Academic Journal
Artificial Intelligence for Diagnosing Normal Anatomical Variants and Pathological Oral Mucosal Lesions: A Prospective Observational Study.
| Τίτλος: | Artificial Intelligence for Diagnosing Normal Anatomical Variants and Pathological Oral Mucosal Lesions: A Prospective Observational Study. |
|---|---|
| Συγγραφείς: | Glavina A; Department of Dental Medicine, University Hospital of Split, 21000 Split, Croatia.; Department of Oral Medicine, Study of Dental Medicine, School of Medicine, University of Split, 21000 Split, Croatia., Galešić M; Department of Oral Medicine, Study of Dental Medicine, School of Medicine, University of Split, 21000 Split, Croatia., Poposki B; Department of Oral and Periodontal Diseases, Faculty of Dentistry-Skopje, Ss. Cyril and Methodius University in Skopje, 1000 Skopje, North Macedonia., Tadin A; Department of Dental Medicine, University Hospital of Split, 21000 Split, Croatia.; Department of Restorative Dental Medicine and Endodontics, Study of Dental Medicine, School of Medicine, University of Split, 21000 Split, Croatia. |
| Πηγή: | Medicina (Kaunas, Lithuania) [Medicina (Kaunas)] 2026 Aug 21; Vol. 62 (8). Date of Electronic Publication: 2026 Aug 21. |
| Τύπος έκδοσης: | Journal Article; Observational Study |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: MDPI Country of Publication: Switzerland NLM ID: 9425208 Publication Model: Electronic Cited Medium: Internet ISSN: 1648-9144 (Electronic) Linking ISSN: 1010660X NLM ISO Abbreviation: Medicina (Kaunas) Subsets: MEDLINE |
| Imprint Name(s): | Publication: 2018- : Basel, Switzerland : MDPI Original Publication: Kaunas : Lietuvos gydytojų sąjunga |
| Ιατρικοί όροι (MeSH): | Artificial Intelligence*/standards , Artificial Intelligence*/trends , Mouth Mucosa*/abnormalities , Mouth Mucosa*/pathology, Humans ; Prospective Studies ; Female ; Adult ; Male ; Middle Aged ; Croatia ; Aged ; Generative Artificial Intelligence |
| Περίληψη: | Background and Objectives: Artificial intelligence (AI) is increasingly used in clinical dentistry, but its diagnostic accuracy for oral mucosal lesions based on clinical photographs remains insufficiently validated. This prospective observational study compared the Top-1 diagnostic accuracy of ChatGPT-4o and ChatGPT-5 in identifying normal anatomical variants and pathological oral mucosal lesions and evaluated their performance across anatomical sites. Materials and Methods: Seventy adults with either normal anatomical variants (n = 21) or pathological oral mucosal lesions (n = 49) were consecutively recruited at the Department of Dental Medicine, University Hospital of Split, Croatia. One standardized clinical photograph per patient was analyzed by ChatGPT-4o and ChatGPT-5 under image-only and image-plus-text conditions using identical prompts. The reference diagnosis was established by an oral medicine specialist, with histopathological examination (HPE) performed when clinically indicated. Diagnostic performance was assessed using Top-1 accuracy and McNemar's test. Results: Both models showed low accuracy with image-only input, but performance improved significantly after clinical information was added (p < 0.001). Overall Top-1 accuracy increased from 19.0% to 69.0% for ChatGPT-4o and from 9.0% to 51.0% for ChatGPT-5. For normal anatomical variants, accuracy increased from 14.3% to 81.0% and from 14.3% to 76.2%, respectively. For pathological oral mucosal lesions, accuracy increased from 20.4% to 