A5082298204- Robot Manipulation and Learning
- Soft Robotics and Applications
- Parkinson's Disease Mechanisms and Treatments
- Robotic Mechanisms and Dynamics
- Teleoperation and Haptic Systems
- Stroke Rehabilitation and Recovery
- COVID-19 diagnosis using AI
- Neurological disorders and treatments
- EEG and Brain-Computer Interfaces
- Muscle activation and electromyography studies
- Autism Spectrum Disorder Research
- Prosthetics and Rehabilitation Robotics
- Advanced Control Systems Design
- Non-Invasive Vital Sign Monitoring
- Advanced Sensor and Energy Harvesting Materials
- AI in cancer detection
- Brain Tumor Detection and Classification
- Robotic Path Planning Algorithms
- Advanced Malware Detection Techniques
- Privacy, Security, and Data Protection
- Stability and Controllability of Differential Equations
- Artificial Intelligence in Healthcare
- ECG Monitoring and Analysis
- Internet Traffic Analysis and Secure E-voting
- Network Security and Intrusion Detection
Universitatea Națională de Știință și Tehnologie Politehnica București
2016-2025
Clinical Emergency Hospital Bucharest
2019
Parkinson’s disease patients face numerous motor symptoms that eventually make their life different from those of normal healthy controls. Out these symptoms, tremor and bradykinesia, are relatively prevalent in all stages this disease. The assessment is usually performed by traditional methods where the accuracy results still an open question. This research proposed a solution for objective bradykinesia subjects with PD (10 older adults aged greater than 60 years 10 bradykinesia) 20 years....
Human Action Recognition (HAR) is a branch of computer vision that deals with the identification human actions at various levels including low level, action and interaction level. Previously, number HAR algorithms have been proposed based on handcrafted methods for recognition. However, techniques are inefficient in case recognizing level as they involve complex scenarios. Meanwhile, traditional deep learning-based approaches take entire image an input later extract volumes features, which...
Predicting diseases in the early stages is extremely important. By taking advantage of advances deep learning and fuzzy logic techniques, a new model proposed this paper for disease evaluation depending on adaptive neuro-fuzzy inference system (ANFIS) with genetic algorithm (GA) classification, pre-trained DenseNet-201 feature extraction, addition to whale optimization (WOA) selection. Two medical databases (chest X-ray MRI brain tumor) diagnosis two types were used as input suggested model....
Parkinson’s disease (PD) is one of the most unremitting and dynamic neurodegenerative human diseases. Various wearable IoT devices have emerged for detecting, diagnosing, quantifying PD, predominantly utilizing inertial sensors computational algorithms. However, their proliferation poses novel challenges concerning security, privacy, connectivity, power optimization. Clinically, continuous monitoring patients’ motor function imperative optimizing Levodopa (L-dopa) dosage while mitigating...
While Electronic Prescription Systems (EPS) adoption varies across EU Member States, there's a lack of comprehensive comparative analysis. Existing studies focus on single EPSs, employ diverse methodologies, and up-to-date data. This study fills this gap by providing overview EPS development, functionalities, usage statistics in each State. Most States widely adopted 2022, with exceptions including Germany, France, Luxembourg, where pilot projects or just plans existed at that time. Out the...
This study explores the development of an Open Artificial Intelligence (AI) decision-making solution, integrating blockchain technology with artificial intelligence to streamline organizational processes. Blockchain’s characteristics transparency, incorruptibility, and decentralized validation are leveraged build a platform that ensures secure transparent decision-making. The platform’s architecture integrates user-friendly frontend robust backend, enabling users create accounts, manage...
Classifying visual information is an apparently simple and effortless task in our everyday routine, but can we automatically predict what see from signals emitted by the brain? While other researchers have already attempted to answer this question, are first show that a commercially available BCI could be effectively used for image classification real-world scenarios – when testing takes place at completely different time than training data collection. The difficult, as it requires relating...
COVID-19 typically known as Coronavirus disease is an infectious caused by a newly discovered coronavirus. Currently detection of coronovirus depends on factors like the patients' signs and symptoms, location where person lives, travelling history close contact with any patient. In order to test patient, healthcare provider uses long swab take nasal sample. The sample then tested in laboratory setting. If coughing up saliva (sputum), emitted for testing. diagnosis becomes even more critical...
This paper overviews the state–of–the–art in upper limb robot–supported approaches, focusing on advancements related mechatronic devices for patients' rehabilitation and/or assistance. Dedicated to technical, comprehensively methodological and global effectiveness improvement this inter–disciplinary field of research, it includes information beyond therapy administrated clinical settings – but with no diminished safety requirements. Our systematic review, based PRISMA guidelines, searched...
