A5041636394- Embedded Systems Design Techniques
- Artificial Intelligence in Healthcare
- Emergency and Acute Care Studies
- Machine Learning in Healthcare
- Context-Aware Activity Recognition Systems
- Healthcare Technology and Patient Monitoring
- Parallel Computing and Optimization Techniques
- Energy Efficient Wireless Sensor Networks
- Non-Invasive Vital Sign Monitoring
- Electronic Health Records Systems
- Artificial Intelligence in Healthcare and Education
- Microbial Metabolic Engineering and Bioproduction
- COVID-19 diagnosis using AI
- Gene Regulatory Network Analysis
- Advancements in PLL and VCO Technologies
- Fire Detection and Safety Systems
- Real-time simulation and control systems
- Fire dynamics and safety research
- Healthcare Operations and Scheduling Optimization
- Cardiac Health and Mental Health
- Mobile Health and mHealth Applications
- Medical Coding and Health Information
- Electromagnetic Compatibility and Noise Suppression
- Atomic and Subatomic Physics Research
- Computational Physics and Python Applications
Aristotle University of Thessaloniki
2019-2024
Infineon Technologies (Austria)
2018
National and Kapodistrian University of Athens
2006-2016
CISC Semiconductor (Austria)
2016
Athens State University
2007
Institute of Nuclear and Particle Physics
2006
<sec> <title>BACKGROUND</title> A significant number of individuals undergoing Coronary Computed Tomography Angiography (CCTA) for suspected Artery Disease (CAD) have non-obstructive or no CAD. There is a need clinically proven models that can predict the pre-test probability (PTP) stable CAD and help to identify low-risk individuals. Optimizing patient stratification paramount importance improve diagnostic yield cost-effectiveness. </sec> <title>OBJECTIVE</title> study being carried out...
The digitization of the healthcare domain has potential to drastically improve services, reduce time diagnosis, and lower costs. However, digital applications for need be interoperable maximize their potential. Additionally, with rapid expansion Artificial Intelligence (AI) and, specifically, Machine Learning (ML), large amounts diverse types data are being utilized. Thus, achieve interoperability in such applications, adoption common semantic models becomes imperative. In this paper, we...
We have designed a modular SOM systolic architecture that can classify data vectors with thousands of elements in real time. The is described as soft IP core synthesizable VHDL. neural network size, the input dimension, weight and element bitwidth precision etc. are all designer tunable parameters. Several instances been synthesized their performance evaluated for different Xilinx Virtex-II Virtex II-Pro FPGAs. Moderate to large size networks process many 4096 fit into single FPGA device...
Although software engineers have high performance algorithms that could be implemented power-efficiently as embedded Systems on Chip (SoC) with modern FPGAs, there is still no easy path for them to a hardware realization, mainly due the lack of appropriate design tools. We present an overview tool we developed boost productivity processor-centric SoC designs FPGAs. Our called SysPy (System Python) exploits strengths popular Python scripting language and acts "glue software" between mature...
We present a small scale sensor network application as testbench to explore different setups in terms of hardware/ software, protocol and data processing/storage scheme options, focusing on alert-based systems with long idle times. Our specifications required the deployment using MQTT over an encrypted TLS connection. focused nodes low-power consumption profile, well formalization protocol's basic elements, by clearly defining topic/message used across network. In addition, we experimented...
We present SysPy (System Python) a tool which exploits the strengths of popular Python scripting language to boost design productivity embedded System on Chips for FPGAs. acts as “glue” software between mature HDLs, ready-to-use VHDL components and programmable processor soft IP cores. can be used to: (i) automatically translate hardware described in into synthesizable VHDL, (ii) capture top-level structural descriptions processor-centric SoCs Python, (iii) implement all steps necessary...
System Python (SysPy) is a public domain design tool using to facilitate all prototyping phases of processor-centric SoCs for FPGAs. In previous work we used as high-level description mechanism hardware modules and connect them embedded processors. this paper show how SysPy can also functional verification SoC when an Architectural Description Language (ADL), helping designer make decisions about key architectural features early in the phase. To best our knowledge unique supporting...
