Evaluation of a retrieval-augmented generation system using a Japanese Institutional Nuclear Medicine Manual and large language model-automated scoring

Recent advances in large language models (LLMs) enable domain-specific question answering using external knowledge. However, addressing information that is not included in training data remains a challenge, particularly in nuclear medicine, where examination protocols are frequently updated and vary...

Full description

Saved in:

Bibliographic Details
Published in:	Radiological physics and technology Vol. 18; no. 3; pp. 861 - 876
Main Authors:	Fukui, Yusuke, Kawata, Yuhei, Kobashi, Kazumasa, Nagatani, Yukihiro, Iguchi, Harumi
Format:	Journal Article
Language:	English
Published:	Singapore Springer Nature Singapore 01.09.2025 Springer Nature B.V
Subjects:	Application programming interface Artificial intelligence Automation Certification Datasets Documents Drug dosages Humans Imaging Information Storage and Retrieval - methods Japan Keywords Kidneys Knowledge Language Large Language Models Manuals Medical and Radiation Physics Medicine Medicine & Public Health Nuclear Medicine Performance evaluation Questions Radiology Radiotherapy Research Article Retrieval Scintigraphy Japan Nuclear medicine Medical question answering Large language models (LLMs) Retrieval-augmented generation (RAG)
ISSN:	1865-0333, 1865-0341, 1865-0341
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Recent advances in large language models (LLMs) enable domain-specific question answering using external knowledge. However, addressing information that is not included in training data remains a challenge, particularly in nuclear medicine, where examination protocols are frequently updated and vary across institutions. In this study, we developed a retrieval-augmented generation (RAG) system using 40 internal manuals from a single Japanese hospital, each corresponding to a different examination in nuclear medicine. These institution-specific documents were segmented and indexed using a hybrid retrieval strategy combining dense vector search (text-embedding-3-small) and sparse keyword search (BM25). GPT-3.5 and GPT-4o were used with the OpenAI application programming interface (API) for response generation. The quality of the generated answers was assessed using a four-point Likert scale by three certified radiological technologists, of which one held an additional certification in nuclear medicine and another held an additional certification in medical physics. Automated evaluation was conducted using RAGAS metrics, including factual correctness and context recall. The GPT-4o model combined with hybrid retrieval achieved the highest performance, as per expert evaluations. Although traditional string-based metrics such as ROUGE and the Levenshtein distance poorly align with human ratings, RAGAS provided consistent rankings across system configurations, despite showing only a modest correlation with manual scores. These findings demonstrate that integrating examination-specific institutional manuals into RAG frameworks can effectively support domain-specific question answering in nuclear medicine. Moreover, LLM-based evaluation methods such as RAGAS may serve as practical tools to complement expert reviews in developing healthcare-oriented artificial intelligence systems.
