Academic Journal
Audiologist-Guided Multimodal AI for Pure-Tone Audiometry and Tympanometry Interpretation and Reporting.
| Τίτλος: | Audiologist-Guided Multimodal AI for Pure-Tone Audiometry and Tympanometry Interpretation and Reporting. |
|---|---|
| Συγγραφείς: | Wu X; Department of Otolaryngology-Head and Neck Surgery, Ningbo Hospital of Integrated Traditional Chinese and Western Medicine, Ningbo, China., Shen X; School of TCM and Pharmacology Health and Early Childhood Care, Ningbo College of Health Sciences, Ningbo, China. shenxiaoli0574@outlook.com., Mo C; Orka Health Limited, Hong Kong, Hong Kong SAR, China.; Department of Special Education and Counselling, The Education University of Hong Kong, Hong Kong, Hong Kong SAR, China., Shao S; Department of Otorhinolaryngology, Ningbo Yinzhou District No.3 Hospital, Ningbo, China., Wang J; Department of Otolaryngology, The Third People's Hospital of Zhengzhou, Zhengzhou, China., Wang S; Faculty of Education, The University of Hong Kong, Hong Kong, Hong Kong SAR, China. sqgwang@hku.hk. |
| Πηγή: | Journal of medical systems [J Med Syst] 2026 Oct 01; Vol. 50 (1). Date of Electronic Publication: 2026 Oct 01. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Kluwer Academic/Plenum Publishers Country of Publication: United States NLM ID: 7806056 Publication Model: Electronic Cited Medium: Internet ISSN: 1573-689X (Electronic) Linking ISSN: 01485598 NLM ISO Abbreviation: J Med Syst Subsets: MEDLINE |
| Imprint Name(s): | Publication: 1999- : New York, NY : Kluwer Academic/Plenum Publishers Original Publication: New York, Plenum Press. |
| Ιατρικοί όροι (MeSH): | Audiometry, Pure-Tone*/methods , Acoustic Impedance Tests*/methods , Artificial Intelligence* , Audiologists*, Humans ; Female ; Middle Aged ; Male ; Adult ; Aged |
| Περίληψη: | Multimodal artificial intelligence (AI) could support audiology reporting, but an end-to-end score can obscure whether errors arise from image transcription, specialty-rule execution, or communication. We evaluated these as separate modules in 155 de-identified outpatient records representing 151 patients. In Module 1, schema-only and audiologist-skill prompts were compared in 50 development records; the locked skill was then tested with a Codex GPT5.5 agentic workflow in 101 independent patients. A post hoc robustness analysis tested matched schema-only and skill-assisted direct-API conditions with DeepSeek-V4-Flash-Vision-Exp in the same 101 patients. The skill increased hearing-loss-type agreement from 85.0% to 94.0% and acoustic-reflex agreement from 70.4% to 98.4% in development. Codex validation yielded 1809/1820 (99.4%) exact numeric thresholds, but only 9/15 (60.0%) no-response entries were correct. In the DeepSeek robustness analysis, the skill increased type agreement from 37.6% to 70.3%, while exact numeric-threshold agreement remained 72.1% and no-response agreement remained 0%. In Module 2, AI workflows received adjudicated structured values, not Module 1 outputs. After two audiologists refined reports using 60 records, 91 independent patients were evaluated. All 25 rule-derived fields were correct in 91/91 Codex and 87/91 DeepSeek cases; both audiologists rated all Chinese reports accurate or basically accurate. These were different operational workflows, not a matched foundation-model comparison. The findings support prospective evaluation of modular, auditable clinician assistance, not autonomous diagnosis, end-to-end performance, or patient usability. (© 2026. The Author(s).) |