63.3% and from 6.1% to 40.8%, respectively. ChatGPT-4o showed numerically higher accuracy than ChatGPT-5, particularly for pathological oral mucosal lesions, but no statistically significant difference was found between the models in the corresponding paired comparisons. Conclusions: Diagnostic performance was limited with image-only input but improved substantially when standardized clinical information accompanied the images. The numerical differences between models, particularly for pathological oral mucosal lesions, may be clinically relevant but do not establish superiority or equivalence. Neither model can currently replace conventional clinical diagnosis, and AI should be regarded as a clinical decision-support tool for evaluating oral mucosal lesions. |
| References: | Medicina (Kaunas). 2025 Jul 30;61(8):. (PMID: 40870424) Oral Oncol. 2024 Aug;155:106873. (PMID: 38833826) Otolaryngol Head Neck Surg. 2024 Jun;170(6):1492-1503. (PMID: 37595113) Int Dent J. 2006 Aug;56(4):203-8. (PMID: 16972394) Front Dent Med. 2023 Feb 20;4:1085251. (PMID: 39935549) Bioengineering (Basel). 2024 Nov 18;11(11):. (PMID: 39593819) Eur Arch Otorhinolaryngol. 2024 Nov;281(11):6123-6131. (PMID: 38703195) Dentomaxillofac Radiol. 2022 May 01;51(4):20210515. (PMID: 35113725) Cureus. 2023 Oct 30;15(10):e47941. (PMID: 38034167) Front Med (Lausanne). 2024 Jan 08;10:1305954. (PMID: 38259845) Int J Comput Vis. 2024;132(9):3753-3769. (PMID: 39211895) J Voice. 2025 Mar 26;:. (PMID: 40148204) CA Cancer J Clin. 2026 Jul-Aug;76(4):e70090. (PMID: 42417444) Acta Stomatol Croat. 2023 Mar;57(1):70-84. (PMID: 37288152) J Clin Med. 2020 Oct 12;9(10):. (PMID: 33053724) JAMA. 2025 Jan 7;333(1):71-74. (PMID: 39425955) Sci Rep. 2020 Mar 31;10(1):5711. (PMID: 32235882) Vis Comput Ind Biomed Art. 2024 Aug 5;7(1):20. (PMID: 39101954) Diagnostics (Basel). 2025 Jan 24;15(3):. (PMID: 39941210) Dent J (Basel). 2025 Apr 30;13(5):. (PMID: 40422618) Med Sci Monit. 2025 Apr 08;31:e946676. (PMID: 40195079) Nat Med. 2020 Jun;26(6):807-808. (PMID: 32514173) Cancers (Basel). 2021 Dec 01;13(23):. (PMID: 34885164) |
| Contributed Indexing: | Keywords: artificial intelligence; clinical; decision support systems; diagnosis; mouth mucosa; oral; photography |
| Entry Date(s): | Date Created: 20260827 Date Completed: 20260827 Latest Revision: 20260903 |
| Update Code: | 20260903 |
| PubMed Central ID: | PMC13515991 |
| DOI: | 10.3390/medicina62081610 |
| PMID: | 42654507 |
| Βάση Δεδομένων: | MEDLINE |
| FullText | Text: Availability: 0 CustomLinks: – Url: https://www.doi.org/10.3390/medicina62081610? 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=16489144&ISBN=&volume=62&issue=8&date=20260821&spage=&pages=&title=Medicina (Kaunas, Lithuania)&atitle=Artificial%20Intelligence%20for%20Diagnosing%20Normal%20Anatomical%20Variants%20and%20Pathological%20Oral%20Mucosal%20Lesions%3A%20A%20Prospective%20Observational%20Study.&aulast=Glavina%20A&id=DOI:10.3390/medicina62081610 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: 42654507 AccessLevel: 3 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Artificial Intelligence for Diagnosing Normal Anatomical Variants and Pathological Oral Mucosal Lesions: A Prospective Observational Study. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AU" term="%22Glavina+A%22">Glavina A</searchLink>; Department of Dental Medicine, University Hospital of Split, 21000 Split, Croatia.; Department of Oral Medicine, Study of Dental Medicine, School of Medicine, University of Split, 21000 Split, Croatia.