AbstractList	Recent advances in large language models (LLMs) enable domain-specific question answering using external knowledge. However, addressing information that is not included in training data remains a challenge, particularly in nuclear medicine, where examination protocols are frequently updated and vary across institutions. In this study, we developed a retrieval-augmented generation (RAG) system using 40 internal manuals from a single Japanese hospital, each corresponding to a different examination in nuclear medicine. These institution-specific documents were segmented and indexed using a hybrid retrieval strategy combining dense vector search (text-embedding-3-small) and sparse keyword search (BM25). GPT-3.5 and GPT-4o were used with the OpenAI application programming interface (API) for response generation. The quality of the generated answers was assessed using a four-point Likert scale by three certified radiological technologists, of which one held an additional certification in nuclear medicine and another held an additional certification in medical physics. Automated evaluation was conducted using RAGAS metrics, including factual correctness and context recall. The GPT-4o model combined with hybrid retrieval achieved the highest performance, as per expert evaluations. Although traditional string-based metrics such as ROUGE and the Levenshtein distance poorly align with human ratings, RAGAS provided consistent rankings across system configurations, despite showing only a modest correlation with manual scores. These findings demonstrate that integrating examination-specific institutional manuals into RAG frameworks can effectively support domain-specific question answering in nuclear medicine. Moreover, LLM-based evaluation methods such as RAGAS may serve as practical tools to complement expert reviews in developing healthcare-oriented artificial intelligence systems. Recent advances in large language models (LLMs) enable domain-specific question answering using external knowledge. However, addressing information that is not included in training data remains a challenge, particularly in nuclear medicine, where examination protocols are frequently updated and vary across institutions. In this study, we developed a retrieval-augmented generation (RAG) system using 40 internal manuals from a single Japanese hospital, each corresponding to a different examination in nuclear medicine. These institution-specific documents were segmented and indexed using a hybrid retrieval strategy combining dense vector search (text-embedding-3-small) and sparse keyword search (BM25). GPT-3.5 and GPT-4o were used with the OpenAI application programming interface (API) for response generation. The quality of the generated answers was assessed using a four-point Likert scale by three certified radiological technologists, of which one held an additional certification in nuclear medicine and another held an additional certification in medical physics. Automated evaluation was conducted using RAGAS metrics, including factual correctness and context recall. The GPT-4o model combined with hybrid retrieval achieved the highest performance, as per expert evaluations. Although traditional string-based metrics such as ROUGE and the Levenshtein distance poorly align with human ratings, RAGAS provided consistent rankings across system configurations, despite showing only a modest correlation with manual scores. These findings demonstrate that integrating examination-specific institutional manuals into RAG frameworks can effectively support domain-specific question answering in nuclear medicine. Moreover, LLM-based evaluation methods such as RAGAS may serve as practical tools to complement expert reviews in developing healthcare-oriented artificial intelligence systems.Recent advances in large language models (LLMs) enable domain-specific question answering using external knowledge. However, addressing information that is not included in training data remains a challenge, particularly in nuclear medicine, where examination protocols are frequently updated and vary across institutions. In this study, we developed a retrieval-augmented generation (RAG) system using 40 internal manuals from a single Japanese hospital, each corresponding to a different examination in nuclear medicine. These institution-specific documents were segmented and indexed using a hybrid retrieval strategy combining dense vector search (text-embedding-3-small) and sparse keyword search (BM25). GPT-3.5 and GPT-4o were used with the OpenAI application programming interface (API) for response generation. The quality of the generated answers was assessed using a four-point Likert scale by three certified radiological technologists, of which one held an additional certification in nuclear medicine and another held an additional certification in medical physics. Automated evaluation was conducted using RAGAS metrics, including factual correctness and context recall. The GPT-4o model combined with hybrid retrieval achieved the highest performance, as per expert evaluations. Although traditional string-based metrics such as ROUGE and the Levenshtein distance poorly align with human ratings, RAGAS provided consistent rankings across system configurations, despite showing only a modest correlation with manual scores. These findings demonstrate that integrating examination-specific institutional manuals into RAG frameworks can effectively support domain-specific question answering in nuclear medicine. Moreover, LLM-based evaluation methods such as RAGAS may serve as practical tools to complement expert reviews in developing healthcare-oriented artificial intelligence systems.
