Suppr超能文献

人工智能驱动的龋齿检测的伦理见解:一项范围综述。

Ethical insights into AI-driven caries detection: a scoping review.

作者信息

Yousuf Tahoora, Khan Madiha, Ghafoor Robia

机构信息

Operative Dentistry & Endodontics, Department of Surgery, Aga Khan University Hospital, Karachi, Pakistan.

出版信息

BDJ Open. 2025 Sep 9;11(1):78. doi: 10.1038/s41405-025-00366-0.

Abstract

BACKGROUND

Artificial Intelligence (AI) has become increasingly integrated into dental diagnostics, particularly for detecting carious lesions. While AI offers benefits such as improved accuracy and efficiency, its use raises important ethical concerns, including transparency, patient privacy, autonomy, diversity and accountability. This scoping review aims to identify these ethical concerns using a structured ethical framework.

METHODOLOGY

Three databases were searched for papers regarding caries detection using AI. An established ethical framework was used to screen each paper for potential areas of concern.

RESULTS

A total of 351 abstracts were screened, of which 7 articles were included in this review. Each article was screened for established ethical principles including transparency, diversity, wellness, autonomy, privacy, accountability, equity, prudence, sustainable development, solidarity and governance. Diversity was the main ethical concern. Concerns related to accountability, equity and transparency were identified in 2 of the articles whereas ethical issue of privacy was identified in 4 of the articles. Only one study mentioned that no ethical approval was taken prior to commencement of study.

CONCLUSION

AI in caries detection faces ethical issues like data bias, privacy risks, and equity concerns, potentially leading to flawed AI models. These issues can be addressed by creating a more specialized ethical framework that is specific to AI in dentistry.

CLINICAL RELEVANCE

Understanding ethical challenges in AI-driven caries detection is critical to ensure accurate diagnostics, maintain patient trust, protect privacy, and support informed decision-making. Clinicians must be equipped to navigate these challenges as AI tools become more prevalent in dental practice.

摘要

背景

人工智能(AI)已越来越多地融入牙科诊断中,尤其是用于检测龋损。虽然人工智能具有提高准确性和效率等优点,但其应用引发了重要的伦理问题,包括透明度、患者隐私、自主性、多样性和问责制。本范围综述旨在使用结构化伦理框架识别这些伦理问题。

方法

检索了三个数据库中有关使用人工智能进行龋病检测的论文。使用既定的伦理框架筛选每篇论文,以确定潜在的关注领域。

结果

共筛选了351篇摘要,其中7篇文章纳入本综述。每篇文章都根据既定的伦理原则进行筛选,包括透明度、多样性、健康、自主性、隐私、问责制、公平、审慎、可持续发展、团结和治理。多样性是主要的伦理关注点。在2篇文章中发现了与问责制、公平和透明度相关的问题,而在4篇文章中发现了隐私伦理问题。只有一项研究提到在研究开始前未获得伦理批准。

结论

龋病检测中的人工智能面临数据偏差、隐私风险和公平问题等伦理问题,可能导致人工智能模型存在缺陷。这些问题可以通过创建一个更专门针对牙科人工智能的伦理框架来解决。

临床相关性

了解人工智能驱动的龋病检测中的伦理挑战对于确保准确诊断、维持患者信任、保护隐私和支持明智决策至关重要。随着人工智能工具在牙科实践中越来越普遍,临床医生必须有能力应对这些挑战。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/81a8/12420780/27bd844aea70/41405_2025_366_Fig1_HTML.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验