CIRCA: comprehensible online system in support of chest X-rays-based screening by COVID-19 example

Chest X-rays (CXRs) are widely used for diagnosing respiratory diseases, including the recent example of COVID-19. Supervised deep learning techniques can help detect cases faster and monitor disease progression. However, they are usually developed using coarser data annotations, which may insufficiently capture the heterogeneous disease portrait. We propose the pipeline called CIRCA ( https://circa.aei.polsl.pl ) for a CXR-based screening support system, developed using 6 diverse datasets. Our tool includes lung segmentation, quantitative assessment of data heterogeneity, and a hierarchical three-class decision system using a convolutional network and radiomic features. Lung segmentation showed an accuracy of ~ 94% in the validation and test sets, while classification accuracy was equal 86%, 83%, and 72% for normal, COVID-19, and other pneumonia classes in the independent test set. Three radiomically distinct subtypes were identified per class. In the hold-out set, the classification subtype-specific cross-dataset NPV ranged from 95 to 100%, with PPV from 86 to 100% for all subtypes except N3 (early stage or convalescent) and both C3 and P3 (probable co-occurrence of COVID-19). Using an independent test set gave similar results. The dataset-specific subtype proportions combined with various predictive qualities of subtypes partly explain the widely reported poor generalization of AI-based prediction systems.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2025-11-28
DOI
https://doi.org/10.1038/s41598-025-26895-0
Primary Topic
COVID-19 diagnosis using AI
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

CIRCA: comprehensible online system in support of chest X-rays-based screening by COVID-19 example

Katarzyna Sznajder, Anna Kozanecka, Wojciech Prażuch, Krzysztof Simon et al.
Scientific Reports
COVID-19 diagnosis using AI
article

CIRCA: comprehensible online system in support of chest X-rays-based screening by COVID-19 example

Katarzyna Sznajder, Anna Kozanecka, Wojciech Prażuch, Krzysztof Simon, Aleksandra Suwalska, Justyna Kozub, Agnieszka Oronowicz-Jaśkowiak, Jan Baron, Edyta Szurowska, Katarzyna Rataj, Joanna Polańska, Katarzyna Gruszczyńska, Tadeusz Jan Popiela, Paweł Rajewski, Jerzy Walecki, Barbara Giżycka, Andrzej Cieszanowski, Jerzy Jaroszewicz, Piotr Fiedor, Piotr Blewaska, Joanna Tobiasz, Bogumił Gołębiewski, Sebastian Hildebrandt, Robert Flisiak, Gabriela Zapolska, Piotr Wasilewski, Marek Socha, Małgorzata Pawłowska, Damian Piotrowski, Paweł Foszner, Michał Marczyk, Grzegorz Przybylski, Robert Pleśniak, Mateusz Rataj, Przemyslaw Chmielarz, Katarzyna Krutul-Walenciej, Krzysztof Klaude, Piotr Rabiko, Mateusz Nowak, S. Mazur, Magdalena Śliwińska, Grzegorz Drabik, Adrianna Tur
article en

Abstract

Chest X-rays (CXRs) are widely used for diagnosing respiratory diseases, including the recent example of COVID-19. Supervised deep learning techniques can help detect cases faster and monitor disease progression. However, they are usually developed using coarser data annotations, which may insufficiently capture the heterogeneous disease portrait. We propose the pipeline called CIRCA ( https://circa.aei.polsl.pl ) for a CXR-based screening support system, developed using 6 diverse datasets. Our tool includes lung segmentation, quantitative assessment of data heterogeneity, and a hierarchical three-class decision system using a convolutional network and radiomic features. Lung segmentation showed an accuracy of ~ 94% in the validation and test sets, while classification accuracy was equal 86%, 83%, and 72% for normal, COVID-19, and other pneumonia classes in the independent test set. Three radiomically distinct subtypes were identified per class. In the hold-out set, the classification subtype-specific cross-dataset NPV ranged from 95 to 100%, with PPV from 86 to 100% for all subtypes except N3 (early stage or convalescent) and both C3 and P3 (probable co-occurrence of COVID-19). Using an independent test set gave similar results. The dataset-specific subtype proportions combined with various predictive qualities of subtypes partly explain the widely reported poor generalization of AI-based prediction systems.

Scientific ReportsVol. 15(1)
Silesian University of Technology (PL), Jagiellonian University (PL), Medical University of Białystok (PL), Medical University of Silesia (PL), University of Opole (PL), Medical University of Warsaw (PL), Yale Cancer Center (US), Nicolaus Copernicus University (PL), Wroclaw Medical University (PL), Central Clinical Hospital (PL), Collegium Medicum in Bydgoszcz (PL), Państwowa Wyższa Szkoła Zawodowa w Raciborzu (PL), The Maria Sklodowska-Curie National Research Institute of Oncology (PL), Gdańsk Medical University (PL), University of Rzeszów (PL)
European Commission, Ministerstwo Edukacji i Nauki, Silesian University of Technology, Narodowe Centrum Badań i Rozwoju
Openalex Percentile: Top 48%
COVID-19 diagnosis using AI
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.