Author	Nagatani, Yukihiro Fukui, Yusuke Iguchi, Harumi Kawata, Yuhei Kobashi, Kazumasa
Author_xml	– sequence: 1 givenname: Yusuke orcidid: 0000-0002-0267-6675 surname: Fukui fullname: Fukui, Yusuke email: fyusuke@belle.shiga-med.ac.jp organization: Department of Radiology, Shiga University of Medical Science Hospital – sequence: 2 givenname: Yuhei surname: Kawata fullname: Kawata, Yuhei organization: Department of Radiology, Shiga University of Medical Science Hospital – sequence: 3 givenname: Kazumasa surname: Kobashi fullname: Kobashi, Kazumasa organization: Department of Radiology, Shiga University of Medical Science Hospital – sequence: 4 givenname: Yukihiro orcidid: 0000-0002-1110-963X surname: Nagatani fullname: Nagatani, Yukihiro organization: Department of Radiology, Shiga University of Medical Science Hospital – sequence: 5 givenname: Harumi surname: Iguchi fullname: Iguchi, Harumi organization: Department of Radiology, Shiga University of Medical Science Hospital
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/40683982$$D View this record in MEDLINE/PubMed
BookMark	eNp9kktv1DAQgC1URNuFP8ABWeLCJeBnNjkhVBUoauECZ2vWmQRXib3YcaX9JfxdHFKWx4GLZzz-5uk5Jyc-eCTkKWcvOWPbV4kL3qqKCV0x1ipeHR6QM97U5SoVPznqUp6S85RuGau5EOIROVWsbmTbiDPy_fIOxgyzC56GngKNOEeHxVhBHib0M3Z0QI9xZdIhzTjRnJwfCv0B9uAxIb3yaXZzXhgY6cdsR4RIb7Bz1nmkN-BzsYPv6AhxwHL6IUNRptDhkmsOEyy5kg2xxH5MHvYwJnxyLzfky9vLzxfvq-tP764u3lxXVolmrgChra1tVGtRq6bTUjWCcV3rLee214w1AmTX9lzrflvbrWWMd6wXu13XQ6vkhrxe4-7zbsLOloYjjGYf3QTxYAI48_eLd1_NEO4MF1K2tahLhBf3EWL4ljHNZnLJ4lg6xJCTkULyWm1FkRvy_B_0NuRYBrZQutVclo8p1LM_SzrW8uvTCiBWwMaQUsT-iHBmls0w62aYshnm52aYQ3GSq1PaL_PF-Dv3f7x-ABZPvxU
Cites_doi	10.5555/3495724.3496517 10.5551/jat.65240 10.3389/fnume.2023.1213714 10.1145/3637528.3671470 10.1109/MetroXRAINE62247.2024.10797032 10.1177/20552076251337177 10.2967/jnumed.120.254532 10.3390/healthcare13060603 10.18653/v1/2024.eacl-demo.16 10.18653/v1/2023.findings-emnlp.1057 10.3390/medicina60030445 10.2139/ssrn.4719185 10.1253/circj.CJ-23-0308 10.1162/tacl_a_00615 10.2196/58041 10.1101/2025.02.28.25323115
ContentType	Journal Article
Copyright	The Author(s) 2025 2025. The Author(s). The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. The Author(s) 2025 2025
Copyright_xml	– notice: The Author(s) 2025 – notice: 2025. The Author(s). – notice: The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. – notice: The Author(s) 2025 2025
DBID	C6C AAYXX CITATION CGR CUY CVF ECM EIF NPM K9. NAPCQ 7X8 5PM
DOI	10.1007/s12194-025-00941-y
DatabaseName	Springer Nature OA Free Journals CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed ProQuest Health & Medical Complete (Alumni) Nursing & Allied Health Premium MEDLINE - Academic PubMed Central (Full Participant titles)
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) ProQuest Health & Medical Complete (Alumni) Nursing & Allied Health Premium MEDLINE - Academic
DatabaseTitleList	CrossRef MEDLINE - Academic MEDLINE ProQuest Health & Medical Complete (Alumni)
