- EEG and Brain-Computer Interfaces
- ECG Monitoring and Analysis
- Blind Source Separation Techniques
- Non-Invasive Vital Sign Monitoring
- Heart Rate Variability and Autonomic Control
- Phonocardiography and Auscultation Techniques
- Neuroscience and Neural Engineering
- Neural Networks and Applications
- Advanced Sensor and Energy Harvesting Materials
- Analog and Mixed-Signal Circuit Design
- Neural dynamics and brain function
- AI in cancer detection
- Muscle activation and electromyography studies
- Neonatal and fetal brain pathology
- Context-Aware Activity Recognition Systems
- IoT and Edge/Fog Computing
- Artificial Intelligence in Healthcare
- Advanced Control Systems Optimization
- Medical Image Segmentation Techniques
- Functional Brain Connectivity Studies
- Infrared Thermography in Medicine
- Advanced Memory and Neural Computing
- Force Microscopy Techniques and Applications
- Telemedicine and Telehealth Implementation
- Preterm Birth and Chorioamnionitis
Universitatea Națională de Știință și Tehnologie Politehnica București
2008-2018
University of Bucharest
2016-2017
National Research & Development Institute for Textiles and Leather
2016
Fetal ECG (FECG) monitoring using abdominal maternal signals is a non-invasive technique that allows early detection of changes in fetal wellbeing. Several other signal components have stronger energy than the FECG, most important being (MECG) and, especially during labor, uterine EMG. This study proposes new method to subtract MECG after detecting and removing segments with high-amplitude variations due contractions. The removes from an approximation current segment based on linear...
Motor imagery is the mental simulation of a motor act that includes preparation for movement, passive observations action and operations representations implicitly or explicitly. as immediate movement likely involves executive brain regions. Implicit are considered to underlie cognitive functions. Another problem concerning neuro-imaging studies on performance imagination very difficult control. The ability an individual control its EEG may enable him communicate without being able their...
Abstract An innovative concept for synchronization analysis between heart rate (HR) components and rhythms in EEG envelopes is represented; it applies time-variant analyses to variability (HRV) EEG, was tested children with temporal lobe epilepsy (TLE). After a removal of ocular movement-related artifacts, band activity computed by means the frequency-selective Hilbert transform providing frequency bands. Synchronization HRV quantified Morlet wavelet coherence. A surrogate data approach...
Detection of pectoral muscle in mammograms is an important pre-processing segmentation step. The one the few anatomical features that appears clearly and reliably medio-lateral oblique view mammograms. This new method overcomes limitation straight-line representation considered our initial investigation using Hough transform.
The electrical activity of the uterus, i.e. electrohysterogram (EHG), is one most prominent tool for preterm labour. There no standard acquisition set up and often EHG corrupted with different types noise: maternal fetal electrocardiogram (mECG, fECG), skeletal muscles, movement artifacts, power line interference (PLI) etc. Moreover, some these noises overlap in frequency domain EHG. Thus, simple linear filtering approaches are not adequate. In this paper empirical mode decomposition (EMD),...
The respiratory rate is a vital parameter that can provide valuable information about the health condition of patient. extraction from photoplethysmographic signal (PPG) was actually encouraged by reported results, our main goal being to obtain accurate estimation PPG signal. We developed fusion algorithm identifies best derived signals, which possible extract rate; based on these, global computed using proposed algorithm. qualitatively tested real signals recorded an acquisition system we...
The fetal ECG can be detected in the recorded abdominal signals. A new procedure to compute heart rate (FHR) is proposed. signal first preprocessed order remove baseline and uterine contractions. Then of mother (MECG) removed using coherent averaging optimizing averaged MECG template. channels containing clearest (FECG) are identified by autocorrelation function. FECG enhanced cross correlation between two that show strongest FECG. This enhancement possible since residual noise after removal...
Fetal ECG (fECG) signal offers valuable information about the health state of fetus during pregnancy that can assist physicians taking timely appropriate decisions and labor. Thus, FECG extraction from composite abdominal signals (ADS) by powerful advance processing methods is becoming more an important requirement for fetal monitoring. This paper presents evaluation four Independent Component Analysis (ICA) algorithms, when used fECG extraction, their performance comparison considering...
This paper presents a wearable system for heart rate monitoring based on photoplethysmography (PPG). method sensing the cardiovascular pulse analyzing of light variations biological tissues. The interaction with tissue is complex phenomenon, therefore, involves scattering, absorption and reflection. In our work, we used PPG reflectance mode. Blood absorbs more than others tissues reduction blood amount shows an increased intensity scattered back reflected from tissue. approach very...
This paper presents the test of skin electrical conductance (SC) in function temperature and body physiology. It is known that electrodermal response becomes a good conductor electricity when an external or internal stimulus occurs. For measuring moisture we used humidity smart sensors. The data acquisition system used, consist sensors, signal conditioning circuitry analog to digital converter. discretization continuous lower power consumption sampled e after 30 seconds. We propose sampling...
Detection and characterization of cancer tumors in mammograms is vital daily clinical practice. The problem detecting possible areas very complex due, on one hand, to the diversity shape ill tissue other hand poorly defined border between healthy cancerous zone. Even though it has been studied for many years, there are still remaining challenges directions future research such as developing better enhancement segmentation algorithms. performance Self Organizing Map (SOM) suspicious regions...
The impact of stress caused by car driving on heart's activity has been extensively studied for over 60 years. In the present paper, we influence drivers' heart rate. algorithm was applied electrocardiographic signals from 6 drivers, and resulting rate variability parameters in time domain were analyzed comparing those normal conditions to conditions. results show that are closely correlated. This finding could be used create an intelligent system would monitor adapt cabin order relax...
This paper presents the experimental part for discrete time signal acquisition in case of respiration rate monitoring using analog flexible sensor. The hardware physiological consists ARDUINO shield and a tubular sensor connected with electrodes. signals are generated by resistance parameter vary through semiconductor when this is stretched. It known fact that act, diaphragm muscle increase decrease rhythmic way volume. has been made polyester stainless steel threads. electronic system...
According to the World Health Organization, breast cancer is most common suffered by women in world, which during last two decades, has increased mortality developing countries. Mammography best method used for screening; problem of detecting possible areas very complex due, on one hand, diversity shape ill tissue and, other poorly defined border between healthy and cancerous zone. An automated technique alignment right left images been developed use computerized analysis bilateral images....
The fetal heart rate (FHR) is a fundamental tool for clinical monitoring, when investigating the wellbeing of fetus. detection QRS complex electrocardiogram using abdominal signal enables non-invasive evaluation FHR. There are many causes FHR variability, but slight variance indicates good evolution pregnancy. current study evaluates RPA model 5-channels records (ADS), in order to chaotic dynamic and reveal correlations time-domain.