Tadas Žvirblis

Tadas Žvirblis serves as an Associate Professor within the Interdisciplinary Statistical Research Group at Vilnius University's Institute of Data Science and Digital Technologies, maintaining his research office at Akademijos St. 4, room 604A in Vilnius. His academic profile bridges theoretical statistics with practical engineering and medical applications through advanced computational methodologies.

His research program centers on machine learning and deep learning innovations for complex signal analysis, with dual specializations in biomedical diagnostics (EEG, NIRS, cardiovascular monitoring) and industrial systems (conveyor mechanics, gear fault detection, engine emissions). Key methodological contributions include novel data augmentation techniques for time series, generative modeling of vibration signals, and prognostic frameworks for reliability engineering, demonstrating consistent interdisciplinary collaboration across medical and engineering domains.

Analysis of his 13 publications from 2023-2025 reveals a strategic research trajectory applying deep learning to data-scarce scenarios, particularly in biomedical signal interpretation and industrial predictive maintenance. His work shows increasing focus on clinical applications since 2024, including ECMO mortality prediction and aortic morphology studies, while maintaining strong industrial engineering output through IEEE conference publications on conveyor systems and engine diagnostics.

Dr. Žvirblis actively supervises doctoral research as Senior Researcher for Gajane Mikalkėnienė's project (2023-2027) developing EEG-based depression diagnosis methods under Informatics field N 009. His grant portfolio includes multiple industry-collaborative projects evidenced by co-authorship with clinical researchers and engineering teams across Lithuania, Poland, and Germany.

As a core member of the Interdisciplinary Statistical Research Group, he contributes to the unit's mission of advancing statistical methodologies for real-world data challenges, with particular emphasis on time-series analysis in non-stationary environments. His laboratory work integrates signal processing hardware with deep learning frameworks to address industrial automation and medical monitoring challenges.