Database_xml	– sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
Discipline	Medicine
EISSN	1865-0341
EndPage	876
ExternalDocumentID	PMC12339626 40683982 10_1007_s12194_025_00941_y
Genre	Evaluation Study Journal Article
GeographicLocations	Japan
GeographicLocations_xml	– name: Japan
GroupedDBID	--- -Y2 .86 06C 06D 0R~ 0VY 123 1N0 203 29P 29~ 2JN 2JY 2KG 2VQ 2~H 30V 4.4 406 408 409 40D 5VS 67Z 6NX 8TC 96X AAAVM AABHQ AACDK AAHNG AAIAL AAJBT AAJKR AANXM AANZL AAPKM AARHV AARTL AASML AATNV AATVU AAUYE AAWCG AAYIU AAYQN AAYTO AAYZH ABAKF ABBRH ABBXA ABDBE ABDZT ABECU ABFSG ABFTV ABHQN ABJNI ABJOX ABKCH ABMNI ABMQK ABPLI ABQBU ABRTQ ABSXP ABTEG ABTKH ABTMW ABULA ABWNU ABXPI ACAOD ACDTI ACGFS ACHSB ACKNC ACMDZ ACMLO ACOKC ACOMO ACPIV ACSTC ACZOJ ADHHG ADHIR ADKNI ADKPE ADRFC ADTPH ADURQ ADYFF ADZKW AEBTG AEFQL AEGAL AEGNC AEJHL AEJRE AEMSY AEOHA AEPYU AESKC AETLH AEVLU AEXYK AEZWR AFBBN AFDZB AFHIU AFLOW AFOHR AFQWF AFWTZ AFZKB AGAYW AGDGC AGJBK AGMZJ AGQEE AGQMX AGRTI AGWZB AGYKE AHAVH AHBYD AHPBZ AHSBF AHWEU AHYZX AIAKS AIGIU AIIXL AILAN AITGF AIXLP AJBLW AJRNO AJZVZ AKMHD ALFXC ALMA_UNASSIGNED_HOLDINGS ALWAN AMKLP AMXSW AMYLF AMYQR ANMIH AOCGG ATHPR AUKKA AXYYD AYFIA B-. BA0 BDATZ BGNMA C6C CAG COF CS3 CSCUP DDRTE DNIVK DPUIP EBLON EBS EIOEI EJD EN4 ESBYG F5P FERAY FFXSO FIGPU FINBP FNLPD FRRFC FSGXE FWDCC G-Y G-Z GGCAI GGRSB GJIRD GNWQR GQ7 GQ8 GXS H13 HF~ HG5 HG6 HLICF HMJXF HQYDN HRMNR HZ~ I0C IJ- IKXTQ IWAJR IXC IXD IZIGR IZQ I~X J-C J0Z JBSCW JCJTX JZLTJ KOV KPH LLZTM M4Y MA- NPVJJ NQJWS NU0 O9- O93 O9I O9J OAM P9S PT4 QOR QOS R89 RIG RLLFE RNS ROL RPX RSV S16 S1Z S27 S37 S3B SAP SDH SHX SISQX SJYHP SMD SNE SNPRN SNX SOHCF SOJ SPISZ SRMVM SSLCW SSXJD STPWE SZ9 T13 TSG TSK TT1 U2A U9L UG4 UOJIU UTJUX UZXMN VC2 VFIZW W48 WJK WK8 YLTOR Z45 ZMTXR ZOVNA ~A9 AAYXX CITATION CGR CUY CVF ECM EIF NPM K9. NAPCQ 7X8 5PM
ID	FETCH-LOGICAL-c428t-aea96cc849ce548d5348201565711cf50082a3d9f155f76c7c001d0f2bbdfa943
IEDL.DBID	RSV
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001531230500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1865-0333 1865-0341
IngestDate	Tue Nov 04 02:05:03 EST 2025 Fri Sep 05 15:37:06 EDT 2025 Sat Oct 11 06:57:30 EDT 2025 Sat Aug 16 01:30:54 EDT 2025 Sat Nov 29 07:37:46 EST 2025 Tue Aug 12 01:10:54 EDT 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	3
Keywords	Nuclear medicine Medical question answering Large language models (LLMs) Retrieval-augmented generation (RAG)
Language	English
License	2025. The Author(s). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c428t-aea96cc849ce548d5348201565711cf50082a3d9f155f76c7c001d0f2bbdfa943
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23
ORCID	0000-0002-0267-6675 0000-0002-1110-963X
OpenAccessLink	https://link.springer.com/10.1007/s12194-025-00941-y
PMID	40683982
PQID	3259513406
PQPubID	2044184
PageCount	16
ParticipantIDs	pubmedcentral_primary_oai_pubmedcentral_nih_gov_12339626 proquest_miscellaneous_3231647232 proquest_journals_3259513406 pubmed_primary_40683982 crossref_primary_10_1007_s12194_025_00941_y springer_journals_10_1007_s12194_025_00941_y
