A5102885257- Machine Learning in Bioinformatics
- Developmental Biology and Gene Regulation
- Retinal Development and Disorders
- Protein Structure and Dynamics
- Hedgehog Signaling Pathway Studies
- Computational Drug Discovery Methods
- Neurogenesis and neuroplasticity mechanisms
- Ocular Oncology and Treatments
- Semiconductor Lasers and Optical Devices
- Advanced Fiber Optic Sensors
- Pluripotent Stem Cells Research
- UAV Applications and Optimization
- Advanced Neural Network Applications
- Enzyme Structure and Function
- Renal and related cancers
- Photonic and Optical Devices
- Corneal Surgery and Treatments
- Biochemical and Molecular Research
- Neural Networks and Applications
- Congenital heart defects research
- Air Traffic Management and Optimization
- Video Surveillance and Tracking Methods
- Energy Harvesting in Wireless Networks
- Robotics and Sensor-Based Localization
- Stochastic Gradient Optimization Techniques
University of Cyprus
2010-2025
University of Nicosia
2021-2025
The University of Texas Southwestern Medical Center
2023
Children's Medical Center
2023
University of Cambridge
2006-2009
Hedgehog signaling has been linked to cell proliferation in a variety of systems; however, its effects on the cycle have not closely studied. In vertebrate retina, Hedgehog's are controversial, with some reports emphasizing increased and others pointing role exit. Here we demonstrate novel for speeding up developing retina by reducing length G1 G2 phases. These fast cycling cells tend exit early. Conversely, retinal progenitors blocked more slowly, longer phases, remain longer. may modulate...
Progenitor cells in the central nervous system must leave cell cycle to become neurons and glia, but signals that coordinate this transition remain largely unknown. We previously found Wnt signaling, acting through Sox2, promotes neural competence Xenopus retina by activating proneural gene expression. now report Sox2 inhibit differentiation Notch activation. Independently of stimulates retinal progenitor proliferation this, when combined with block on differentiation, maintains fates....
Protein Secondary Structure Prediction (PSSP) is regarded as a challenging task in bioinformatics, and numerous approaches to achieve more accurate prediction have been proposed. Accurate PSSP can be instrumental inferring protein tertiary structure their functions. Machine Learning particular Deep show promising results for the problem. In this paper, we deploy Convolutional Neural Network (CNN) trained with Subsampled Hessian Newton (SHN) method (a Free Optimisation variant), two-...
Purpose The purpose of the paper is to test use artificial neural networks (ANNs) as a tool in fraud detection. Design/methodology/approach Following review relevant literature on detection by auditors, authors developed questionnaire which they distributed auditors attending seminar. was then used develop seven ANNs usage these models Findings Utilizing exogenous and endogenous factors input variables developing different models, an average 90 per cent accuracy found prediction model. It...
Filtering of Protein Secondary Structure Prediction (PSSP) aims to provide physicochemically realistic results, while it usually improves the predictive performance. We performed a comparative study on this challenging problem, utilizing both machine learning techniques and empirical rules we found that combinations two lead highest improvement.
Machine Learning (ML) is now becoming a key driver empowering the next generation of drone technology and extending its reach to applications never envisioned before. Examples include precision agriculture, crowd detection, even aerial supply transportation. Testing projects before actual deployment usually performed via robotic simulators. However, testing assessment on-board ML algorithms daunting task. practitioners are required dedicate vast amounts time for development configuration...
We are now witnessing the extensive deployment of drones in a diverse set applications with Machine Learning (ML) constituting key enabler empowering uptake drone technology. With advancements robotics and edge computing, on-board ML is on uprise. However, testing solutions for before release to production daunting task practitioners. This usually involves emulator collect various performance indicators ranging from algorithm correctness resource utilization. Thus, thoroughly evaluate...
Trying to extract features from complex sequential data for classification and prediction problems is an extremely difficult task. This task even more challenging when both the upstream downstream information of a time-series important process sequence at specific time-step. One typical problem which falls in this category Protein Secondary Structure Prediction (PSSP). Recurrent Neural Networks (RNNs) have been successful handling data. These methods are demanding terms time space...
As drone technology penetrates even more application domains, Machine Learning (ML) is becoming a key driver enabling intelligence in the sky. However, ML Practitioners and Drone Application Operators are faced with several challenges when wanting to test ML-driven applications early design phase. These include development configuration of experiment use-cases over robotics simulator along collection assessment desired KPIs which can range from algorithm accuracy resource utilization impact...
We examine the use of state-of-the-art distributed sensing systems to extract temperature information from optical fibre infrastructure already Electricity Authority Cyprus power distribution network (~25-year old installation); as a means in underground cables. The fibres are collocated with existing cables, for purpose line monitoring cable joints that prone failure, along general unusual behaviour and potential fault conditions. Detection is achieved using DTS: Distributed Temperature...
In this work we utilize multimode optical fibers for the detection of simulated errors or failures in underground power cables. It is known that cases failure transmission cables overheat locally, they become a hot-spot, and it extremely difficult to detect locate problem. The proposed methodology as follows, having an electric cable simulate various temperature profiles whilst fiber was placed selected distances away from our fault examine performance fiber. way aim stabilize operation...
We present a study on the application of machine learning to optical fibre distributed sensing, with data recovered using state-of-the-art, commercial BOTDR sensing system; temperature information was extracted from power line distribution networks that are part Electricity Authority Cyprus. A approach implemented for prediction task finding points abnormal behaviour, mimicking cable joints prone failure, along general monitoring unusual behaviour and potential fault conditions; is binary...