Academic Journal
Προβολή σε EDS

Accuracy and hallucination behaviors of artificial intelligence-based large language models for dental implant identification on radiographs.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Accuracy and hallucination behaviors of artificial intelligence-based large language models for dental implant identification on radiographs.
Συγγραφείς: Hlaing NHMM; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea., Park K; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea., Wan Q; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea., Lee SJ; Department of Restorative Dentistry and Biomaterials Sciences, Harvard School of Dental Medicine, Boston, Mass, United States., Gallucci GO; Department of Restorative Dentistry and Biomaterials Sciences, Harvard School of Dental Medicine, Boston, Mass, United States., Lee JH; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea; Department of Restorative Dentistry and Biomaterials Sciences, Harvard School of Dental Medicine, Boston, Mass, United States; Dental Research Institute, Seoul National University School of Dentistry, Seoul, Republic of Korea. Electronic address: jaehyun_lee@hsdm.harvard.edu.
Πηγή: Journal of dentistry [J Dent] 2026 Oct; Vol. 173, pp. 106754. Date of Electronic Publication: 2026 May 15.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: Elsevier Country of Publication: England NLM ID: 0354422 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1879-176X (Electronic) Linking ISSN: 03005712 NLM ISO Abbreviation: J Dent Subsets: MEDLINE
Imprint Name(s): Publication: Kidlington : Elsevier
Original Publication: Bristol, Eng., Wright.
Ιατρικοί όροι (MeSH): Large Language Models* , Dental Implants* , Artificial Intelligence*, Humans ; Retrospective Studies ; Generative Artificial Intelligence ; Radiography, Panoramic
Περίληψη: Objectives: This study aimed to evaluate the accuracy and error patterns of four artificial intelligence-based large language models (LLMs) in identifying dental implant characteristics including implant level, connection type, brand, and manufacturer, on radiographic images.
Methods: In total, 120 standardized radiographic images (60 periapical and 60 panoramic) representing six implant systems (Straumann BL, Straumann TL, Osstem TS [Hiossen ET], Osstem [Hiossen] US, Osstem [Hiossen] SS, and Dentium SuperLine) were retrospectively collected and analyzed. For each image, four multimodal LLMs (ChatGPT-5, ChatGPT-4o, Gemini 2.5 Pro, and Grok 4) were queried six times with a standardized prompt that requested the implant level, connection type, brand, and manufacturer. The primary outcome was response-level accuracy. Responses were classified as successful or unsuccessful identification; unsuccessful responses were further categorized as definitive incorrect (hallucination), tentative incorrect, ambiguous, or abstention. Statistical analysis used generalized estimating equations for accuracy and Chi-square tests for error types (α=0.05).
Results: Structural characteristics (implant level and connection type) were identified with accuracies ranging from 51.67% to 82.5%, whereas brand and manufacturer recognition remained consistently poor (4.17%-32.78%). Hallucinations represented the predominant error type (>87% of errors in all models), while ChatGPT-5 showed a comparatively higher rate of abstention and lower hallucination frequency.
Conclusions: Current LLMs have limited capability for the definitive radiographic identification of dental implant systems, particularly for specific brands. Recent model iterations show a promising trend toward reduced hallucinations and increased reliability through abstention; however, their high error rates indicate that they can only be considered as supplementary tools requiring clinician verification.
Clinical Significance: Current LLMs are not sufficiently reliable for definitive radiographic implant identification, and their outputs should be interpreted with clinician verification. However, the increasing tendency of recent models to acknowledge uncertainty indicates promising progress toward safer clinical decision support systems.
(Copyright © 2026 Elsevier Ltd. All rights reserved.)
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Contributed Indexing: Keywords: Artificial intelligence; Computer-assisted diagnosis; Dental implants; Dental radiography; Hallucinations; Large language models
Substance Nomenclature: 0 (Dental Implants)
Entry Date(s): Date Created: 20260516 Date Completed: 20260715 Latest Revision: 20260715
Update Code: 20260715
DOI: 10.1016/j.jdent.2026.106754
PMID: 42142612
Βάση Δεδομένων: MEDLINE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://www.doi.org/10.1016/j.jdent.2026.106754?
    Name: ScienceDirect (all content) (s7799221)
    Category: fullText
    Text: View record from ScienceDirect
    MouseOverText: View record from ScienceDirect
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:cmedm&genre=article&issn=1879176X&ISBN=&volume=173&issue=&date=20261001&spage=106754&pages=106754&title=Journal of dentistry&atitle=Accuracy%20and%20hallucination%20behaviors%20of%20artificial%20intelligence-based%20large%20language%20models%20for%20dental%20implant%20identification%20on%20radiographs.&aulast=Hlaing%20NHMM&id=DOI:10.1016/j.jdent.2026.106754
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: cmedm
DbLabel: MEDLINE
An: 42142612
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Accuracy and hallucination behaviors of artificial intelligence-based large language models for dental implant identification on radiographs.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Hlaing+NHMM%22">Hlaing NHMM</searchLink>; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea.<br /><searchLink fieldCode="AU" term="%22Park+K%22">Park K</searchLink>; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea.<br /><searchLink fieldCode="AU" term="%22Wan+Q%22">Wan Q</searchLink>; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea.<br /><searchLink fieldCode="AU" term="%22Lee+SJ%22">Lee SJ</searchLink>; Department of Restorative Dentistry and Biomaterials Sciences, Harvard School of Dental Medicine, Boston, Mass, United States.<br /><searchLink fieldCode="AU" term="%22Gallucci+GO%22">Gallucci GO</searchLink>; Department of Restorative Dentistry and Biomaterials Sciences, Harvard School of Dental Medicine, Boston, Mass, United States.<br /><searchLink fieldCode="AU" term="%22Lee+JH%22">Lee JH</searchLink>; Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Republic of Korea; Department of Restorative Dentistry and Biomaterials Sciences, Harvard School of Dental Medicine, Boston, Mass, United States; Dental Research Institute, Seoul National University School of Dentistry, Seoul, Republic of Korea. Electronic address: jaehyun_lee@hsdm.harvard.edu.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%220354422%22">Journal of dentistry</searchLink> [J Dent] 2026 Oct; Vol. 173, pp. 106754. <i>Date of Electronic Publication: </i>2026 May 15.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22Elsevier%22">Elsevier </searchLink><i>Country of Publication: </i>England <i>NLM ID: </i>0354422 <i>Publication Model: </i>Print-Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1879-176X (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2203005712%22">03005712 </searchLink><i>NLM ISO Abbreviation: </i>J Dent <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Publication</i>: Kidlington : Elsevier<br /><i>Original Publication</i>: Bristol, Eng., Wright.