PublicationCentury	2000
PublicationDate	2025-09-01
PublicationDateYYYYMMDD	2025-09-01
PublicationDate_xml	– month: 09 year: 2025 text: 2025-09-01 day: 01
PublicationDecade	2020
PublicationPlace	Singapore
PublicationPlace_xml	– name: Singapore – name: Japan – name: Heidelberg
PublicationTitle	Radiological physics and technology
PublicationTitleAbbrev	Radiol Phys Technol
PublicationTitleAlternate	Radiol Phys Technol
PublicationYear	2025
Publisher	Springer Nature Singapore Springer Nature B.V
Publisher_xml	– name: Springer Nature Singapore – name: Springer Nature B.V
References	941_CR5 941_CR4 941_CR1 NM Guerreiro (941_CR21) 2023; 11 941_CR24 941_CR25 941_CR3 941_CR22 941_CR2 T Hisamatsu (941_CR16) 2024 941_CR23 941_CR20 M Djekidel (941_CR7) 2023 J Vrdoljak (941_CR6) 2025; 13 WA Weber (941_CR8) 2020; 61 D Wang (941_CR12) 2024; 26 M Tomczak (941_CR27) 2014; 21 941_CR19 941_CR17 941_CR18 941_CR13 941_CR14 941_CR10 OK Gargari (941_CR26) 2025; 11 P Lewis (941_CR9) 2020; 33 K Kusunose (941_CR15) 2023 J Miao (941_CR11) 2024; 60
References_xml	– volume: 33 start-page: 9459 year: 2020 ident: 941_CR9 publication-title: Adv Neural Inf Process Syst doi: 10.5555/3495724.3496517 – ident: 941_CR24 – ident: 941_CR2 – ident: 941_CR18 – year: 2024 ident: 941_CR16 publication-title: J Atheroscler Thromb doi: 10.5551/jat.65240 – ident: 941_CR14 – year: 2023 ident: 941_CR7 publication-title: Front Nucl Med doi: 10.3389/fnume.2023.1213714 – ident: 941_CR10 doi: 10.1145/3637528.3671470 – ident: 941_CR3 doi: 10.1109/MetroXRAINE62247.2024.10797032 – ident: 941_CR5 – volume: 11 start-page: 205520762513371 year: 2025 ident: 941_CR26 publication-title: Digit Health doi: 10.1177/20552076251337177 – volume: 61 start-page: 263S issue: Suppl 2 year: 2020 ident: 941_CR8 publication-title: J Nucl Med doi: 10.2967/jnumed.120.254532 – ident: 941_CR1 – volume: 13 start-page: 603 issue: 6 year: 2025 ident: 941_CR6 publication-title: Healthcare doi: 10.3390/healthcare13060603 – ident: 941_CR20 doi: 10.18653/v1/2024.eacl-demo.16 – ident: 941_CR22 doi: 10.18653/v1/2023.findings-emnlp.1057 – ident: 941_CR23 – volume: 60 start-page: 445 issue: 3 year: 2024 ident: 941_CR11 publication-title: Medicina doi: 10.3390/medicina60030445 – ident: 941_CR13 doi: 10.2139/ssrn.4719185 – ident: 941_CR19 – year: 2023 ident: 941_CR15 publication-title: Circ J doi: 10.1253/circj.CJ-23-0308 – ident: 941_CR17 – volume: 21 start-page: 19 issue: 1 year: 2014 ident: 941_CR27 publication-title: Trends Sport Sci – volume: 11 start-page: 1500 year: 2023 ident: 941_CR21 publication-title: Trans Assoc Comput Linguist doi: 10.1162/tacl_a_00615 – volume: 26 year: 2024 ident: 941_CR12 publication-title: J Med Internet Res doi: 10.2196/58041 – ident: 941_CR4 – ident: 941_CR25 doi: 10.1101/2025.02.28.25323115
SSID	ssj0061222
Score	2.3312836
Snippet	Recent advances in large language models (LLMs) enable domain-specific question answering using external knowledge. However, addressing information that is not...
