Open positions

Postdoc positions

Intracerebral recordings, microelectrodes, epilepsy, spike sorting

 

DURATION 2 years (possible renewal)
STARTING DATE 01/11/2017
BRIEF DESCRIPTION OF THE PROJECT 

Single unit and LFP analysis in human epileptic patients

The general goal of this project is to improve knowledge on pathophysiology of epilepsy, in particular the interictal events and the epileptic seizures. With the development of intracerebral recordings with microelectrodes in our epileptic patients, we are now able since several years to perform continuous intracerebral recordings at different scales, from usual macroelectrodes to single neuron recordings, in the interictal period but also during seizures, in multiple intracerebral areas.

The main role of the position will be to perform electrophysiological analysis of intracerebral signals at different periods: during the interictal period and preictal period and to study the link with multiunit/single unit activities in cortical areas.

Main activities The doc /post-doc will be integrated in a multidisciplinary team including researchers, engineers, clinical electrophysiologists, neuroradiologists and neurosurgeons. He/she will receive a full assistance from an engineer for the data acquisition and storage. In addition, he/she will have inputs from several researchers/methodologists in signal analysis.
Qualification & experience required The candidate should have a PhD or equivalent in Signal analysis or Neuroscience related field. Expertise in signal processing and good analytical skills (MATLAB) are required, notably in spike sorting analysis. Since the analytic approaches needed for this study have been extensively developed and used in animals, previous experience in animal in vivo or in vitro experiments is a plus.
Application Process Candidates should submit curriculum vitae, a cover letter and contact information for two referees. Application should be sent to Pr Vincent Navarro
 (vincent.navarro@aphp.fr).

Contact: Vincent Navarro

PhD positions

Undergrad intern positions

Mécanismes neuronaux néocorticaux de la mort et de la réanimation

La question abordée par l’étudiant(e) en Master, stagiaire dans notre équipe, sera double : quel est l’effet in vivo
d’une anoxie cérébrale prolongée, cause ultime de la mort, sur les propriétés électriques neuronales et les activités
synaptiques corticales ? et, comment ces mêmes propriétés sont-elles restaurées lors d’une réanimation cardiorespiratoire ?

Il s’agira donc d’identifier pour la première fois en « temps réel » et in vivo, dans un modèle
expérimental déjà existant dans notre équipe [1], les mécanismes neuronaux accompagnant l’arrêt des fonctions
cérébrales lors d’une anoxie complète et leur récupération lors d’une ré-oxygénation contrôlée. En particulier,
l’étudiant(e) tentera d’identifier, au cours des mêmes expériences, les mécanismes neuronaux et synaptiques soustendant : les activités corticales à hautes-fréquences survenant dans l’EEG au cours de la période anoxique précoce[2], la transition vers un état cérébral isoélectrique [électroencéphalogramme (EEG) « plat »], la survenue d’une onde cérébrale appelée « Wave-of-Death » [3] considérée comme un marqueur de mort neuronale et, les processus de récupération neuronale et synaptique (largement inconnus) après restauration de l’apport en oxygène.

Les expériences seront réalisées in vivo chez des rats anesthésiés, curarisés et ventilés artificiellement. Les
constantes physiologiques (Sp02, rythme cardiaque, expiration de C02) seront mesurées tout au long des
expériences. Les activités électrophysiologiques néocorticales (cortex somatosensoriel) seront recueillies par un
électrocorticogramme et par l’enregistrement intracellulaire simultané des neurones corticaux sous-jacents. Cette
approche permettra de mesurer à chaque instant les propriétés d’excitabilité des neurones, leurs patrons d’activités
synaptiques et la capacité du cortex à traiter des informations sensorielles (utilisées comme indice
neurophysiologique des fonctions corticales). Après une période contrôle, l’anoxie cérébrale sera induite par un
arrêt de la respiration artificielle. La réanimation consistera, après des durées d’anoxie variables, à rétablir la
respiration et donc l’oxygénation cérébrale.

Ce travail de recherche pourra être poursuivi par la réalisation d’un doctorat sur le même sujet dans notre
laboratoire.

Il est recommandé de consulter les références suivantes :
1. Altwegg-Boussac T et coll. Cortical neurons and networks are dormant but fully responsive during isoelectric
brain state. Brain, 140: 2381–2398, 2017.
2. Borjigin J et al. Surge of neurophysiological coherence and connectivity in the dying brain. Proc Natl Acad Sci.
110:14432–14437, 2013.
3. van Rijn CM et al. Decapitation in rats: latency to unconsciousness and the ‘wave of death’. PLoS One. 6:e16514,
2011.

