Products
What we build
Intelligent tools designed to enhance clinical efficiency and elevate the quality of healthcare.
-
ABIGAIL
ABIGAIL develops an AI-based system for noninvasive classification of high-grade brain gliomas by integrating MRI imaging with clinical and laboratory data. This approach aims to accelerate diagnosis by replacing invasive biopsies, reduce complications and healthcare costs, and enable faster, more precise treatment decisions. Current Status: TRL 2 → TRL 3.
-
BRAIN
In brAIn, we are developing an AI-powered system for automatic detection of intracranial bleeding from CT scans. The solution will prioritize critical cases for immediate radiologist review, enabling faster diagnosis and timely intervention in life-threatening emergencies. Current Status: TRL 2 → TRL 3.
-
BRIDGE
BRIDGE is an advanced data harmonization platform that unifies heterogeneous and unstructured medical databases, enabling seamless AI training and deployment. Its research arm, CLEAN (Clinical Language Engine for Automated Narratives), leverages AI to automatically summarize patients' disease evolution for shift handovers, discharge reports, and examination requests. Current Status: TRL 3.
-
CLEAN
Project CLEAN (Clinical Language Engine for Automated Narratives) is an AI-driven system designed to automatically generate structured clinical documents—such as discharge summaries or DRG codes—directly from unstructured hospital data like admission letters and daily clinical notes. Current Status: TRL 3.
-
GhostEMR
GhostEMR is a specialized, GDPR-compliant software solution designed for the secure anonymization and de-identification of Electronic Health Records (EHR). It processes a wide range of medical documentation — from outpatient consults and admission letters to daily clinical notes, laboratory and imaging results, inpatient consults, discharge summaries, and DRG codes. GhostEMR uses advanced algorithms to remove or mask personally identifiable information while preserving the statistical and clinical integrity of datasets, enabling healthcare institutions to safely share and analyze patient data for research, innovation, and quality improvement. Current Status: TRL 9.
Team
The people behind it
A group of passionate researchers and clinicians committed to innovation and excellence.
Peter Drotar
Peter Drotar, MSc, PhD, serves as a full professor at the Technical University of Kosice. His expertise lies in machine learning, medical imaging, and biomedical signal processing. A former Honeywell scientist and ESET Science Award finalist, he was named Scientific Person of the Year at the Technical University of Kosice in 2025. He has also acted as principal investigator in several research projects with a total budget exceeding €2 million.
Jakub Gazda
Jakub Gazda, MD, PhD, is a physician and researcher at Pavol Jozef Safarik University and L. Pasteur University Hospital in Kosice, Slovakia. His work focuses on liver cirrhosis, its complications, and AI-driven diagnostic tools. He completed research stays at Sapienza University of Rome and Hospital Clínic Barcelona and serves as an external consultant for Signant Health.
Matej Gazda
Matej Gazda, MSc, PhD, is a researcher at the Technical University of Kosice specializing in artificial intelligence for medical imaging, with a focus on advanced image segmentation and diagnostic applications. He completed a one-year research stay at Polytechnique Montréal. Named the Top Student Personality of Slovakia for 2021/2022, he contributes to the development of innovative technologies in medical AI.
Research
Publications
Our contributions to advancing medical AI and clinical science.
Prospective evaluation of speech as a digital biomarker for covert hepatic encephalopathy
J. Gazda, J.C. García-Pagán, S. Drazilova, P. Drotar, M. Hires, M. Gazda, M. Janicko, A. Baiges, P. Jarcuska
npj Digital Medicine (2025)
Convolutional neural network ensemble for Parkinson's disease detection from voice recordings
M. Hireš, M. Gazda, P. Drotár, N. D. Pah, M. A. Motin, D. K. Kumar
Computers in Biology and Medicine (2022)
Multiple-Fine-Tuned Convolutional Neural Networks for Parkinson's Disease Diagnosis From Offline Handwriting
M. Gazda, J. Gazda, S. Kadoury, R. Kanasz, P. Drotar
Computer Methods and Programs in Biomedicine (2025)
echoGAN: Extending the field of view in Transthoracic Echocardiography through conditional GAN-based outpainting
J. Gazda, J.C. García-Pagán, S. Drazilova, P. Drotar, M. Hires, M. Gazda, M. Janicko, A. Baiges, P. Jarcuska
Computer Methods and Programs in Biomedicine (2025)
Handwriting-Based Classification of Hepatic Encephalopathy Using Nonlinear Complexity Features
K. Demcakova, P. Drotar, M. Hires, J. Gazda, P. Jarcuska
IEEE Symposium on Computer-Based Medical Systems (2025)
Artificial Intelligence and Its Application to Minimal Hepatic Encephalopathy Diagnosis
J. Gazda, P. Drotar, S. Drazilova, J. Gazda, M. Gazda, M. Janicko, P. Jarcuska
Journal of Personalized Medicine (2021)
Self-Supervised Deep Convolutional Neural Network for Chest X-Ray Classification
M. Gazda, J. Plavka, J. Gazda, P. Drotar
IEEE Access (2021)
Treatment response to ursodeoxycholic acid in primary biliary cholangitis: A systematic review and meta-analysis
J. Gazda, S. Drazilova, M. Gazda, M. Janicko, T. Koky, M. Macej, M. Carbone, P. Jarčuška
Digestive and Liver Disease (2023)
Towards annotation-efficient segmentation via image-to-image translation
E. Vorontsov, P. Molchanov, M. Gazda, C. Beckham, J. Kautz, S. Kadoury
Medical Image Analysis (2022)
Partners
Collaborations
Working with leading institutions to advance medical AI.