SourceID	pubmedcentral proquest pubmed crossref springer
SourceType	Open Access Repository Aggregation Database Index Database Publisher
StartPage	861
SubjectTerms	Application programming interface Artificial intelligence Automation Certification Datasets Documents Drug dosages Humans Imaging Information Storage and Retrieval - methods Japan Keywords Kidneys Knowledge Language Large Language Models Manuals Medical and Radiation Physics Medicine Medicine & Public Health Nuclear Medicine Performance evaluation Questions Radiology Radiotherapy Research Article Retrieval Scintigraphy
Title	Evaluation of a retrieval-augmented generation system using a Japanese Institutional Nuclear Medicine Manual and large language model-automated scoring
URI	https://link.springer.com/article/10.1007/s12194-025-00941-y https://www.ncbi.nlm.nih.gov/pubmed/40683982 https://www.proquest.com/docview/3259513406 https://www.proquest.com/docview/3231647232 https://pubmed.ncbi.nlm.nih.gov/PMC12339626
Volume	18
WOSCitedRecordID	wos001531230500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAVX databaseName: SpringerLINK Contemporary 1997-Present customDbUrl: eissn: 1865-0341 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0061222 issn: 1865-0333 databaseCode: RSV dateStart: 20080101 isFulltext: true titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22 providerName: Springer Nature
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Rb9MwED6xgiZeNsYGhJXKSLxBpCROYucRoVYIqRVigPoWuXbcVUIpalOk_hL-7u6cuFUpPLDX-BIn9p3vc-78HcCbYlbJ3CYUdNcmTLMoQpMyNrRCJSLKZVLZyBWbEJOJnE6Lz92hsLXPdvchSbdS7w-7oXGlIZVfpXS4ONyewEN0d5IKNny5-e7XX3TZLnYQyxwlOefdUZm_P-PQHR1hzONUyT_ipc4Njc7v9wFP4KyDnex9qycX8KCqn8LpuAusX8Lv4Y71my0tU2zlKm3hxVBt5o6407C5o6h2Mi0BNKOs-TlKf0KXS6Usmc89cD8Y2YTIktWK-X7YWBEDKlO1YT8oBZ3536XMVeTBvpolQmjsa61dZuAVfBsNv374GHY1G0KNG5kmVJUqcq1lWugKN0MmI_IcOq6diTjWNiPIobgpLOIYK3ItNPpJE9lkNjNWFSl_Br16WVcvgHHNucm51dZkaZIKouKPMxspB0qEDeCtn7ryZ0vNUe5JmGmsSxzr0o11uQ2g72e37Mx0XXLc_GUxR1ATwOtdMxoYRU1w3JYbkuHEuYbIM4DnrTLsusMbEWBKbJEHarITIPLuw5Z6cetIvBEx8AJ3kwG889qyf69_f8bL_xO_hseJUzhKiutDr1ltqlfwSP9qFuvVAE7EVA6c9dwBrVUXQg
linkProvider	Springer Nature