Preterm birth (PTB) is defined as delivery occurring before 37 weeks of gestation. In this paper, Artificial Intelligence (AI)-based predictive models are adapted to accurately estimate the probability PTB. doing so, pregnant women' objective results and variables extracted from screening procedure in combination with demographics, medical history, social other data used. A dataset consisting 375 women used a number alternative Machine Learning (ML) algorithms applied predict The ensemble...
Recent statistics have demonstrated that Emergency Departments (EDs) in Greece lack organization and service. In most cases, patient prioritization is not automatically implemented. The main objective of this paper to present IntelTriage, a smart triage system, dynamically assigns priorities patients an ED monitors their vital signs location during stay the clinic through wearable biosensors. Initital scenarios functional requirements are presented as preliminary results.
Obstructive coronary artery disease (CAD) is characterized as significant upon detection of stenosis diameter. In this paper, we adapt Artificial Intelligence (AI)-based predictive models to accurately estimate the pretest likelihood obstructive CAD on computed tomography angiography (CCTA) in patients with suspected CAD. doing so, use patients' objective results and variables extracted from screening procedure combination demographics, medical history, social other data. We a dataset...
The overwhelming volume of patients in emergency departments (EDs) is a significant problem that hinders the delivery high quality healthcare. Despite their great value, triage protocols are challenging to put into practice. This paper examines utility prediction models as tool for clinical decision support, with focus on medium-severity defined by ESI algorithm. 689 cases medium-risk were gathered from AHEPA hospital, evaluated, and data fed three classifiers: XGBoost (XGB), Random Forest...
Emergency Department (ED) overcrowding is a major global issue of public health concerning patients' safety and quality care delivery, leading to increased mortality, costs due prolonged in-hospital length stay readmissions The main goal this paper present IntelTriage its functional non-functional requirements. smart triage system that automatically prioritizes ED's patients, continuously monitors their vital signs also tracks location through wearable device intelligent clinical decision...
Appointment Scheduling (AS), typically serves as the basis for majority of non-urgent healthcare services and is a fundamental healthcare-related procedure which, if done correctly effectively, can lead to significant benefits facility. The main objective this work present ClinApp, an intelligent system able schedule manage medical appointments collect data directly from patients.
We present a set of tools for digital design, integration and verification mixed-signal Application Specific Integrated Circuits (ASIC) developed within our design team. have chosen Python the development tools. By drawing on Python's features we targeting many steps required across flow: a) Complex blocks are auto-generated, where generates, according to input files thousands Register Transfer Level (RTL) lines code. b) Using tool full RTL hierarchy can be parsed using extracted information...
Massive incident response in health environments, usually overwhelms local and regional resources. Even developed countries, growing evidence exists that the lack of organisation service provision is one major issues Emergency Departments (EDs). In such highly demanding most cases, patients are not prioritised an automated intelligent way. Hence, critical time spent preparation processes rather addressing ones. this paper, we introduce architecture IntelTriage, which a novel smart triage...
Emergency Departments globally suffer overcrowding due to the lack of adequate capacity and/or guidelines and policies for triaging patients. Medical service quality improvement requires optimal patient prioritization according their level urgency. This study aims evaluate classification models predicting admission or discharge incidents triaged as 3 Severity Index algorithm. As such, adult visits were examined from a publicly available dataset. Feature Importance was used assessment each...
Clin App is a platform streamlining medical appointment management and patient data collection using conversational agent. Focused on healthcare professionals patients, it offers automation, questionnaire creation, management. This work showcases ClinApp's microservices-based architecture its user-centered design.
A soft IP core, expressed in parametric VHDL, has been developed and used to synthesize different multiprocessor Systems on Chip (SoCs) for the stochastic simulation of large-size biochemical reaction networks based Gillespie's First Reaction Method (FRM). The SoCs can be configured have up N=8 Processing Elements simulate efficiently hardware biomolecular with m=16K reactions. FPGA implementations are communicating a host PC via serial Ethernet ports control data transmission respectively....