Optimizing GPT-4 Turbo Diagnostic Accuracy in Neuroradiology through Prompt Engineering and Confidence Thresholds
Open accessAbstract
Background and Objectives: Integrating large language models (LLMs) such as GPT-4 Turbo into diagnostic imaging faces a significant challenge, with current misdiagnosis rates ranging from 30–50%. This study evaluates how prompt engineering and confidence thresholds can improve diagnostic accuracy in neuroradiology. Methods: We analyze 751 neuroradiology cases from the American Journal of Neuroradiology using GPT-4 Turbo with customized prompts to improve diagnostic precision. Results: Initially, GPT-4 Turbo achieved a baseline diagnostic accuracy of 55.1%. By reformatting responses to list five diagnostic candidates and applying a 90% confidence threshold, the highest precision of the diagnosis increased to 72.9%, with the candidate list providing the correct diagnosis at 85.9%, reducing the misdiagnosis rate to 14.1%. However, this threshold reduced the number of cases that responded. Conclusions: Strategic prompt engineering and high confidence thresholds significantly reduce misdiagnoses and improve the precision of the LLM diagnostic in neuroradiology. More research is needed to optimize these approaches for broader clinical implementation, balancing accuracy and utility.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Chain-of-thought prompting has been used to decompose complex diagnostic tasks into linear steps, which aids in integrating context and delivering comprehensive diagnostic reasoning.
“Chain-of-thought (CoT) prompting has been used to decompose complex diagnostic tasks into linear steps, which aids in integrating context and delivering comprehensive diagnostic reasoning33.”
From Enhancing diagnostic capability with multi-agents conversational large language models (2025) · cites this paper as [33]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
ChatGPT and other large language models are double-edged swords
Shen, Y., Heacock, L., Elias, J., Hentel, K. D., Reig, B., Shih, G., & Moy, L. (2023) · Radiology · doi:10.1148/radiol.230163 · Cited by 1,008 · see citing works
Self-citationChatGPT and the future of medical writing
Biswas, S. (2023) · Radiology · doi:10.1148/radiol.223312 · Cited by 571 · see citing works
Antaki, F., Touma, S., Milad, D., El-Khoury, J., & Duval, R. (2023) · doi:10.1101/2023.01.22.23284882 · Cited by 563 · see citing works
Evaluating the Performance of ChatGPT in Ophthalmology
Antaki, F., Touma, S., Milad, D., El-Khoury, J., & Duval, R. (2023) · Ophthalmology Science · doi:10.1016/j.xops.2023.100324 · Cited by 563 · see citing works
ChatGPT makes medicine easy to swallow: an exploratory case study on simplified radiology reports
Jeblick, K., Schachtner, B., Dexl, J., Mittermeier, A., Stüber, A. T., Topalis, J., Weber, T., Wesp, P., Sabel, B. O., Ricke, J., et al. (2023) · European Radiology · doi:10.1007/s00330-023-10213-1 · Cited by 549 · see citing works
Bhayana, R., Krishna, S., & Bleakney, R. R. (2023) · Radiology · doi:10.1148/radiol.230582 · Cited by 460 · see citing works
Confident Adaptive Language ModelingPreprint
Schuster, T., Fisch, A., Gupta, J., et al. (2022) · Neural Information Processing Systems · doi:10.48550/arxiv.2207.07061 · Cited by 419 · see citing works
Hirosawa, T., Harada, Y., Yokose, M., Sakamoto, T., Kawamura, R., & Shimizu, T. (2023) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph20043378 · Cited by 344 · see citing works
Bhayana, R. (2024) · Radiology · doi:10.1148/radiol.232756 · Cited by 340 · see citing works
Potential of ChatGPT and GPT-4 for Data Mining of Free-Text CT Reports on Lung Cancer
