Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologists
Objectives - The aim of this study was to determine whether ChatGPT-4 can correctly suggest MRI protocols and additional MRI sequences based on real-world Radiology Request Forms (RRFs) as well as to investigate the ability of ChatGPT-4 to suggest time saving protocols. - Material & methods - Re...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
December 2025
|
| In: |
European journal of radiology
Year: 2025, Volume: 193, Pages: 1-7 |
| ISSN: | 1872-7727 |
| DOI: | 10.1016/j.ejrad.2025.112416 |
| Online Access: | Verlag, kostenfrei, Volltext: https://doi.org/10.1016/j.ejrad.2025.112416 Verlag, kostenfrei, Volltext: https://www.sciencedirect.com/science/article/pii/S0720048X25005029 
|
| Author Notes: | Zeynep Bendella, Barbara Daria Wichtmann, Ralf Clauberg, Vera C. Keil, Nils C. Lehnen, Robert Haase, Laura C. Sáez, Isabella C. Wiest, Jakob Nikolas Kather, Christoph Endler, Alexander Radbruch, Daniel Paech, Katerina Deike |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 194076081X | ||
| 003 | DE-627 | ||
| 005 | 20251120193723.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 251110s2025 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.ejrad.2025.112416 |2 doi | |
| 035 | |a (DE-627)194076081X | ||
| 035 | |a (DE-599)KXP194076081X | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 33 |2 sdnb | ||
| 100 | 1 | |a Bendella, Zeynep |d 1987- |e VerfasserIn |0 (DE-588)1068225300 |0 (DE-627)819808164 |0 (DE-576)42740262X |4 aut | |
| 245 | 1 | 0 | |a Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologists |c Zeynep Bendella, Barbara Daria Wichtmann, Ralf Clauberg, Vera C. Keil, Nils C. Lehnen, Robert Haase, Laura C. Sáez, Isabella C. Wiest, Jakob Nikolas Kather, Christoph Endler, Alexander Radbruch, Daniel Paech, Katerina Deike |
| 264 | 1 | |c December 2025 | |
| 300 | |a 7 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Online verfügbar: 13. September 2025, Artikelversion: 16. September 2025 | ||
| 500 | |a Gesehen am 10.11.2025 | ||
| 520 | |a Objectives - The aim of this study was to determine whether ChatGPT-4 can correctly suggest MRI protocols and additional MRI sequences based on real-world Radiology Request Forms (RRFs) as well as to investigate the ability of ChatGPT-4 to suggest time saving protocols. - Material & methods - Retrospectively, 1,001 RRFs of our Department of Neuroradiology (in-house dataset), 200 RRFs of an independent Department of General Radiology (independent dataset) and 300 RRFs from an external, foreign Department of Neuroradiology (external dataset) were included. Patients’ age, sex, and clinical information were extracted from the RRFs and used to prompt ChatGPT- 4 to choose an adequate MRI protocol from predefined institutional lists. Four independent raters then assessed its performance. Additionally, ChatGPT-4 was tasked with creating case-specific protocols aimed at saving time. - Results - Two and 7 of 1,001 protocol suggestions of ChatGPT-4 were rated “unacceptable” in the in-house dataset for reader 1 and 2, respectively. No protocol suggestions were rated “unacceptable” in both the independent and external dataset. When assessing the inter-reader agreement, Coheńs weighted ĸ ranged from 