Research project
The LCOMS laboratory at Université de Lorraine is seeking an outstanding Postdoctoral Researcher to work on contactless photoplethysmography (remote/imaging PPG, rPPG/iPPG) and its application to the estimation of physiological and cardiovascular parameters.
Photoplethysmography (PPG) is an optical technique that provides access to cardiovascular information through subtle variations in blood volume. In a contactless setting, these variations can be captured from ordinary video cameras, potentially enabling physiological measurements without electrodes, wearable sensors or dedicated medical equipment.
The project aims to investigate how video-based PPG signals can be processed and interpreted to extract physiological information beyond conventional heart-rate estimation. Particular attention will be paid to PPG waveform analysis, signal quality, robustness and artificial intelligence.
The research is situated within a long-standing research programme at LCOMS on contactless physiological measurements. Previous work has investigated heart rate and heart-rate variability, PPG waveform morphology, signal quality, blood pressure and arterial stiffness from camera-based measurements. Deep-learning approaches, including U-Net architectures and 3D convolutional neural networks, have also been developed for extracting and modelling physiological information from video.
Scientific objectives
Depending on the candidate's expertise and interests, the postdoctoral project may address one or several of the following questions:
- How can the quality and physiological content of contactless PPG signals be improved?
- Which features of the PPG waveform can be reliably recovered from video?
- Can contactless PPG provide access to cardiovascular parameters beyond heart rate?
- How can artificial intelligence improve the estimation of blood pressure or vascular properties from camera-based PPG?
- How can knowledge learned from contact PPG measurements be transferred to contactless measurements?
- Can new physiological biomarkers be inferred from subtle information contained in camera-based PPG signals?
A particular research direction will be defined jointly with the successful candidate, taking into account their background and scientific interests. The project may build on previous work showing similarities between contact and contactless PPG waveform characteristics and on deep-learning approaches designed to transform or directly interpret camera-based PPG signals. Augmentation and generation of synthetic iPPG signals can also be a relevant research direction. The successful candidate will therefore have the opportunity to contribute to the development of a broader scientific framework for camera-based, non-invasive and contactless physiological monitoring.
Tasks
The postdoctoral researcher will:
- Conduct a state-of-the-art review of contactless PPG and relevant physiological biomarkers.
- Design and implement signal-processing and/or deep-learning methods for extracting physiological information from video.
- Design experiments and evaluate the proposed methods using appropriate physiological reference measurements and/or existing datasets.
- Contribute to the design of new experimental protocols when required.
- Publish the research results in international peer-reviewed journals and conferences.
- Contribute to the preparation of future research projects and collaborations.
Candidate profile
We are looking for a highly motivated researcher with a PhD in one of the following fields:
- Signal processing;
- Computer vision;
- Artificial intelligence / machine learning;
- Biomedical engineering;
- or a closely related discipline.
Strong expertise in one or more of the following areas will be particularly appreciated:
- physiological signal processing;
- PPG / rPPG / iPPG;
- biomedical signal analysis;
- deep learning;
- time-series modelling;
- image and video processing;
- computer vision;
- experimental data acquisition and analysis.
Experience with Python and deep-learning frameworks such as PyTorch or TensorFlow would be advantageous.
A background in physiological measurements is welcome but is not mandatory. We particularly encourage applications from candidates with strong expertise in signal processing, computer vision or AI who are interested in applying their skills to biomedical problems.
Research environment
The successful candidate will join LCOMS (Laboratoire de Conception, Optimisation et Modélisation des Systèmes) at Université de Lorraine.
The project will be supervised by Frédéric Bousefsaf, Associate Professor at Université de Lorraine. His research focuses on biomedical signal and image processing, contactless physiological measurements, affective computing and artificial intelligence. His research programme on contactless physiological measurements has been developed at LCOMS since 2017 and builds on earlier doctoral research in the same field.
The research will benefit from an established scientific environment combining biomedical engineering, signal and image processing, computer vision and artificial intelligence, as well as existing collaborations and research activities in contactless physiological measurement.
What we offer
- A research position at Université de Lorraine, France.
- An interdisciplinary project at the intersection of AI, computer vision, signal processing and biomedical engineering.
- A research topic with strong potential for high-impact scientific publications.
- Access to an established research programme in contactless physiological measurement.
- The opportunity to develop an independent scientific contribution within a broader long-term research programme.
- A stimulating research environment in the LCOMS laboratory.
Application
Applicants should submit:
- a CV, including a list of publications;
- a short description of their research background;
- a research statement (maximum 2 pages) describing their experience and how it relates to the proposed project;
- contact details of one or two academic referees.
The candidate must satisfy the eligibility conditions of the Université de Lorraine AM2I Excellence Postdoctoral Programme 2027. In particular, candidates who obtained their PhD in Lorraine or whose PhD was supervised by a member of Université de Lorraine are not eligible for this programme.
The institutional application itself requires a CV, a scientific activity report and a scientific project describing its articulation with the host laboratory.
Informal enquiries:
Frédéric Bousefsaf
Université de Lorraine – LCOMS