<br /><searchLink fieldCode="AU" term="%22Galešić+M%22">Galešić M</searchLink>; Department of Oral Medicine, Study of Dental Medicine, School of Medicine, University of Split, 21000 Split, Croatia.<br /><searchLink fieldCode="AU" term="%22Poposki+B%22">Poposki B</searchLink>; Department of Oral and Periodontal Diseases, Faculty of Dentistry-Skopje, Ss. Cyril and Methodius University in Skopje, 1000 Skopje, North Macedonia.<br /><searchLink fieldCode="AU" term="%22Tadin+A%22">Tadin A</searchLink>; Department of Dental Medicine, University Hospital of Split, 21000 Split, Croatia.; Department of Restorative Dental Medicine and Endodontics, Study of Dental Medicine, School of Medicine, University of Split, 21000 Split, Croatia. – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%229425208%22">Medicina (Kaunas, Lithuania)</searchLink> [Medicina (Kaunas)] 2026 Aug 21; Vol. 62 (8). <i>Date of Electronic Publication: </i>2026 Aug 21. – Name: TypePub Label: Publication Type Group: TypPub Data: Journal Article; Observational Study – Name: Language Label: Language Group: Lang Data: English – Name: TitleSource Label: Journal Info Group: Src Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22MDPI%22">MDPI </searchLink><i>Country of Publication: </i>Switzerland <i>NLM ID: </i>9425208 <i>Publication Model: </i>Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1648-9144 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%221010660X%22">1010660X </searchLink><i>NLM ISO Abbreviation: </i>Medicina (Kaunas) <i>Subsets: </i>MEDLINE – Name: PublisherInfo Label: Imprint Name(s) Group: PubInfo Data: <i>Publication</i>: 2018- : Basel, Switzerland : MDPI<br /><i>Original Publication</i>: Kaunas : Lietuvos gydytojų sąjunga – Name: SubjectMESH Label: MeSH Terms Group: Su Data: <searchLink fieldCode="MM" term="%22Artificial+Intelligence%22">Artificial Intelligence*</searchLink>/<searchLink fieldCode="MM" term="%22Artificial+Intelligence+standards%22">standards</searchLink> <br /><searchLink fieldCode="MM" term="%22Artificial+Intelligence%22">Artificial Intelligence*</searchLink>/<searchLink fieldCode="MM" term="%22Artificial+Intelligence+trends%22">trends</searchLink> <br /><searchLink fieldCode="MM" term="%22Mouth+Mucosa%22">Mouth Mucosa*</searchLink>/<searchLink fieldCode="MM" term="%22Mouth+Mucosa+abnormalities%22">abnormalities</searchLink> <br /><searchLink fieldCode="MM" term="%22Mouth+Mucosa%22">Mouth Mucosa*</searchLink>/<searchLink fieldCode="MM" term="%22Mouth+Mucosa+pathology%22">pathology</searchLink><br /><searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Prospective+Studies%22">Prospective Studies</searchLink> ; <searchLink fieldCode="MH" term="%22Female%22">Female</searchLink> ; <searchLink fieldCode="MH" term="%22Adult%22">Adult</searchLink> ; <searchLink fieldCode="MH" term="%22Male%22">Male</searchLink> ; <searchLink fieldCode="MH" term="%22Middle+Aged%22">Middle Aged</searchLink> ; <searchLink fieldCode="MH" term="%22Croatia%22">Croatia</searchLink> ; <searchLink fieldCode="MH" term="%22Aged%22">Aged</searchLink> ; <searchLink fieldCode="MH" term="%22Generative+Artificial+Intelligence%22">Generative Artificial Intelligence</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Background and Objectives: Artificial intelligence (AI) is increasingly used in clinical dentistry, but its diagnostic accuracy for oral mucosal lesions based on clinical photographs remains insufficiently