0.88 to 0.98 (each p < 0.001). ChatGPT-4′s freely composed protocols were approved in 766/1,001 (76.5 %) and 140/300 (46.67 %) cases of the in-house and external dataset with mean time savings (standard deviation) of 3:51 (minutes:seconds) (±2:40) minutes and 2:59 (±3:42) minutes per adopted in-house and external MRI protocol. - Conclusion - ChatGPT-4 demonstrated a very high agreement with board-certified (neuro-)radiologists in selecting MRI protocols and was able to suggest approved time saving protocols from the set of available sequences. | ||
| 650 | 4 | |a ChatGPT-4 | |
| 650 | 4 | |a Large language model (LLM) | |
| 650 | 4 | |a MRI protocol | |
| 650 | 4 | |a Radiology request form | |
| 700 | 1 | |a Wichtmann, Barbara |d 1992- |e VerfasserIn |0 (DE-588)113837461X |0 (DE-627)895889684 |0 (DE-576)492514135 |4 aut | |
| 700 | 1 | |a Clauberg, Ralf |d 1974- |e VerfasserIn |0 (DE-588)129382272 |0 (DE-627)394200373 |0 (DE-576)297632663 |4 aut | |
| 700 | 1 | |a Keil, Vera Catharina Wilma |d 1985- |e VerfasserIn |0 (DE-588)141573228 |0 (DE-627)704014130 |0 (DE-576)324785232 |4 aut | |
| 700 | 1 | |a Lehnen, Nils Christian |e VerfasserIn |0 (DE-588)1068546816 |0 (DE-627)820445215 |0 (DE-576)427853028 |4 aut | |
| 700 | 1 | |a Haase, Robert |d 1983- |e VerfasserIn |0 (DE-588)1103529463 |0 (DE-627)86128559X |0 (DE-576)470658436 |4 aut | |
| 700 | 1 | |a Sáez, Laura C. |e VerfasserIn |4 aut | |
| 700 | 1 | |a Wiest, Isabella |d 1992- |e VerfasserIn |0 (DE-588)1198882956 |0 (DE-627)168103638X |4 aut | |
| 700 | 1 | |a Kather, Jakob Nikolas |d 1989- |e VerfasserIn |0 (DE-588)1064064914 |0 (DE-627)812897587 |0 (DE-576)423589091 |4 aut | |
| 700 | 1 | |a Endler, Christoph Hans-Jürgen |e VerfasserIn |0 (DE-588)1067933727 |0 (DE-627)819366838 |0 (DE-576)427074339 |4 aut | |
| 700 | 1 | |a Radbruch, Alexander |d 1977- |e VerfasserIn |0 (DE-588)1022344501 |0 (DE-627)716953951 |0 (DE-576)362851166 |4 aut | |
| 700 | 1 | |a Paech, Daniel |d 1986- |e VerfasserIn |0 (DE-588)1080278214 |0 (DE-627)844124893 |0 (DE-576)453464742 |4 aut | |
| 700 | 1 | |a Deike-Hofmann, Katerina |d 1989- |e VerfasserIn |0 (DE-588)1121245439 |0 (DE-627)874191173 |0 (DE-576)48078213X |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t European journal of radiology |d Amsterdam [u.a.] : Elsevier Science, 1990 |g 193(2025) vom: Dez., Artikel-ID 112416, Seite 1-7 |h Online-Ressource |w (DE-627)32044483X |w (DE-600)2005350-2 |w (DE-576)099718138 |x 1872-7727 |7 nnas |a Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologists |
| 773 | 1 | 8 | |g volume:193 |g year:2025 |g month:12 |g elocationid:112416 |g pages:1-7 |g extent:7 |a Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologists |
| 856 | 4 | 0 | |u https://doi.org/10.1016/j.ejrad.2025.112416 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 856 | 4 | 0 | |u https://www.sciencedirect.com/science/article/pii/S0720048X25005029 |x Verlag |z kostenfrei |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20251110 | ||
| 993 | |a Article | ||
| 994 | |a 2025 | ||
| 998 | |g 1080278214 |a Paech, Daniel |m 1080278214:Paech, Daniel |d 50000 |e 50000PP1080278214 |k 0/50000/ |p 12 | ||
| 998 | |g 1022344501 |a Radbruch, Alexander |m 1022344501:Radbruch, Alexander |d 50000 |e 50000PR1022344501 |k 0/50000/ |p 11 | ||
