Abstract
Objective
Radiology reporting is an essential component of clinical diagnosis and decision-making. With the advent of advanced artificial intelligence (AI) models like GPT-4 (Generative Pre-trained Transformer 4), there is growing interest in evaluating their potential for optimizing or generating radiology reports. This study aimed to compare the quality and content of radiologist-generated and GPT-4 AI-generated radiology reports.
Methods
A comparative study design was employed in the study, where a total of 100 anonymized radiology reports were randomly selected and analyzed. Each report was processed by GPT-4, resulting in the generation of a corresponding AI-generated report. Quantitative and qualitative analysis techniques were utilized to assess similarities and differences between the two sets of reports.
Results
The AI-generated reports showed comparable quality to radiologist-generated reports in most categories. Significant differences were observed in clarity (p = 0.027), ease of understanding (p = 0.023), and structure (p = 0.050), favoring the AI-generated reports. AI-generated reports were more concise, with 34.53 fewer words and 174.22 fewer characters on average, but had greater variability in sentence length. Content similarity was high, with an average Cosine Similarity of 0.85, Sequence Matcher Similarity of 0.52, BLEU Score of 0.5008, and BERTScore F1 of 0.8775.
Conclusion
The results of this proof-of-concept study suggest that GPT-4 can be a reliable tool for generating standardized radiology reports, offering potential benefits such as improved efficiency, better communication, and simplified data extraction and analysis. However, limitations and ethical implications must be addressed to ensure the safe and effective implementation of this technology in clinical practice.
Clinical relevance statement
The findings of this study suggest that GPT-4 (Generative Pre-trained Transformer 4), an advanced AI model, has the potential to significantly contribute to the standardization and optimization of radiology reporting, offering improved efficiency and communication in clinical practice.
Key Points
• Large language model–generated radiology reports exhibited high content similarity and moderate structural resemblance to radiologist-generated reports.
• Performance metrics highlighted the strong matching of word selection and order, as well as high semantic similarity between AI and radiologist-generated reports.
• Large language model demonstrated potential for generating standardized radiology reports, improving efficiency and communication in clinical settings.
Similar content being viewed by others
Abbreviations
- AI:
-
Artificial intelligence
- BERT:
-
Bidirectional Encoder Representations from Transformers
- BLEU:
-
Bilingual Evaluation Understudy
- CT:
-
Computed tomography
- GPT-4:
-
Generative Pre-trained Transformer 4
- LLMs:
-
Large language models
- MRI:
-
Magnetic resonance imaging
- NLG:
-
Natural language generation
- NLP:
-
Natural language processing
- PACS:
-
Picture Archiving and Communication Systems
- SDV:
-
Standard deviation
- TF-IDF:
-
Term Frequency–Inverse Document Frequency
References
Srinivasa Babu A, Brooks ML (2015) The malpractice liability of radiology reports: minimizing the risk. Radiographics 35:547–554
Larson DB (2018) Strategies for implementing a standardized structured radiology reporting program. Radiographics 38:1705–1716
Adams LC, Truhn D, Busch F et al (2023) Leveraging GPT-4 for post hoc transformation of free-text radiology reports into structured reporting: a multilingual feasibility study. Radiology. https://doi.org/10.1148/radiol.230725:230725
Jeblick K, Schachtner B, Dexl J et al (2023) ChatGPT makes medicine easy to swallow: an exploratory case study on simplified radiology reports. Eur Radiol. https://doi.org/10.1007/s00330-023-10213-1
Gaube S, Suresh H, Raue M et al (2021) Do as AI say: susceptibility in deployment of clinical decision-aids. NPJ Digit Med 4:31
Rao A, Kim J, Kamineni M, Pang M, Lie W, Succi MD (2023) Evaluating ChatGPT as an adjunct for radiologic decision-making. medRxiv. https://doi.org/10.1101/2023.02.02.23285399
Choudhury A, Asan O (2020) Role of artificial intelligence in patient safety outcomes: systematic literature review. JMIR Med Inform 8:e18599
Aggarwal R, Sounderajah V, Martin G et al (2021) Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis. NPJ Digit Med 4:65
Alfarghaly O, Khaled R, Elkorany A, Helal M, Fahmy A (2021) Automated radiology report generation using conditioned transformers. Inform Med Unlocked 24:100557
Monshi MMA, Poon J, Chung V (2020) Deep learning in generating radiology reports: a survey. Artif Intell Med 106:101878
Wiggins WF, Kitamura F, Santos I, Prevedello LM (2021) Natural language processing of radiology text reports: interactive text classification. Radiol Artif Intell 3:e210035
Lee P, Bubeck S, Petro J (2023) Benefits, limits, and risks of GPT-4 as an AI chatbot for medicine. N Engl J Med 388:1233–1239
Mabotuwana T, Lee MC, Cohen-Solal EV (2013) An ontology-based similarity measure for biomedical data – application to radiology reports. J Biomed Inform 46:857–868
Lyu Q, Tan J, Zapadka ME et al (2023) Translating radiology reports into plain language using ChatGPT and GPT-4 with prompt learning: promising results, limitations, and potential. https://doi.org/10.48550/arXiv.2303.09038
Jing B, Xie P, Xing EP (2017) On the automatic generation of medical imaging reports Annual meeting of the Association for Computational Linguistics
Tejani AS, Ng YS, Xi Y, Fielding JR, Browning TG, Rayan JC (2022) Performance of multiple pretrained BERT models to automate and accelerate data annotation for large datasets. Radiol Artif Intell 4:e220007
Yan A, McAuley J, Lu X et al (2022) RadBERT: adapting transformer-based language models to radiology. Radiol Artif Intell 4:e210258
Li J, Lin Y, Zhao P et al (2022) Automatic text classification of actionable radiology reports of tinnitus patients using bidirectional encoder representations from transformer (BERT) and in-domain pre-training (IDPT). BMC Med Inform Decis Mak 22:200
Nishigaki D, Suzuki Y, Wataya T et al (2023) BERT-based transfer learning in sentence-level anatomic classification of free-text radiology reports. Radiol Artif Intell 5:e220097
Olthof AW, Shouche P, Fennema EM et al (2021) Machine learning based natural language processing of radiology reports in orthopaedic trauma. Comput Methods Programs Biomed 208:106304
OpenAI (2023) GPT-4 Technical Report. Arxiv abs/2303.08774
Alkaissi H, McFarlane SI (2023) Artificial hallucinations in ChatGPT: implications in scientific writing. Cureus 15:e35179
Li J, Cheng X, Zhao WX, Nie J-Y, Wen J-R (2023) HELMA: a large-scale hallucination evaluation benchmark for large language models. arXiv preprint arXiv:230511747
Funding
This study has received funding by National Institutes of Health.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Guarantor
The scientific guarantor of this publication is Ashkan Malayeri.
Conflict of interest
The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
Statistics and biometry
No complex statistical methods were necessary for this paper.
Informed consent
Written informed consent was not required for this study because radiology reports were anonymized.
Ethical approval
The National Institutes of Health’s Institutional Review Board (IRB) approved the protocol as a retrospective study.
Study subjects or cohorts overlap
N/A.
Methodology
• Comparative quantitative and qualitative analysis
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Hasani, A.M., Singh, S., Zahergivar, A. et al. Evaluating the performance of Generative Pre-trained Transformer-4 (GPT-4) in standardizing radiology reports. Eur Radiol (2023). https://doi.org/10.1007/s00330-023-10384-x
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00330-023-10384-x