validated. This prospective observational study compared the Top-1 diagnostic accuracy of ChatGPT-4o and ChatGPT-5 in identifying normal anatomical variants and pathological oral mucosal lesions and evaluated their performance across anatomical sites. Materials and Methods: Seventy adults with either normal anatomical variants (n = 21) or pathological oral mucosal lesions (n = 49) were consecutively recruited at the Department of Dental Medicine, University Hospital of Split, Croatia. One standardized clinical photograph per patient was analyzed by ChatGPT-4o and ChatGPT-5 under image-only and image-plus-text conditions using identical prompts. The reference diagnosis was established by an oral medicine specialist, with histopathological examination (HPE) performed when clinically indicated. Diagnostic performance was assessed using Top-1 accuracy and McNemar's test. Results: Both models showed low accuracy with image-only input, but performance improved significantly after clinical information was added (p &lt; 0.001). Overall Top-1 accuracy increased from 19.0% to 69.0% for ChatGPT-4o and from 9.0% to 51.0% for ChatGPT-5. For normal anatomical variants, accuracy increased from 14.3% to 81.0% and from 14.3% to 76.2%, respectively. For pathological oral mucosal lesions, accuracy increased from 20.4% to 63.3% and from 6.1% to 40.8%, respectively. ChatGPT-4o showed numerically higher accuracy than ChatGPT-5, particularly for pathological oral mucosal lesions, but no statistically significant difference was found between the models in the corresponding paired comparisons. Conclusions: Diagnostic performance was limited with image-only input but improved substantially when standardized clinical information accompanied the images. The numerical differences between models, particularly for pathological oral mucosal lesions, may be clinically relevant but do not establish superiority or equivalence. Neither model can currently replace conventional clinical diagnosis, and AI should be regarded as a clinical decision-support tool for evaluating oral mucosal lesions. – Name: Ref Label: References Group: RefInfo Data: Medicina (Kaunas). 2025 Jul 30;61(8):. (PMID: <searchLink fieldCode="PM" term="%2240870424%22">40870424)</searchLink><br />Oral Oncol. 2024 Aug;155:106873. (PMID: <searchLink fieldCode="PM" term="%2238833826%22">38833826)</searchLink><br />Otolaryngol Head Neck Surg. 2024 Jun;170(6):1492-1503. (PMID: <searchLink fieldCode="PM" term="%2237595113%22">37595113)</searchLink><br />Int Dent J. 2006 Aug;56(4):203-8. (PMID: <searchLink fieldCode="PM" term="%2216972394%22">16972394)</searchLink><br />Front Dent Med. 2023 Feb 20;4:1085251. (PMID: <searchLink fieldCode="PM" term="%2239935549%22">39935549)</searchLink><br />Bioengineering (Basel). 2024 Nov 18;11(11):. (PMID: <searchLink fieldCode="PM" term="%2239593819%22">39593819)</searchLink><br />Eur Arch Otorhinolaryngol. 2024 Nov;281(11):6123-6131. (PMID: <searchLink fieldCode="PM" term="%2238703195%22">38703195)</searchLink><br />Dentomaxillofac Radiol. 2022 May 01;51(4):20210515. (PMID: <searchLink fieldCode="PM" term="%2235113725%22">35113725)</searchLink><br />Cureus. 2023 Oct 30;15(10):e47941. (PMID: <searchLink fieldCode="PM" term="%2238034167%22">38034167)</searchLink><br />Front Med (Lausanne). 2024 Jan 08;10:1305954. (PMID: <searchLink fieldCode="PM" term="%2238259845%22">38259845)</searchLink><br />Int J Comput Vis. 2024;132(9):3753-3769. (PMID: <searchLink fieldCode="PM" term="%2239211895%22">39211895)</searchLink><br />J Voice. 