Premature babies: new monitoring and diagnostic tools

Thermal cameras, sound sensors coupled with advanced data analysis techniques: the monitoring tools developed by the DigiNewB project aim to detect the very first signs of infection in premature babies. The expected benefit is twofold: to save precious time for the start of treatment (every half hour counts), and to reduce as much as possible the need for invasive medical procedures on premature babies (blood tests, for example).
Photo : CHU de Rennes

The device

Conceived and coordinated by Prof. Patrick Pladys and Prof. Guy Carrault from the University Hospital of Rennes, the Digi-NewB project, supported by the European Union to the tune of €4.4 million (Horizon 2020 programme), will comprise 4 closely connected axes:

    The creation of a database from the 'physiological signals' and health information of 700 children born prematurely
    The definition and design of relevant multivariate indices for the early detection of risk factors for the health, development and well-being of premature babies
    The design of a new "user friendly" monitoring tool capable of immediately producing relevant alerts for rapid management and better response to identified risks
    The evaluation, throughout the project, by the neonatology departments of the CHU of the Grand Ouest region of France of the indices and new tools, before their diffusion

Pr Patrick PLADYS is the head of the paediatrics department and the "woman-child" unit at the University Hospital of Rennes, and Pr Guy CARRAULT of the University of Rennes 1 (Signal and Image Processing Laboratory LTSI, UR1/Inserm U 1099) is the technical coordinator of the Clinical Investigation-Technological Innovation Centre.

A non-invasive study including 700 preterm babies

The 6 CHU of the Grand Ouest (HUGO network: Angers, Brest, Nantes, Poitiers, Rennes and Tours) will mobilise the actors of the HUGOPEREN network of paediatricians of the Grand Ouest, who are used to working together, to include 700 newborns in the study, to verify the hypotheses of the project on this large number of children and to test in real time the diagnostic aid tool and its ease of use

The clinical study will be based on anonymised data from young inpatients and their analysis.

This includes observed clinical data, signals collected during the follow-up of these children, and sounds linked to videos from infrared thermal cameras.

This non-invasive study will therefore not involve any additional medical procedures for the children concerned.

Project and partners

6 French university hospitals, 3 European universities and 2 French and Irish companies join forces in a European consortium

  • HUGO, the scientific cooperation grouping of the six university hospitals of the Grand Ouest region (Angers, Brest, Nantes, Poitiers, Rennes and Tours), is leading the project and will coordinate the European consortium;
  • the CHU of Rennes, which will be the promoter of the clinical study, will ensure, with the 6 University Hospital of the Grand Ouest, the collection of patient data within the neonatology departments;
  • the Université de Rennes 1 will be responsible for the creation of the database, its exploitation, its analysis and the design of the signal processing and video image analysis algorithms to create the decision-making system;
  • Tampere University of Technology in Finland will provide expertise in developing the infection risk prediction model;
  • The National University of Ireland in Galway will be tasked with identifying user needs, enabling the design of a user-friendly monitoring station;
  • The University of Porto, a specialist in signal processing, will propose non-linear indices extracted from the analysis of recorded physiological signals such as heart rate variability;
  • Voxygen Health, a Breton company specialising in speech processing, will contribute its know-how in the design of the data acquisition system and the computer processing of the babies' vocalisation;
  • Syncrophi, an Irish company, will produce the prototypes and prepare them for marketing.


At the end of 2015, the European Union confirmed the funding of Digi-NewB to the tune of €4.4M as part of the Horizon 2020 programme, the aim of which is to accelerate the transfer of innovation to care.

On 31 March 2016, the University Hospital of Rennes welcomed the 7 partners of the project and the 6 university hospitals of the Grand Ouest region for the official launch of Digi-NewB, the only French project selected in a call dedicated to personalized medicine.

The clinical services, university research teams and companies from France, Ireland, Portugal and Finland, which have joined forces to pool their expertise and bring this project to fruition, are all present for a first working seminar.

Long-term work

For many years, the teams of Prof. Pladys of the University Hospital of Rennes and Prof. Guy Carrault of the University of Rennes 1 have been working on the analysis of risk factors for the health of children born prematurely. They are behind several nationally funded clinical studies. The DigiNewB project will be able to rely on their work and on the collaborative tools already set up in the Grand Ouest region, including

    the HUGOPEREN network of paediatricians, which brings together 200 clinicians in over 50 hospital departments and paediatric units;
    Physiodev/Ascent Grand-Ouest, a network and research platform for the asynchronous collection of physiological data for signal processing studies of physiological rhythms during development;

This work has enabled proof of concept to be demonstrated, the relevance of the various signals collected to be demonstrated, links to the health status of children to be established and the feasibility of the project to be demonstrated.

A competitive environment

This large-scale project, supported by the HUGO network of university hospitals in the West of France, is one of the five proposals selected by the European Union within the framework of the Horizon 2020 programme (in a highly competitive context, as this call for projects received no less than 100 applications). The project set-up and start-up was supported by the Plateforme Projets Européens (2PE Bretagne).