Analysis of Pupil Fluctuations from Video Monitoring Using Deep Learning

 

Host laboratory: The “Network dynamics and cellular excitability” Team[1], in collaboration with the Movement Investigation and Therapeutics Team, of the Brain and Spine Institute (INSERM U1127, CNRS-UMR-7225) at the Hôpital de la Pitié-Salpêtrière in Paris proposes a M2/engineering internship in the field of physiological monitoring using deep learning of video recordings.

 

Introduction: Spontaneous oscillations of pupil size are one-dimensional time series that provide a clear proxy of neural dynamics in the brain autonomous system [1-2]. At rest, pupil size is modulated by the activity fluctuations of sympathetic and parasympathetic projections [1-2]. The use of video for non-contact and low-cost methods for measurement physiological parameters have a significant potential in remote sensing or tele-monitoring. Thus, pupil detection, characterization and tracking from video monitoring provides a non-invasive, quantitative and sensitive biomarker that can reflect the activity of brain circuits that are altered during different cognitive states or brain disorders [2-3].

 

Current solutions for video-based analysis of pupil remain challenging for images with artifacts and under   naturalistic non-controlled light conditions. The whole pupil is not always clearly visible to the camera, and its appearance may suffer from reflection of external light sources on the cornea, from low and changing illumination, or simply from low contrast, camera defocusing or motion blurs [1]. In medical image analysis and computer vision, deep learning algorithms have become a powerful method for robustly extracting complex features from medical videos [4-5]. Deep learning algorithms, in particular convolutional neural networks (CNN), can achieve a robust pupil segmentation and tracking and automatically compensate for several image/video artifacts [6].

 

Objective of the stage: During the internship, the student will evaluate the accuracy and relevance of deep learning algorithms to detect, characterize and track pupil-size from video recordings of benchmark databases (e.g. the Labelled Pupil in the Wild [8]). The student will then assess the robustness of the video-based monitoring of pupil size/fluctuations to different image artifacts and lighting conditions in different real video settings (with mono-focal standard and infrared cameras) at our laboratories. The pupil segmentation and tracking algorithm, including automatic procedures to remove or correct image artifacts, will be implemented as open-source code (in Python or C++) for free usage in academia.

 

Specific requirements:

  • Minimal duration: 4-6 months
  • Academic background in image processing or bio-engineering. Basic knowledge of machine learning (classification, neural networks) and image processing tools (filtering, segmentation, etc.) are required,
  • Good programming skills (C++ or Python) are required for this position,
  • A minimal level of scientific English is preferable,
  • The candidate should have excellent interpersonal skills and the ability to work independently.

 

The successful candidate will have the opportunity to join a dynamic and scientifically grounded research team. His/her tasks will be directly supervised by M Chavez[2] (CNRS) and P. Pouget[3] (CNRS).

 

Applicants should send their CV and covering letter to the 2 following addresses:

mario.chavez@upmc.fr

pierre.pouget@upmc.fr

 

Bibliographic references:

[1] AR. Mitz et al. (2017). Using pupil size and heart rate to infer affective states during behavioral neuropsychology experiments. ournal of Neuroscience Methods279: 1-12.

[2] J. Stoll (2013). Pupil responses allow communication in locked-in syndrome patients. Current Biology, 23(15), R647-R648.

[3] P. Artoni, et al. (2019). Deep learning of spontaneous arousal fluctuations detects early cholinergic defects across neurodevelopmental mouse models and patients. Proceedings of the National Academy of Sciences, 201820847.

[4] Y. LeCun, et al. (2015). Deep Learning. Nature 521, pages436–444

[5] G. Litjens (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60-88.

[6] Y. H. Yiu (2019). DeepVOG: Open-source pupil segmentation and gaze estimation in neuroscience using deep learning. Journal of Neuroscience Methods, 324, 108307.

[7] M. Weygandt Mathis & A. Mathis (2020). Deep learning tools for the measurement of animal behavior in neuroscience. Current Opinion in Neurobiology, 60, 1-11.

[8] TM. Tonsen (2016). Labelled pupils in the wild: a dataset for studying pupil detection in unconstrained environments. In Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications (pp. 139-142).

