Medical Science

  • Home

Volume 29, Issue 163, September 2025

Artificial Intelligence in Emergency Radiology - A Review

Maria Jasiewicz1♦, Anita Warzocha2, Anita Krowiak3, Piotr Pitrus3, Karolina Krowiak3, Wiktoria Hander4, Gabriela Majka4, Aleksandra Karnas4, Karol Bednarz3, Magdalena Kowalczyk4

1Medical Center in Lancut, ul. Ignacego Paderewskiego 5, 37-100 Lancut, Poland
2University Clinical Hospital F. Chopin in Rzeszów, Poland
3Clinical Provincial Hospital No. 2 in Rzeszow, Poland Lwowska 60, 35-301 Rzeszów, Poland
4College of Medical Sciences, University of Rzeszów Tadeusza Rejtana 16 C Ave, 35-310 Rzeszów, Poland

♦Corresponding author
Maria Jasiewicz, Medical Center in Lancut, ul. Ignacego Paderewskiego 5, 37-100 Lancut, Poland

ABSTRACT

This article discusses the growing role of artificial intelligence (AI) in radiology, with a particular focus on its application in emergency departments. Due to the rise in imaging tests, which often involve medical emergencies, radiologists are experiencing an increase in workload. That is why artificial intelligence has great potential in developing new algorithms based on different machine learning methods. Recent clinical studies show that artificial intelligence can match, and sometimes even surpass, human specialists in detecting conditions such as pulmonary embolism, stroke, fractures, and small bowel obstruction. Despite promising research results, we have to take into account the irregularities that AI may exhibit, such as regulations concerning data privacy, bias in AI training, and the lack of transparency in how it makes decisions, known as the "black box" problem. Further research should focus on preparing AI protocols with medical professionals and algorithm programmers. Researchers should carry the work forward to validate the sample volume and its diversity.

Keywords: artificial intelligence, radiology, emergency diagnostic, machine learning

Medical Science, 2025, 29, e179ms3688
PDF
DOI: https://doi.org/10.54905/disssi.v29i163.e179ms3688

Published: 25 September 2025

Creative Commons License

© The Author(s) 2025. Open Access. This article is licensed under a Creative Commons Attribution License 4.0 (CC BY 4.0).