Fink, M. A., Bischoff, A., Fink, C. A., Moll, M., Kroschke, J., Dulz, L., Heußel, C. P., Kauczor, H., & Weber, T. F. (2023) · Radiology · doi:10.1148/radiol.231362 · Cited by 278 · see citing works
Savage, T., Nayak, A., Gallo, R., Rangan, E., & Chen, J. H. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01010-1 · Cited by 265 · see citing works
Appropriateness of breast cancer prevention and screening recommendations provided by ChatGPT
Haver, H. L., Ambinder, E. B., Bahl, M., Oluyemi, E. T., Jeudy, J., & Yi, P. H. (2023) · Radiology · doi:10.1148/radiol.230424 · Cited by 247 · see citing works
Large Language Models Can Self-Improve
Huang, J., Gu, S., Hou, L., Wu, Y., Wang, X., Yu, H., & Han, J. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.67 · Cited by 230 · see citing works
Large Language Models Can Self-Improve
Huang, J., Gu, S. S., Hou, L., Wu, Y., Wang, X., Yu, H., & Han, J. (2022) · arXiv · doi:10.48550/arxiv.2210.11610 · Cited by 230 · see citing works
Progressive-Hint Prompting Improves Reasoning in Large Language ModelsPreprint
Zheng, C., Liu, Z.-Y., Xie, E.-Z., Li, Z.-G., & Li, Y. (2023) · arXiv.org · doi:10.48550/arxiv.2304.09797 · Cited by 165 · see citing works
Prompt Engineering a Prompt EngineerPreprint
Ye, Q., Axmed, M., Pryzant, R., & Khani, F. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2311.05661 · Cited by 154 · see citing works
Ueda, D., Mitsuyama, Y., Takita, H., Horiuchi, D., Walston, S. L., Tatekawa, H., & Miki, Y. (2023) · Radiology · doi:10.1148/radiol.231040 · Cited by 139 · see citing works
Gertz, R. J., Bunck, A. C., Lennartz, S., Dratsch, T., Iuga, A.-I., Maintz, D. C., & Kottlors, J. (2023) · Radiology · doi:10.1148/radiol.230877 · Cited by 134 · see citing works
GPT-4 in Radiology: Improvements in Advanced Reasoning
Bhayana, R., Bleakney, R. R., & Krishna, S. (2023) · Radiology · doi:10.1148/radiol.230987 · Cited by 127 · see citing works
Horiuchi, D., Tatekawa, H., Shimono, T., Walston, S. L., Takita, H., Matsushita, S., Oura, T., Mitsuyama, Y., Miki, Y., & Ueda, D. (2023) · Neuroradiology · doi:10.1007/s00234-023-03252-4 · Cited by 112 · see citing works
Feasibility of Differential Diagnosis Based on Imaging Patterns Using a Large Language Model
Kottlors, J., Bratke, G., Rauen, P., Kabbasch, C., ten Persigehl, T., Schlamann, M., & Lennartz, S. (2023) · Radiology · doi:10.1148/radiol.231167 · Cited by 101 · see citing works
Self‐Monitoring in Social Interaction: The Centrality of Self‐Affect
Ickes, W. J., Holloway, R. A., Stinson, L. L., & Graham, T. L. (2006) · Journal of Personality · doi:10.1111/j.1467-6494.2006.00388.x · Cited by 72 · see citing works
Suthar, P. P., Kounsal, A., Chhetri, L., Saini, D., & Dua, S. G. (2023) · Cureus · doi:10.7759/cureus.43958 · Cited by 58 · see citing works
Lallas, A., Lallas, K., Tschandl, P., Kittler, H., Apalla, Z., Longo, C., & Argenziano, G. (2020) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2020.06.085 · Cited by 40 · see citing works
Yang, Z., Yao, Z., Tasmin, M., Vashisht, P., Jang, W. S., Ouyang, F., Wang, B., Berlowitz, D. R., & Yu, H. (2023) · medRxiv · doi:10.1101/2023.10.26.23297629 · Cited by 37 · see citing works
Benchmarking the symptom-checking capabilities of ChatGPT for a broad range of diseases
Chen, A., Chen, D. O., & Tian, L. (2023) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocad245 · Cited by 25 · see citing works
Algorithm Learning and Problem Solving
Scandura, J. M. (1966) · The Journal of Experimental Education · doi:10.1080/00220973.1966.11010945 · Cited by 20 · see citing works
Engineering neural systems for high-level problem solving
Sylvester, J., & Reggia, J. A. (2016) · Neural Networks · doi:10.1016/j.neunet.2016.03.006 · Cited by 18 · see citing works
