A5066095758- EEG and Brain-Computer Interfaces
- ECG Monitoring and Analysis
- Infant Development and Preterm Care
- Non-Invasive Vital Sign Monitoring
- Blind Source Separation Techniques
- Neonatal and fetal brain pathology
- Infant Health and Development
- Seismic Imaging and Inversion Techniques
- Underwater Acoustics Research
- Neuroscience of respiration and sleep
- Speech and Audio Processing
- Obstructive Sleep Apnea Research
- Neonatal Respiratory Health Research
- Heart Rate Variability and Autonomic Control
- Context-Aware Activity Recognition Systems
- Muscle activation and electromyography studies
- Time Series Analysis and Forecasting
- Reservoir Engineering and Simulation Methods
- French Language Learning Methods
- Hydrocarbon exploration and reservoir analysis
- Advanced Sensor and Energy Harvesting Materials
- Underwater Vehicles and Communication Systems
- Target Tracking and Data Fusion in Sensor Networks
- Parkinson's Disease Mechanisms and Treatments
- Phonocardiography and Auscultation Techniques
Centre Hospitalier Universitaire de Rennes
2019-2025
Inserm
2013-2025
Université de Rennes
2013-2025
Laboratoire Traitement du Signal et de l'Image
2011-2021
Institut de Recherche en Informatique et Systèmes Aléatoires
2003
Centre National de la Recherche Scientifique
2003
Milieux environnementaux, transferts et interactions dans les hydrosystèmes et les sols
2003
Institut national de recherche en informatique et en automatique
2003
Monitoring sleep of premature infants is a vital aspect clinical care, as it can reveal potential future pathologies and health issues. This study presents novel approach to automatically estimate track Quiet Sleep (QS) in 33 newborns using ECG, respiration, video motion features. Using an annotated dataset from 15 neonates (10 preterm, 5 full-term) encompassing 127.2 hours, comprehensive feature extraction selection process was employed. Three classifiers (Random Forest, Logistic...
In this paper, we perform complex network analysis on a connectivity dataset retrieved from monitoring system in order to classify simple daily activities. The is composed of set wearable sensing modules positioned the subject's body and data consists correlation between each pair modules. A number measures are then computed followed by application statistical significance feature selection methods. These methods were implemented for purpose reducing total required provide accurate activity...
This paper deals with the conception of a new system for sleep staging in ambulatory conditions. Sleep recording is performed by means five electrodes: two temporal, frontal and reference. configuration enables to avoid chin area enhance quality muscular signal hair region patient convenience. The electroencephalopgram (EEG), eletromyogram (EMG), elctrooculogram (EOG) signals are separated using Independent Component Analysis approach. compared standard analysis polysomnographic recordings...
In neonatal electroencephalography (EEG) heart activity is a major source of artifacts which can lead to misleading results in automated analysis if they are not properly eliminated. this work we propose combination empirical mode decomposition (EMD) and adaptive filtering (AF) cancel electrocardiogram (ECG) noise simplified EEG montage for preterm infants. The introduction EMD prior AF allows selectively remove ECG preserving at maximum the original characteristics EEG. Cleaned signals...
Background: Sleep is an important determinant of brain development in preterm infants. Its temporal organization varies with gestational age (GA) and post-menstrual (PMA) but little known about how sleep develops very The objective was to study the correlation between quiet (QS) maturation premature infants without severe complications during their neonatal hospitalization. Methods: Percentage time spent QS average duration intervals (ADI) were analyzed from a cohort newborns no included...
Video-based motion analysis recently appeared to be a promising approach in neonatal intensive care units for monitoring the state of preterm newborns since it is contact-less and noninvasive. However important remove periods when newborn absent or an adult present from analysis. In this paper, we propose method automatic detection presence incubator open bed. We learn specific model each bed type as camera placement differs lot encountered situations are different between both. break...
<b><i>Background:</i></b> The influence of the first immunization on cardiorespiratory (CR) stability in very preterm infants is still a controversial subject. <b><i>Objectives:</i></b> To describe changes induced by heart and respiratory rate variability (HRV-RRV) to test potential association between preimmunization profiles postimmunization CR events. <b><i>Methods:</i></b> Continuous 72-hour recordings 2.5-hour...
This study proposes a method to facilitate the remote follow up of patients suffering from cardiac pathologies and treated with an implantable device, by synthesizing 12-lead surface ECG intracardiac electrograms (EGM) recorded device. Two methods (direct indirect), based on dynamic time-delay artificial neural networks (TDNNs) are proposed compared classical linear approaches. The direct aims estimate 12 different transfer functions between EGM each signal. indirect is preliminary...
Current cardiac implantable devices offer improved processing power and recording capabilities. Some of these already provide basic telemonitoring features that may help to reduce health care expenditure. A challenge is posed in particular for the patient's electrical activity. Indeed, only intracardiac electrograms (EGMs) are acquired by implanted device signals difficult analyze directly clinicians. In this paper, we propose a patient-specific method synthesize surface electrocardiogram...
Cry analysis has been proven to be an inescapable tool evaluate the development of preterm infants. However, date, only a few authors proposed automatically extract spontaneous cry events in real context Neonatal Intensive Care Units. In fact, this is challenging since wide variety sounds can also occur (e.g., alarms, adult voice). communication, new method for extraction from life recordings long duration presented. A strategy based on initial segmentation between silence and sound events,...
In this paper, we present an activity classification-based algorithm for the automatic detection of Levodopa Induced Dyskinesia in Parkinson's Disease (PD) patients. Two PD patients experiencing motor fluctuations related to chronic therapy performed a protocol simple daily life activities on at least two different occasions. A Random Forest classifier was able classify by with overall accuracy 86%. Based detected activity, K Nearest Neighbor presence dyskinesia ranging from 75% 88%.
Cry analysis is an important tool to evaluate the development of preterm infants. However, context Neonatal Intensive Care Units challenging, since a wide variety sounds can occur (e.g., alarms and adult voices). In this paper, method extract cries proposed. It based on initial segmentation between silence sound events, followed by feature extraction resulting audio segments cry non-cry classification. A database 198 events coming from 21 newborns 439 was created. Then, set...