Today, medical image-based diagnosis has advanced significantly in the world. The number of studies being conducted this field is enormous, and they are producing findings with a significant impact on humanity. databases created skyrocketing. Examining these data crucial to find important underlying patterns. Classification an effective method for identifying This work proposes deep investigation analysis evaluate diagnose image using various classification methods critically methods’...
Early detection of COVID-19 positive people are now extremely needed and considered to be one the most effective ways how limit spreading infection. Commonly used screening methods reverse transcription polymerase chain reaction (RT-PCR) or antigen tests, which need periodically repeated. This paper proposes a methodology for detecting disease in non-invasive way using wearable devices analysis bio-markers artificial intelligence. have reused publicly available dataset containing COVID-19,...
In the context in which it was demonstrated that humanoid robots are efficient helping children diagnosed with autism exploring their affective state, this paper underlines and proves efficiency of a previously developed machine learning-based mobile application called PandaSays, improved integrated an Alpha 1 Pro robot, discusses performance evaluations using deep convolutional neural networks residual networks. The model trained MobileNet network had accuracy 56.25%, performing better than...
The control problem of a class tentacle arm, with variable length, that can achieve any position and orientation in 3D space increase the length order to get better constraint operator is presented. First, dynamic model system inferred. In avoid difficulties generated by complexity nonlinear integral - differential model, based on artificial potential method. Then, method used for constrained motion an environment obstacles. Numerical simulations spatial planar models are presented...
The paper focuses on hyper-redundant arms with continuum elements that perform the grasping function by coiling. First, there is concern dynamic model of arm for position control during non-contact operations environment. A frequency stability criterion based Kahman - Yakubovich Popov Lemma and P PD algorithms proposed. Then, in contact a load analyzed. dynamics system are discussed an extension endorsed. SMA actuators introduced. Numerical simulations experimental results motion toward...
This paper is a component of research project regarding the design and development an Intelligent Haptic Robot-Glove for rehabilitation patients that have diagnosis cerebrovascular accident. In this paper, control system hand exoskeleton discussed. order to avoid complex problems given by distributed sensor network, treats problem compliance force using observers: velocity observer, observer disturbance observer. Two solutions are First, effects eliminated cascade closed loop with observers....
The paper treats the control problem of a class robots constituted by chain continuum segments. technological model basis is central, long and thin, highly flexible elastic backbone. segment system decoupled one. main parameters arm are determined curvature gradient. dynamic inferred. primary benefit proposed method that equations represented set ODE's in time instead PDE's space, new gradient lumped parameter used. A sliding mode used order to achieve desired shape arm. stability...
The grasping control problem for a hyperredundant manipulator is presented. dynamic model derived by using Lagrange equations developed infinite dimensional systems. algorithms the position and force are proposed. arm fluid pressure inferred conditions that ensure stability of motion discussed. Numerical simulations
Autism, or autism spectrum disorder (ASD), is a development characterized by difficulties in communication and speech it cannot be cured entirely. The hardest part for individuals with to express their feelings emotions towards others. This paper describes mobile application developed using machine learning algorithms multiclass image classification that needs as input just children's drawings. solution dedicated specially children between 2 5 years old predicts emotional state. has been...
In last decade, machine learning (ML) techniques increased the capability to automatically learn experience without being explicitly programmed. Different methods are used for a number of tasks such as image processing, predictive analytics, data mining, and classification, regression, clustering, dimensionality reduction. ML is widely utilized in medical imaging research field. This paper introduces survey on processing it focuses two main types (supervised unsupervised learning) importance...
This paper deals with the fractional order control for complex systems, hand exoskeleton and sensors, that monitor human behavior. The laws based on physical significance variables, models, delays or without delays, are proposed discussed. Lyapunov techniques methods derive from Yakubovici-Kalman-Popov lemma used frequency criterions ensure asymptotic stability of closed loop system inferred. An observer is sensors. system, hand-observer, studied sector laws. Numerical simulations an...
This paper treats the control problem of a class hyper-redundant robots. The dynamic model arm is described by hyperbolic partial differential equations with uncertain components. By using spatial weighted error control, infinite dimensional system becomes finite-dimensional problem. stability analysis and resulting controllers are obtained concept boundary geometric technique. A robust algorithm that based on sliding mode discussed. tendon determines evolution toward prescribed switching...