| Competing Interests: | Declarations. Ethics Approval: The study was approved by the Human Research Ethics Committee of the University of Hong Kong (EA260116) and the Ethics Committee of Ningbo Hospital of Integrated Traditional Chinese and Western Medicine (YJ260720192039). The Ningbo Hospital approval and applicable institutional and hospital data-governance requirements permitted transmission and processing of fully de-identified clinical records and images through OpenRouter and downstream AI providers under the retention arrangements actually used in this study. Individual patient consent was waived for retrospective secondary use of the clinical records under the hospital’s data-governance framework. Consent to Participate: All expert and lay raters provided written informed consent using the University of Hong Kong HREC-approved consent form. Individual patient consent was waived for retrospective secondary use. Clinical records and audiological test images were de-identified before research use. Consent to Publish: Not applicable. The manuscript contains no identifiable individual participant information. Generative AI and AI-Assisted Technologies in the Writing Process: OpenAI Codex was used to assist with code generation, language editing, reference-format checks, and manuscript formatting. All analyses, interpretations, citations, and manuscript text were reviewed and approved by the authors, who take full responsibility for the content. Competing Interests: The authors declare no competing interests. Trial Registration: Clinical trial number: not applicable. |
| References: | Laryngoscope Investig Otolaryngol. 2024 Feb 08;9(1):e1199. (PMID: 38362190) Radiol Artif Intell. 2024 Jul;6(4):e240300. (PMID: 38809149) Am J Otolaryngol. 2024 Nov-Dec;45(6):104502. (PMID: 39197330) Clin Otolaryngol. 2024 Sep;49(5):595-603. (PMID: 38745553) BMC Med Inform Decis Mak. 2020 Nov 30;20(1):310. (PMID: 33256715) Otolaryngol Head Neck Surg. 2024 May;170(5):1228-1233. (PMID: 38682759) J Med Internet Res. 2026 Sep 11;28:e94755. (PMID: 42727086) Int J Audiol. 2026 Aug;65(8):992-995. (PMID: 41439697) Health Lit Res Pract. 2020 Jun 4;4(2):e129-e137. (PMID: 32510569) Nat Med. 2025 Oct;31(10):3283-3289. (PMID: 40954311) Audiol Res. 2026 Jul 31;16(4):. (PMID: 42645122) Hear Res. 2025 Aug;464:109337. (PMID: 40532492) Sensors (Basel). 2024 Nov 06;24(22):. (PMID: 39598904) J Med Internet Res. 2022 Feb 2;24(2):e32581. (PMID: 34919056) JMIR Med Educ. 2024 Apr 26;10:e55595. (PMID: 38693697) Cureus. 2024 May 8;16(5):e59857. (PMID: 38854312) Otolaryngol Head Neck Surg. 2024 Dec;171(6):1751-1757. (PMID: 39105460) Otolaryngol Head Neck Surg. 2025 Feb;172(2):734-743. (PMID: 39666770) J Clin Med. 2025 Jan 28;14(3):. (PMID: 39941547) J Med Syst. 2025 Feb 27;49(1):32. (PMID: 40011323) BMJ. 2022 May 18;377:e070904. (PMID: 35584845) Eur Arch Otorhinolaryngol. 2026 Apr;283(4):2227-2236. (PMID: 41430450) JAMA Netw Open. 2025 Feb 03;8(2):e2457879. (PMID: 39903463) JMIR Form Res. 2025 Nov 28;9:e79534. (PMID: 41313805) IEEE J Biomed Health Inform. 2022 Jul;26(7):2974-2982. (PMID: 35290196) NPJ Digit Med. 2026 Mar 15;9(1):. (PMID: 41832240) Trends Hear. 2024 Jan-Dec;28:23312165241260041. (PMID: 38870447) OTO Open. 2024 Dec 11;8(4):e70048. (PMID: 39664064) BMC Med. 2025 Aug 1;23(1):447. (PMID: 40745595) Am J Audiol. 2020 Sep 3;29(3):513-527. (PMID: 32551926) Am J Audiol. 2026 Jun 2;35(2):652-667. (PMID: 41949919) BMC Med Educ. 2026 Apr 23;26(1):. (PMID: 42026543) Otol Neurotol. 2026 Jan 1;47(1):64-69. (PMID: 41126436) |
| Contributed Indexing: | Keywords: Artificial intelligence; Clinical decision support; Large language models; Patient communication; Pure-tone audiometry; Tympanometry |
| Entry Date(s): | Date Created: 20261001 Date Completed: 20261001 Latest Revision: 20261004 |
| Update Code: | 20261004 |
| PubMed Central ID: | PMC13630873 |
| DOI: | 10.1007/s10916-026-02463-5 |
| PMID: | 42821199 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1573-689X |
|---|---|
| DOI: | 10.1007/s10916-026-02463-5 |