Terminologia Anatomica Compliance in Generative Artificial Intelligence Outputs: A Structured Narrative Review and Terminology-Specific Verification Checklist.
- 1Basic Science, Lake Erie College of Osteopathic Medicine, Erie, USA.
- 2Medicine, St. George's University, St. George's, GRD.
- 3Anatomy, Tanta University Hospital, Tanta, EGY.
Abstract
Generative artificial intelligence (GenAI) can produce fluent anatomical explanations and persuasive illustrations, but fluency and visual realism do not establish compliance with Terminologia Anatomica, second edition (TA2), or anatomical correctness. Unlike broader anatomy-education governance frameworks, this evidence-informed narrative review isolates terminology compliance as a distinct validation problem. It examines six connected domains: baseline human and editorial use of standardized terminology, text-based GenAI performance, image-generation fidelity, citation reliability, multilingual terminology, and publisher oversight. Direct evidence is heterogeneous but increasingly specific. Recent studies have used dedicated terminology measures or TA2 as a reference standard for text outputs, while image studies consistently identify omissions, invented structures, incorrect relationships, and mislabeled anatomy. One anatomy-specific citation study also found substantial reference hallucination and limited PubMed identifier accuracy. These risks occur against a pre-existing background of incomplete TA adoption among anatomists and journal editors, so GenAI may reproduce or amplify inconsistencies already present in its source environment. Evidence concerning non-English TA2 output remains particularly limited. A four-step terminology-specific checklist is therefore proposed: classify the output and its risk; verify every anatomical term against an authoritative terminology source; verify scientific content, structural fidelity, and source traceability; and document accountable human review and disclose AI assistance. Verification should be tiered by consequence: low for internal brainstorming, moderate for teaching materials, and high for summative assessment, publication, and clinical or procedural use. The checklist is intentionally practical rather than a validated scoring instrument. Current evidence does not support treating any general-purpose GenAI system as reliably TA2-compliant without human verification.
Related
- Making Generative AI Use Visible in a Foundational Health-Sciences Course: Effects on Learning, Verification and Attribution
- Generative Artificial Intelligence Models in Breast Cancer Diagnosis: A Systematic Review
- Nursing students’ experiences with using generative artificial intelligence: a descriptive qualitative study
- Generative Artificial Intelligence in Periodontology: Application of OpenAI Models for the Classification of Periodontitis and Risk Assessment in Patients Undergoing Supportive Periodontal Care
- Simulating Psychotherapeutic Sessions with Generative Artificial Intelligence: A Proof-of-Concept Study of In Silico Psychotherapy Research
BibTeX
@article{sharma2026terminologia,
title = {{Terminologia Anatomica Compliance in Generative Artificial Intelligence Outputs: A Structured Narrative Review and Terminology-Specific Verification Checklist.}},
author = {Deepak Sharma and Anaida Singh and Ahmed Mahgoub},
journal = {Cureus},
year = {2026},
month = sep,
doi = {10.7759/cureus.116639},
url = {https://europepmc.org/article/MED/42910208},
note = {CC BY},
}
RIS
TY - JOUR
TI - Terminologia Anatomica Compliance in Generative Artificial Intelligence Outputs: A Structured Narrative Review and Terminology-Specific Verification Checklist.
AU - Deepak Sharma
AU - Anaida Singh
AU - Ahmed Mahgoub
PY - 2026
DA - 2026/09/21
JO - Cureus
DO - 10.7759/cureus.116639
UR - https://europepmc.org/article/MED/42910208
AB - Generative artificial intelligence (GenAI) can produce fluent anatomical explanations and persuasive illustrations, but fluency and visual realism do not establish compliance with Terminologia Anatomica, second edition (TA2), or anatomical correctness. Unlike broader anatomy-education governance frameworks, this evidence-informed narrative review isolates terminology compliance as a distinct validation problem. It examines six connected domains: baseline human and editorial use of standardized terminology, text-based GenAI performance, image-generation fidelity, citation reliability, multilingual terminology, and publisher oversight. Direct evidence is heterogeneous but increasingly specific. Recent studies have used dedicated terminology measures or TA2 as a reference standard for text outputs, while image studies consistently identify omissions, invented structures, incorrect relationships, and mislabeled anatomy. One anatomy-specific citation study also found substantial reference hallucination and limited PubMed identifier accuracy. These risks occur against a pre-existing background of incomplete TA adoption among anatomists and journal editors, so GenAI may reproduce or amplify inconsistencies already present in its source environment. Evidence concerning non-English TA2 output remains particularly limited. A four-step terminology-specific checklist is therefore proposed: classify the output and its risk; verify every anatomical term against an authoritative terminology source; verify scientific content, structural fidelity, and source traceability; and document accountable human review and disclose AI assistance. Verification should be tiered by consequence: low for internal brainstorming, moderate for teaching materials, and high for summative assessment, publication, and clinical or procedural use. The checklist is intentionally practical rather than a validated scoring instrument. Current evidence does not support treating any general-purpose GenAI system as reliably TA2-compliant without human verification.
SN - 2168-8184
ER -