Evidence map›Paper›PMID 41262497›Full record

ArticlePolish journal of radiology2025

Advancing AI in radiology: a comparative analysis of ChatGPT-o1-preview and ChatGPT-3.5 in the Polish National Specialization Exam.

Adam Mitręga, Michał Bielówka, Dominika Kaczyńska, Natalia Denisiewicz, Mikołaj Magiera, Marcin Rojek, Maja Dreger, Jakub Kufel, Miłosz Zbroszczyk

Abstract read
PubMed Full text at PMC Publisher
In one paragraph

Article in Polish journal of radiology, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.

0numbers the graph read from it
0cells of the map it votes in
0citing papers in PubMed
–field-weighted citation impact
1 · What the graph read from it

What it found

Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.

The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.

2 · The registry

The trial behind it

Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.

Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.

3 · Its place in the literature

Who cites it

0 citing papers in PubMed.

No citing paper in PubMed yet.

4 · The record

Corrections and comments

PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.

5 · Who and what money

Authors and funding

9 authors.

Adam Mitręga *Students' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Michał Bielówka *Students' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Dominika KaczyńskaStudents' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Natalia DenisiewiczStudents' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Mikołaj MagieraStudents' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Marcin RojekStudents' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Maja DregerStudents' Scientific Association of Computer Analysis and Artificial Intelligence at the Department of Radiology and Nuclear Medicine of the Medical University of Silesia in Katowice, Poland.
Jakub KufelDepartment of Radiodiagnostics, Interventional Radiology and Nuclear Medicine, Medical University of Silesia, Katowice, Poland.
Miłosz ZbroszczykDepartment of Radiodiagnostics, Interventional Radiology and Nuclear Medicine, Medical University of Silesia, Katowice, Poland.

Funding

No grant is acknowledged in the PubMed record.

6 · The paper itself

Abstract

Purpose: The aim of this study was to evaluate the performance of the ChatGPT-o1-preview language model in solving the Polish National Specialization Exam (PES) in radiology and imaging diagnostics and compare its results with previous versions of the model. Material and methods: A set of 119 valid radiology exam questions from Spring 2023 was analyzed. Each question was classified by type, subtype, and clinical relevance. ChatGPT answered each question five times using a standardized prompt with a 5-point confidence scale. Performance was assessed using accuracy and declared and calculated difficulty indices. Statistical analysis was performed in Python with a significance level of Results: The model achieved a score of 93.33% correct answers, comparable to the average physician score of 94.86%. ChatGPT-o1-preview showed exceptional accuracy in "memory" questions, with over 96% correct answers. This result, significantly higher than that of the older ChatGPT-3.5 model (52%), demonstrates progress in artificial intelligence (AI) capabilities. The model also exhibited higher confidence in its responses, indicating better adaptation to medical exams. Conclusions: Despite its high accuracy, the study was based on a relatively small set of questions, which limits the ability to fully assess the model's effectiveness. The results indicate the potential of AI as a tool to support clinical work, but further, more extensive research is necessary to evaluate its applicability and reliability in the medical environment.

Indexed as

AIartificial intelligenceChatGPTmedical imagingradiology

Identifiers

PMID41262497
PMCPMC12624502

What OpenQuestion holds

Textmetadata
LicenceCC BY-NC-ND
Read underepoch 390

Registered trials

None linked

Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.