Jiang, S., Xu, Y., & Lü, X. (2023) · Radiology · doi:10.1148/radiol.231335 · Cited by 10 · see citing works
Strength in Numbers: Estimating Confidence of Large Language Models by Prompt Agreement
Portillo-Wightman, G., DeLucia, A., & Dredze, M. H. (2023) · Workshop on Trustworthy Natural Language Processing (TrustNLP) · doi:10.18653/v1/2023.trustnlp-1.28 · Cited by 10 · see citing works
Yang, Y., Li, H., Wang, Y., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2310.04782 · Cited by 2 · see citing works
Flagged first, then by citations
Research works printed but not linked: 3
- [21]Case of the Week Diagnoses|American Journal of Neuroradiology Available online: https://www.ajnr.org/cow/by/diagnosis (accessed on 18 March 2024)
- [23]Nori H. King N. McKinney S.M. Carignan D. Horvitz E. Capabilities of GPT-4 on Medical Challenge Problems arXiv 2023 2303.13375
- [34]Zhao T. Wei M. Preston J.S. Poon H. Pareto Optimal Learning for Estimating Large Language Model Errors arXiv 2023 10.48550/arxiv.2306.16564 · DOI
Cited by
Sorted by citations
Takita, H., Kabata, D., Walston, S. L., Tatekawa, H., Saito, K., Tsujimoto, Y., Miki, Y., & Ueda, D. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01543-z · Cited by 203 · see citing works
Large language models for disease diagnosis: a scoping review
Zhou, S., Xu, Z., Zhang, M., Xu, C., Guo, Y., Zhan, Z., Fang, Y., Ding, S., Wang, J., Xu, K., et al. (2025) · npj Artificial Intelligence · doi:10.1038/s44387-025-00011-z · Cited by 143 · see citing works
Enhancing diagnostic capability with multi-agents conversational large language models
Chen, X., Yi, H., You, M., Liu, W., Wang, L., Li, H., Zhang, X., Guo, Y., Fan, L., Chen, G., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01550-0 · Cited by 117 · see citing works
Large Language Models in Medical Diagnostics: Scoping Review With Bibliometric Analysis
Su, H., Sun, Y., Li, R., Zhang, A., Yang, Y., Xiao, F., Duan, Z., Chen, J., Hu, Q., Yang, T., et al. (2025) · Journal of Medical Internet Research · doi:10.2196/72062 · Cited by 66 · see citing works
Retrieval-augmented generation elevates local LLM quality in radiology contrast media consultation
Wada, A., Tanaka, Y., Nishizawa, M., Yamamoto, A., Akashi, T., Hagiwara, A., Hayakawa, Y. K., Kikuta, J., Shimoji, K., Sano, K., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01802-z · Cited by 58 · see citing works
Self-citationSonoda, Y., Kurokawa, R., Hagiwara, A., Asari, Y., Fukushima, T., Kanzawa, J., Gonoi, W., & Abe, O. (2024) · Japanese Journal of Radiology · doi:10.1007/s11604-024-01712-2 · Cited by 32 · see citing works
Self-citationLarge language models as a diagnostic support tool in neuropathology
Hewitt, K. J., Wiest, I. C., Carrero, Z. I., Bejan, L., Millner, T. O., Brandner, S., & Kather, J. N. (2024) · The Journal of Pathology: Clinical Research · doi:10.1002/2056-4538.70009 · Cited by 27 · see citing works
Accuracy Evaluation of GPT-Assisted Differential Diagnosis in Emergency Department
Shah-Mohammadi, F., & Finkelstein, J. (2024) · Diagnostics · doi:10.3390/diagnostics14161779 · Cited by 22 · see citing works
Bentegeac, R., Guellec, B. L., Kuchcinski, G., Amouyel, P., & Hamroun, A. (2025) · Journal of Medical Internet Research · doi:10.2196/64348 · Cited by 19 · see citing works
Haque, M. A., & Siddique, H. R. (2025) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2025.108611 · Cited by 17 · see citing works
Sato, H., Ogasawara, K., & Sakurai, H. (2025) · JMIR Medical Education · doi:10.2196/76925 · Cited by 12 · see citing works
Gaebe, K., & van der Woerd, B. D. (2025) · PLoS ONE · doi:10.1371/journal.pone.0325803 · Cited by 12 · see citing works
Same Prompt, Different Answer: Why AI Disagrees with Itself
