ALArkadiusz Lewicki

Arkadiusz Lewicki

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Arkadiusz Lewicki has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Standardized 2D Ultrasound Sequences for Fetal Cardiac Screening: A Platform for AI Integration

    by Agnieszka Nocun, Arkadiusz Lewicki, Magdalena Dudzik, Jerzy Wegrzynowski, Malwina Merta, Piotr Bralewski, Krzysztof Serafin, Alicja Zachara, Azer Farajov, Marcin Wiechec

    Objective: The precise and consistent identification of fetal cardiac structures and functional flows is essential for the early diagnosis of congenital heart defects (CHDs), yet interobserver variability remains a significant challenge in clinical practice. Although fetal cardiac magnetic resonance imaging (MRI) has emerged as a valuable adjunct in selected cases, its routine use remains limited by long acquisition times, motion artifacts, and restricted availability in many centers. This underscores the need to optimize ultrasound‐based techniques that can reduce reliance on MRI while still providing comprehensive and reproducible fetal cardiac assessment. This article presents a study investigating the reliability of a layered imaging approach combined with standardized B‐mode and color Doppler ultrasound protocols to improve interobserver agreement in image interpretation among experts in fetal echocardiography. Methods: A dataset comprising 209 B‐mode and 205 color Doppler recordings acquired during midgestational anomaly scans (mean gestational age: 21.2 weeks) was systematically evaluated by five expert supervisors. Interobserver agreement was quantified using the prevalence‐adjusted bias‐adjusted kappa (PABAK) coefficient. To evaluate consistency across anatomical layers, one‐way ANOVA was employed, followed by post hoc analysis where applicable. Results: Most anatomical structures and functional features were consistently identified across observers, certain structures posed challenges, likely owing to their smaller size, lower visibility, or greater variability in presentation. Conclusion: By integrating structured imaging sequences with advanced statistical methodologies, this study explored the potential of improving diagnostic accuracy and standardization in fetal cardiac assessments, offering insights into the development of more reliable protocols for clinical applications and future research.

  • Open Access

    PrAViC: Probabilistic Adaptation Framework for Real-Time Video Classification

    by Magdalena Trędowicz, Marcin Mazur, Szymon Janusz, Arkadiusz Lewicki, Jacek Tabor, Łukasz Struski

    Video processing is generally divided into two main categories: processing of the entire video, which typically yields optimal classification outcomes, and real-time processing, where the objective is to make a decision as promptly as possible. Although the models dedicated to the processing of entire videos are typically well-defined and clearly presented in the literature, this is not the case for online processing, where a plethora of hand-devised methods exist. To address this issue, we present PrAViC, a novel, unified, and theoretically-based adaptation framework for tackling the online classification problem in video data. The initial phase of our study is to establish a mathematical background for the classification of sequential data, with the potential to make a decision at an early stage. This allows us to construct a natural function that encourages the model to return a result much faster. The subsequent phase is to present a straightforward and readily implementable method for adapting offline models to the online setting using recurrent operations. Finally, PrAViC is evaluated by comparing it with existing state-of-the-art offline and online models and datasets. This enables the network to significantly reduce the time required to reach classification decisions while maintaining, or even enhancing, accuracy.

  • Multiple instance learning for medical image classification based on instance importance

    by Łukasz Struski, Szymon Janusz, Jacek Tabor, Michał Markiewicz, Arkadiusz Lewicki


  • Open Access

    ProMIL: Probabilistic Multiple Instance Learning for Medical Imaging

    by Łukasz Struski, Dawid Rymarczyk, Arkadiusz Lewicki, Robert Sabiniewicz, Jacek Tabor, Bartosz Zieliński

    Multiple Instance Learning (MIL) is a weakly-supervised problem in which one label is assigned to the whole bag of instances. An important class of MIL models is instance-based, where we first classify instances and then aggregate those predictions to obtain a bag label. The most common MIL model is when we consider a bag as positive if at least one of its instances has a positive label. However, this reasoning does not hold in many real-life scenarios, where the positive bag label is often a consequence of a certain percentage of positive instances. To address this issue, we introduce a dedicated instance-based method called ProMIL, based on deep neural networks and Bernstein polynomial estimation. An important advantage of ProMIL is that it can automatically detect the optimal percentage level for decision-making. We show that ProMIL outperforms standard instance-based MIL in real-world medical applications. We make the code available.