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3JTsMwEB2xCbiwL2E1EjeIlMRZjwiB2FohNnGLXDsuSChFXZD4En6XGTduVZYDXONJnNgznufM-A3AftYo0lgHFHSXyg0jz0OTUtrViQgSL06DQnum2ERSr6ePj9l1dSisY7PdbUjSrNTDw25oXKFL5VcpHc5338dhMkSPRYz5N7cPdv1Fl21iB34aoyTnvDoq8_MzRt3RN4z5PVXyS7zUuKHT-f99wALMVbCTHfX1ZBHGinIJpmtVYH0ZPk4GrN-spZlgbVNpCy-6otc0xJ2KNQ1FtZHpE0AzyppvovQFulwqZcls7oH5wcjqRJYs2sz2w2qCGFCZKBV7oRR0Zn-XMlORB_vqthBCY18daTIDV-D-9OTu-Mytaja4EjcyXVcUIoulTMNMFrgZUhGR59Bx7SjxfakjghyCq0wjjtFJLBOJflJ5Omg0lBZZyFdhomyVxTowLjlXMddSqygMwoSo-P1Ie8KAkkQ7cGCnLn_tU3PkQxJmGuscxzo3Y52_O7BlZzevzLSTc9z8RT5HUOPA3qAZDYyiJjhurR7JcOJcQ-TpwFpfGQbd4Y0IMFNsSUfUZCBA5N2jLeXzkyHxRsTAM9xNOnBotWX4Xr9_xsbfxHdh5uyudpVfndcvN2E2MMpHCXJbMNFt94ptmJJv3edOe8fY0Cc2rBk-
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT-NADLZ4Ce2FxwJLlrLMStwgapLJ84iAChZaIfEQt2g6kylIKEVtisQv4e-uPUlayuOAuHacTjOx68-x_RlgN-lmcag9SrpLZfuB46BJKW3rSHiRE8Zeph0zbCLqdOLb2-TiVRe_qXavU5JlTwOxNOVF81Hp5qTxDQ3Nt2kUK5XGufbzLMz7VEhP8frlTf1fjO7b5BHcOERJznnVNvPxd0y7pnd4833Z5JvcqXFJreXv38wKLFVwlB2U-rMKM1n-ExbbVcJ9DV6Ox2zgrK-ZYAMzgQs_tMWoZwg9FesZ6mojUxJDM6qm76H0P3TFNOKS1TUJ5sUj6xCJshiweh_WFsSMykSu2AOVprP6NSozk3pwr6KP0Br3GkpTMbgO163jq8MTu5rlYEsMcApbZCIJpYz9RGYYJKmASHWojTuIXFfqgKCI4CrRiG90FMpIov9Ujva6XaVF4vMNmMv7ebYJjEvOVci11CrwPT8iin430I4wYCXSFuzVjzF9LCk70gk5M511imedmrNOny1o1E86rcx3mHIMCgOXI9ix4O94GQ2Psil4bv0RyXDiYkNEasGvUjHG2-GFCDxjXImnVGYsQKTe0yv5_Z0h90YkwROMMi3YrzVn8rs-v43fXxPfgcWLo1Z6fto524IfntE9qptrwFwxGGXbsCCfivvh4I8xp_8DViIi
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Evaluation+of+a+retrieval-augmented+generation+system+using+a+Japanese+Institutional+Nuclear+Medicine+Manual+and+large+language+model-automated+scoring&rft.jtitle=Radiological+physics+and+technology&rft.au=Fukui%2C+Yusuke&rft.au=Kawata%2C+Yuhei&rft.au=Kobashi%2C+Kazumasa&rft.au=Nagatani%2C+Yukihiro&rft.date=2025-09-01&rft.pub=Springer+Nature+B.V&rft.issn=1865-0333&rft.eissn=1865-0341&rft.volume=18&rft.issue=3&rft.spage=861&rft.epage=876&rft_id=info:doi/10.1007%2Fs12194-025-00941-y&rft.externalDBID=NO_FULL_TEXT
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1865-0333&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1865-0333&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1865-0333&client=summon