2025 Mar 26;:. (PMID: <searchLink fieldCode="PM" term="%2240148204%22">40148204)</searchLink><br />CA Cancer J Clin. 2026 Jul-Aug;76(4):e70090. (PMID: <searchLink fieldCode="PM" term="%2242417444%22">42417444)</searchLink><br />Acta Stomatol Croat. 2023 Mar;57(1):70-84. (PMID: <searchLink fieldCode="PM" term="%2237288152%22">37288152)</searchLink><br />J Clin Med. 2020 Oct 12;9(10):. (PMID: <searchLink fieldCode="PM" term="%2233053724%22">33053724)</searchLink><br />JAMA. 2025 Jan 7;333(1):71-74. (PMID: <searchLink fieldCode="PM" term="%2239425955%22">39425955)</searchLink><br />Sci Rep. 2020 Mar 31;10(1):5711. (PMID: <searchLink fieldCode="PM" term="%2232235882%22">32235882)</searchLink><br />Vis Comput Ind Biomed Art. 2024 Aug 5;7(1):20. (PMID: <searchLink fieldCode="PM" term="%2239101954%22">39101954)</searchLink><br />Diagnostics (Basel). 2025 Jan 24;15(3):. (PMID: <searchLink fieldCode="PM" term="%2239941210%22">39941210)</searchLink><br />Dent J (Basel). 2025 Apr 30;13(5):. (PMID: <searchLink fieldCode="PM" term="%2240422618%22">40422618)</searchLink><br />Med Sci Monit. 2025 Apr 08;31:e946676. (PMID: <searchLink fieldCode="PM" term="%2240195079%22">40195079)</searchLink><br />Nat Med. 2020 Jun;26(6):807-808. (PMID: <searchLink fieldCode="PM" term="%2232514173%22">32514173)</searchLink><br />Cancers (Basel). 2021 Dec 01;13(23):. (PMID: <searchLink fieldCode="PM" term="%2234885164%22">34885164)</searchLink> – Name: SubjectMinor Label: Contributed Indexing Group: Data: <i>Keywords: </i>artificial intelligence; clinical; decision support systems; diagnosis; mouth mucosa; oral; photography – Name: DateEntry Label: Entry Date(s) Group: Date Data: <i>Date Created: </i>20260827 <i>Date Completed: </i>20260827 <i>Latest Revision: </i>20260903 – Name: DateUpdate Label: Update Code Group: Date Data: 20260903 – Name: PubmedCentralID Label: PubMed Central ID Group: ID Data: PMC13515991 – Name: DOI Label: DOI Group: ID Data: 10.3390/medicina62081610 – Name: AN Label: PMID Group: ID Data: 42654507 |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42654507 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.3390/medicina62081610 Languages: – Code: eng Text: English Subjects: – SubjectFull: Humans Type: general – SubjectFull: Prospective Studies Type: general – SubjectFull: Female Type: general – SubjectFull: Adult Type: general – SubjectFull: Male Type: general – SubjectFull: Middle Aged Type: general – SubjectFull: Croatia Type: general – SubjectFull: Aged Type: general – SubjectFull: Generative Artificial Intelligence Type: general – SubjectFull: Artificial Intelligence standards Type: general – SubjectFull: Artificial Intelligence trends Type: general – SubjectFull: Mouth Mucosa abnormalities Type: general – SubjectFull: Mouth Mucosa pathology Type: general Titles: – TitleFull: Artificial Intelligence for Diagnosing Normal Anatomical Variants and Pathological Oral Mucosal Lesions: A Prospective Observational Study. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Glavina A – PersonEntity: Name: NameFull: Galešić M – PersonEntity: Name: NameFull: Poposki B – PersonEntity: Name: NameFull: Tadin A IsPartOfRelationships: – BibEntity: Dates: – D: 21 M: 08 Text: 2026 Aug 21 Type: published Y: 2026 Identifiers: – Type: issn-electronic Value: 1648-9144 Numbering: – Type: volume Value: 62 – Type: issue Value: 8 Titles: – TitleFull: Medicina (Kaunas, Lithuania) Type: main |
| ResultId | 1 |