Remote Physiological Monitoring Using Smartphone Videos

Host laboratory: The “Network dynamics and cellular excitability” Team[1], of the Brain and Spine Institute (CNRS-UMR-7225), at the Hôpital de la Pitié-Salpêtrière in Paris, proposes an M2/engineering internship in the field of physiological monitoring using smartphone video recordings.

 

Introduction: Non-contact and low-cost methods for measurement physiological parameters (e.g. heart and respiratory rates) have a significant potential in remote sensing or tele-monitoring. Current solutions for non-contact measurements of heart and respiratory activity include laser Doppler, microwave Doppler radar [1], thermal imaging [2]. Despite their relative success, these proposals require an expensive and specialized hardware. There is a real need for accurate, low-cost and easy to use systems that can be used in tele-monitoring.

 

Recent advances in video and computer vision technologies have allowed video camera systems to become a useful alternative for non-contact monitoring of heart and respiratory rates [3]. Color changes in optical video measurement of the skin have been shown to contain information related to cardiac and respiratory activities [3-4]. Physiological measurements can be thus obtained by statistical analysis of video images. Nevertheless, these approaches generally require multichannel signal inputs (the RGB channels) and normal ambient light conditions.

 

Some smartphones have the capability to process video streams from both the front- and rear-facing cameras simultaneously. Nevertheless, the potential of smartphones for extracting and monitoring vital physiological parameters has not been explored [5]. Motivated by the video processing capabilities of smartphones and their extensive usage, they will be used for the monitoring of heart and respiratory rate.

 

Objective of the stage: The monitoring of cardio-respiratory dynamics from distant video recordings has already been studied in our group. A python-based toolbox has been recently developed for standard video recordings (webcams). In this internship, the student will implement this monitoring system on an Android smartphone. On the basis of our previous work, she/he will evaluate the accuracy and relevance of non-contact physiological measurements (including heart and respiratory rate) from single channel images (one of the RGB channels, gray levels, or brightness) obtained by smartphone videos (under Android OS). She/he will also assess the robustness of this video-based monitoring system to body/head movements/position, subject-camera distance and ambient illumination level. The monitoring system must be implemented using the open-source toolbox OpenCV.

 

Specific requirements:

  • Minimal duration: 4-6 months
  • Knowledge of software development on Android (Android Studio, and Android frameworks) is required,
  • Good programming skills (C++ or Python) are required for this position,
  • Background in informatics, signal/image processing or bioengineering.
  • Basic knowledge of image processing tools (filtering, segmentation, etc.) and OpenCV library
  • A minimal level of scientific English is preferable,
  • Candidate should have excellent interpersonal skills and the ability to work independently.

 

The successful candidate will have the opportunity to join a dynamic and scientifically grounded research team. His/her tasks will be directly supervised by M Chavez[2] (CNRS).

 

Applicants should send their CV and covering letter to: mario.chavez@upmc.fr

 

Bibliographic references:

[1] SS. Ulyanov; VV. Tuchin (1993). Pulse-wave monitoring by means of focused laser beams scattered by skin surface and membranes. In Proc. SPIE “Static and Dynamic Light Scattering in Medicine and Biology”, 1884: 160-167.

[2] M Garbey, N Sun, A Merla, I Pavlidis (2007). Contact-free measurement of cardiac pulse based on the analysis of thermal imagery. IEEE Trans Biomed Eng, 54 (8):1418-1426

[3] W. Verkruysse, LO Svaasand, and J S. Nelson (2008). Remote plethysmographic imaging using ambient light. Opt Express. 16(26): 21434–21445.

[4] Poh MZ, McDuff DJ, Picard RW (2010). Non-contact, automated cardiac pulse measurements using video imaging and blind source separation. Opt Express. 18(10):10762-10774

[5] Nam Y, Kong Y, Reyes B, Reljin N, & Chon KH (2016). Monitoring of heart and breathing rates using dual cameras on a smartphone. PloS one, 11(3).

 

We permanently seek talented undergraduate or graduate students from different backgrounds interested in our scientific research areas. Such disciplines include electrophysiology, clinical neurosciences, or biomedical signals processing.

 

Inquiries or applications should include a cover letter and updated CV. Specific job opportunities are regularly posted in our team website.

 

Here are some useful links to international organisms offering scholar/fellowships:

French Ministry for Europe & Foreign Affairs

Campus France

Eiffel Scholarship Program

Human Frontier Science Program

McDonnell Foundation