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Large+Language+Models%22">Large Language Models*</searchLink> <br /><searchLink fieldCode="MM" term="%22Dental+Implants%22">Dental Implants*</searchLink> <br /><searchLink fieldCode="MM" term="%22Artificial+Intelligence%22">Artificial Intelligence*</searchLink><br /><searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Retrospective+Studies%22">Retrospective Studies</searchLink> ; <searchLink fieldCode="MH" term="%22Generative+Artificial+Intelligence%22">Generative Artificial Intelligence</searchLink> ; <searchLink fieldCode="MH" term="%22Radiography%2C+Panoramic%22">Radiography, Panoramic</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Objectives: This study aimed to evaluate the accuracy and error patterns of four artificial intelligence-based large language models (LLMs) in identifying dental implant characteristics including implant level, connection type, brand, and manufacturer, on radiographic images.<br />Methods: In total, 120 standardized radiographic images (60 periapical and 60 panoramic) representing six implant systems (Straumann BL, Straumann TL, Osstem TS [Hiossen ET], Osstem [Hiossen] US, Osstem [Hiossen] SS, and Dentium SuperLine) were retrospectively collected and analyzed. For each image, four multimodal LLMs (ChatGPT-5, ChatGPT-4o, Gemini 2.5 Pro, and Grok 4) were queried six times with a standardized prompt that requested the implant level, connection type, brand, and manufacturer. The primary outcome was response-level accuracy. Responses were classified as successful or unsuccessful identification; unsuccessful responses were further categorized as definitive incorrect (hallucination), tentative incorrect, ambiguous, or abstention. Statistical analysis used generalized estimating equations for accuracy and Chi-square tests for error types (α=0.05).<br />Results: Structural characteristics (implant level and connection type) were identified with accuracies ranging from 51.67% to 82.5%, whereas brand and manufacturer recognition remained consistently poor (4.17%-32.78%). Hallucinations represented the predominant error type (>87% of errors in all models), while ChatGPT-5 showed a comparatively higher rate of abstention and lower hallucination frequency.<br />Conclusions: Current LLMs have limited capability for the definitive radiographic identification of dental implant systems, particularly for specific brands. Recent model iterations show a promising trend toward reduced hallucinations and increased reliability through abstention; however, their high error rates indicate that they can only be considered as supplementary tools requiring clinician verification.<br />Clinical Significance: Current LLMs are not sufficiently reliable for definitive radiographic implant identification, and their outputs should be interpreted with clinician verification. However, the increasing tendency of recent models to acknowledge uncertainty indicates promising progress toward safer clinical decision support systems.<br /> (Copyright © 2026 Elsevier Ltd. All rights reserved.)
– Name: Abstract
  Label: Competing Interests
  Group: Ab
  Data: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
– Name: SubjectMinor
  Label: Contributed Indexing
  Group:
  Data: <i>Keywords: </i>Artificial intelligence; Computer-assisted diagnosis; Dental implants; Dental radiography; Hallucinations; Large language models
– Name: NumberCAS
  Label: Substance Nomenclature
  Group: ID
  Data: 0 (Dental Implants)
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20260516 <i>Date Completed: </i>20260715 <i>Latest Revision: </i>20260715
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20260715
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1016/j.jdent.2026.106754
– Name: AN
  Label: PMID
  Group: ID
  Data: 42142612
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42142612
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.jdent.2026.106754
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        StartPage: 106754
    Subjects:
      – SubjectFull: Humans
        Type: general
      – SubjectFull: Retrospective Studies
        Type: general
      – SubjectFull: Generative Artificial Intelligence
        Type: general
      – SubjectFull: Radiography, Panoramic
        Type: general
      – SubjectFull: Large Language Models
        Type: general
      – SubjectFull: Dental Implants
        Type: general
      – SubjectFull: Artificial Intelligence
        Type: general
    Titles:
      – TitleFull: Accuracy and hallucination behaviors of artificial intelligence-based large language models for dental implant identification on radiographs.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Hlaing NHMM
      – PersonEntity:
          Name:
            NameFull: Park K
      – PersonEntity:
          Name:
            NameFull: Wan Q
      – PersonEntity:
          Name:
            NameFull: Lee SJ
      – PersonEntity:
          Name:
            NameFull: Gallucci GO
      – PersonEntity:
          Name:
            NameFull: Lee JH
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 10
              Text: 2026 Oct
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 1879-176X
          Numbering:
            – Type: volume
              Value: 173
          Titles:
            – TitleFull: Journal of dentistry
              Type: main
ResultId 1