Mee, T., Choi, J. J., & Baduashvili, A. (2025) · Journal of General Internal Medicine · doi:10.1007/s11606-025-10071-1 · Cited by 7 · see citing works
Chetla, N., Samayamanthula, S., Chang, J., Leigh, A., Akosman, S., Tandon, M., Hage, T., & Cusick, M. (2025) · Clinical Ophthalmology · doi:10.2147/opth.s517238 · Cited by 5 · see citing works
Japanese Radiology 2025 Updates
Iima, M., Saida, T., Yamada, Y., Kurokawa, R., Ueda, D., Honda, M., Nishioka, K., Ito, R., Sofue, K., & Naganawa, S. (2025) · Canadian Association of Radiologists Journal · doi:10.1177/08465371251374557 · Cited by 3 · see citing works
AI in Medical Imaging and Image Processing
Nurzyǹska, K., Strzelecki, M., Piórkowski, A., & Obuchowicz, R. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14124153 · Cited by 3 · see citing works
Nguyen, D., Bronson, I., Chen, R., & Kim, Y. H. (2025) · Frontiers in Radiology · doi:10.3389/fradi.2025.1670517 · Cited by 2 · see citing works
Song, C. W., Baek, B. H., Kim, S. K., Yoon, W., Lee, Y. Y., Park, I., Park, J. H., Park, S. B., & Choi, I. W. (2026) · Journal of Clinical Medicine · doi:10.3390/jcm15041673 · Cited by 2 · see citing works
Context-Guided Mixture-of-Experts with Multimodal LLMs for Lesion Detection in Buccal Mucosa Images
Nayak, A., Diddigi, R. B., Kestur, R., Kumar, P. M., Ranganathan, K., Lavanya, C., Rajeshwari, S., & Behera, S. S. (2026) · International Dental Journal · doi:10.1016/j.identj.2026.109575 · Cited by 1 · see citing works
Application of large language models in medical diagnosis: A bibliometric review
Zhang, Q., Wang, H., Li, H., Jiao, F., Zhou, H., & Li, M. (2026) · Digital Health · doi:10.1177/20552076261462624 · Cited by 1 · see citing works
Confidence Calibration for Multimodal LLMs: An Empirical Study Through Medical VQA
Du, Y., Wang, Y., Kong, M. C., Liang, T., Long, Q., Chen, B., & Zhu, Q. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-04978-0_9
Siddiqui, E., & Kumari, S. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-45875-0.00011-9
Paulino, D., Netto, A. T. C., Jardim, R. R.-A. J., Rocha, A., & Paredes, H. (2026) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2026.2690095
Schiller, D., Pristoupil, J., Polaskova, P., Šterclová, M., & Lambert, L. (2026) · European Radiology Experimental · doi:10.1186/s41747-026-00790-3
Generative AI versus physicians in diagnostic radiology: a systematic review and meta-analysis
Takita, H., Kabata, D., Walston, S. L., Mitsuyama, Y., Tatekawa, H., Saito, K., Miki, Y., & Ueda, D. (2026) · Japanese Journal of Radiology · doi:10.1007/s11604-026-02090-7
- Authors
- Wada, A., Akashi, T., Shih, G., Hagiwara, A., Nishizawa, M., Hayakawa, Y., Kikuta, J., Shimoji, K., Sano, K., Kamagata, K., Nakanishi, A., & Aoki, S.
- Affiliations
- Juntendo University (JP)
- Weill Cornell Medicine (US)
- Corresponding author
- Wada, Akihiko
- ORCID
- Venue
- Diagnostics
- Publisher
- MDPI AG
- Published
- 2024-07-17
- Volume
- 14
- Issue
- 14
- Pages
- 1541
- ISSN
- 2075-4418 (ISSN-L)
- PMC
- PMC11276551
- OpenAlex
- W4400727806
- Cited by
- 25
- References
- 26
- Citations by year
- 2026: 7 · 2025: 15 · 2024: 3
- FWCI
- 2.34
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Diagnostics (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Funders
- Japan Society for the Promotion of Science (JSPS) KAKENHI (22K07674)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiology practices and education
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Keywords
- GPT-4 Turbo, large language models, neuroradiology, prompt engineering, confidence threshold, diagnostic accuracy, misdiagnosis, AI-assisted diagnosis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