| 998 | |g 1064064914 |a Kather, Jakob Nikolas |m 1064064914:Kather, Jakob Nikolas |d 910000 |d 910100 |e 910000PK1064064914 |e 910100PK1064064914 |k 0/910000/ |k 1/910000/910100/ |p 9 | ||
| 998 | |g 1198882956 |a Wiest, Isabella |m 1198882956:Wiest, Isabella |d 60000 |d 61100 |e 60000PW1198882956 |e 61100PW1198882956 |k 0/60000/ |k 1/60000/61100/ |p 8 | ||
| 999 | |a KXP-PPN194076081X |e 4801920063 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"id":{"eki":["194076081X"],"doi":["10.1016/j.ejrad.2025.112416"]},"name":{"displayForm":["Zeynep Bendella, Barbara Daria Wichtmann, Ralf Clauberg, Vera C. Keil, Nils C. Lehnen, Robert Haase, Laura C. Sáez, Isabella C. Wiest, Jakob Nikolas Kather, Christoph Endler, Alexander Radbruch, Daniel Paech, Katerina Deike"]},"physDesc":[{"extent":"7 S."}],"recId":"194076081X","note":["Online verfügbar: 13. September 2025, Artikelversion: 16. September 2025","Gesehen am 10.11.2025"],"language":["eng"],"type":{"bibl":"article-journal","media":"Online-Ressource"},"title":[{"title":"Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologists","title_sort":"Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologists"}],"relHost":[{"disp":"Chat GPT-4 shows high agreement in MRI protocol selection compared to board-certified neuroradiologistsEuropean journal of radiology","recId":"32044483X","physDesc":[{"extent":"Online-Ressource"}],"title":[{"subtitle":"EJR","title":"European journal of radiology","title_sort":"European journal of radiology"}],"note":["Gesehen am 05.02.20"],"type":{"bibl":"periodical","media":"Online-Ressource"},"language":["eng"],"origin":[{"publisher":"Elsevier Science","publisherPlace":"Amsterdam [u.a.]","dateIssuedKey":"1990","dateIssuedDisp":"1990-"}],"part":{"pages":"1-7","year":"2025","text":"193(2025) vom: Dez., Artikel-ID 112416, Seite 1-7","volume":"193","extent":"7"},"id":{"issn":["1872-7727"],"eki":["32044483X"],"zdb":["2005350-2"]},"titleAlt":[{"title":"EJR"}],"pubHistory":["Nachgewiesen 10.1990 -"]}],"person":[{"display":"Bendella, Zeynep","family":"Bendella","given":"Zeynep","role":"aut"},{"role":"aut","given":"Barbara","family":"Wichtmann","display":"Wichtmann, Barbara"},{"display":"Clauberg, Ralf","family":"Clauberg","given":"Ralf","role":"aut"},{"display":"Keil, Vera Catharina Wilma","family":"Keil","given":"Vera Catharina Wilma","role":"aut"},{"role":"aut","given":"Nils Christian","family":"Lehnen","display":"Lehnen, Nils Christian"},{"given":"Robert","role":"aut","family":"Haase","display":"Haase, Robert"},{"family":"Sáez","given":"Laura C.","role":"aut","display":"Sáez, Laura C."},{"display":"Wiest, Isabella","given":"Isabella","role":"aut","family":"Wiest"},{"display":"Kather, Jakob Nikolas","role":"aut","given":"Jakob Nikolas","family":"Kather"},{"display":"Endler, Christoph Hans-Jürgen","family":"Endler","given":"Christoph Hans-Jürgen","role":"aut"},{"role":"aut","given":"Alexander","family":"Radbruch","display":"Radbruch, Alexander"},{"display":"Paech, Daniel","family":"Paech","given":"Daniel","role":"aut"},{"family":"Deike-Hofmann","role":"aut","given":"Katerina","display":"Deike-Hofmann, Katerina"}],"origin":[{"dateIssuedDisp":"December 2025","dateIssuedKey":"2025"}]} | ||
| SRT | |a BENDELLAZECHATGPT4SH2025 | ||