensuring proper compliance with his or her treatment and for managing adverse reactions on a daily basis. To ensure adequate follow-up of cancer patients treated with oral chemotherapy at home, health professionals must have sufficient information at their disposal, such as the correct intake of treatment, nutritional status, the existence of a fever, and the maintenance of social activities.

The objective of the project is to provide models for the analysis and exploitation of data coming from heterogeneous sources. The project tasks will benefit from the scientific skills of the MIND/LITIS and MAD/GREYC groups in the fields of behaviour detection, decision theory and human-agent interaction, in complementarity with the Henri Becquerel Centre providing its medical expertise and access to data from objects deployed in real conditions.

The objectives of the project are as follows:
analysing the temporal evolution from different sources (connected objects, questionnaires completed by the patient) taken in isolation to define a normal behaviour model and a deviant situation detection mechanism;
proposing a model for integrating these data into a patient profile that allows them to be linked in order to extend the effectiveness of detection by combining weak signals;
defining an adaptive automatic decision-making system for alert removal, initialized with decision rules provided by health professional experts, and which will evolve according to previously detected situations to decide earlier on the actions to be taken;
developing an interactive virtual agent that substitutes forms used as questionnaires so that the feedback from the patient is facilitated and contextualized according to his or her situation;
implementing these models on the platform provided by the Henri Becquerel Centre to evaluate, validate and quantify the effectiveness of the system in patient follow-up.

Missions and profile of the post-doctoral position:

The post-doctoral fellow will be recruited to contribute to a project workpackage entitled "Representation and detection of a patient's profile" whose objective is to precisely identify the data collected and to propose an integrative representation model characterizing the patient's activity and condition over a given period. Representation in the form of discrete dynamic systems will be preferred, possibly on different time scales if the heterogeneity of the values collected so requires. Comparative analysis by calculating distance with reference models will make it possible to identify abnormal situations, as can be done in the field of trust management for the detection of malicious actions/agents.

Candidates must have a PhD in computer science. Skills related to activity detection using discrete dynamic systems, automatic classification techniques and/or time series analysis are expected.

The contract is for a period of 18 months, starting in January 2019. The gross monthly salary is about 2800 euros.

Missions and profile of the engineer's position:

The engineer will be recruited to contribute to the workpackages "Patient-Agent Interactions" and "Application to the follow-up of cancer patients treated with oral chemotherapy". Its role will be to develop the interaction and data collection environment necessary for the project. His/Her task will be to develop a conversational agent to interact with the patient using existing platforms and to integrate the data collected by different connected devices within a unified platform.

Candidates must have a MSc degree in computer science. The expected skills are mainly computer programming and web technologies. A desirable experience in networking for connected objects or in programming conversational agents is a plus.

The contract is for a period of 12 months, starting in January 2019. The gross monthly salary is about 2700 euros.

Applications must include a CV and a covering letter and must be sent by e-mail to Laurent.Vercouter at insa-rouen.fr by 31 December 2018 at the latest.

—
*Laurent Vercouter*
Co-directeur du LITIS EA 4108
Coordinateur du Pôle Normand Stratégique de Recherche et de Formation Sciences du Numérique
Professeur des Universités en Informatique
Fédération CNRS NormaSTIC FR 3638
LITIS, INSA Rouen Normandie, France
Ph: (+33) 2 32 95 99 86

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cs.umbc.edu/pipermail/agents/attachments/20181212/7af4035f/attachment.html>