IMT Atlantique, internationally recognized for the quality of its research, is a leading general engineering school under the authority of the Ministry of Industry and Digital Technologies, ranked in the 3 main international rankings (THE, SHANGAI, QS). With three campuses in Brest, Nantes and Rennes, IMT Atlantique aims to combine digital technology and energy to transform society and industry through training, research and innovation.
The postdoctorate will be located at Smart Macadam offices in Nantes (80%) and at IMT Atlantique Brest campus in the RAMBO team (20%). RAMBO is a member of Lab-STICC laboratory (mixed research unit of CNRS #6285). The postdoc will be supervised by the associate professors Mihai Andries and Christophe Lohr (IMT Atlantique) and Julie Golliot (research director at Smart Macadam). The research domains of the RAMBO team are Cognitive Robotics, Robot Learning, Ambient assisted living and Human-robot interaction. This project is part of the RAMBO research activities on ageing-at-home.
In France, more than one million people are affected by a neurodegenerative disease (2014) such as Alzheimer's disease, Parkinson's disease, Huntington's disease and multiple sclerosis. Early detection of the characteristic clinical signs of the onset of these diseases would enable the persons concerned to receive appropriate care and plan certain tasks while they are able to make important decisions concerning their personal ﬁnances, legal provisions or relating to their health. Early detection of Alzheimer's disease would allow to slow the progression of symptomatic manifestations (cognitive loss, memory impairment) while limiting the exposure of sufferers to risk factors.
Smart Macadam is a company created in Nantes in 2018, focused on the design and development of innovative solutions for people affected by neurodegenerative diseases. As part of its flagship project, named Mementop, Smart Macadam has developed a certiﬁed medical device, based on a personalized Gazette, a smartphone application that offers a set of simple functions (diary, memos, memory albums, games, spatio-temporal location...) and a support service. This device should allow sick users (referred hereafter as assisted person) to find their way in space and time, stimulate their memory and preserve their executive functions. The device is also intended for the relatives of the assisted persons (called carers) to whom it offers a unified channel of communication with their assisted person, as well as a structuring means to contribute to the preservation of the latter's autonomy at home.
We aim to increase the capacities of the Mementop device by extracting significant sounds from environmental noise which correspond to daily human activities, and identify them (noise from chairs, walking, opening cupboards, toilets, etc.). These noises, by their presence and regularity over the course of the day, are reassuring markers of a priori normal human activity. On the other hand, a prolonged absence or a drift in time may be the first signs or a suspicion of disorders in the user and may give rise to a procedure for alerting and removing doubts by a carer. This ambient sound classification tool will allow to identify certain alert situations, such as the falling or breaking of an object, or cries for help in case of injury.