大型语言模型是否可以检测前牙中的牙周病变? 一个比较研究
Diulia Pereira Bubna1, Natanael Henrique Ribeiro Mattos1, Luana Beatriz das Portas Luiz1
1Tuiuti University of Paraná, Curitiba, PR, Brazil.
Brazilian dental journal
|January 15, 2026
概括
像ChatGPT 5.0和Gemini Flash 2.0这样的大型语言模型 (LLM) 显示出高灵敏度,但在X射线图上检测周周病变时的特异性较差. 这些通用AI工具还不适合用于放射性诊断应用.
科学领域:
- 人工智能在牙科中的应用
- 放射性诊断 放射性诊断 放射性诊断 放射性诊断
- 医学成像分析 医学成像分析
背景情况:
- 大型语言模型 (LLM) 越来越多地被用于医学应用.
- 周围牙病变需要精确的放射检测,以便及时进行内牙治疗.
- 评估牙科LLM的诊断能力至关重要.
研究的目的:
- 评估ChatGPT 5.0和Gemini Flash 2.0的诊断性能,用于在周围放射图上识别周围损伤.
- 为了比较这两种LLM的准确性,敏感性和特异性.
- 确定通用LLM适用于牙周内科的放射诊断任务的适用性.
主要方法:
- 通过使用标准化提示器,通过ChatGPT 5.0和Gemini Flash 2.0分析了75张周周放射 (带有和没有病变).
- 一位经过校准的内牙专家提供了参考诊断.
- 包括平衡精度,灵敏度,特异性和F1分数在内的性能指标被计算并使用McNemar测试进行比较.
主要成果:
- 聊天GPT 5.0表现出比双子闪 2.0 (0.48) 更高的平衡精度 (0.54),具有更高的灵敏度 (0.97 与 0.84 相比).
- 两种模型的特异性都很差 (两种模型的特异性都为0.11),错误阳性率很高.
- 模型之间发现了显著的差异 (p < 0.05),支持ChatGPT 5.0,但两者都在牙本地化方面遇到了困难.
结论:
- 虽然LLM在检测周周病变方面表现出高灵敏度,但它们的低特异性限制了它们的诊断效用.
- 一般用途的LLM目前对积极分类有偏见,并且在内牙科的放射性诊断中不可靠.
- 需要进一步开发以提高牙科放射学解释的AI模型的准确性和特异性.
相关概念视频
01:15Teeth
The formation of teeth, also known as odontogenesis, is a complex process that begins in utero, around the sixth week of embryonic development. There are three stages to this process: the bud stage, the cap stage, and the bell stage.
In the bud stage, the tooth germ (an aggregation of cells) starts to form in the developing jawbone. During the cap stage, the tooth germ differentiates into enamel organ, dental papilla, and dental sac, which will later develop into the tooth's enamel, dentin and...
In the bud stage, the tooth germ (an aggregation of cells) starts to form in the developing jawbone. During the cap stage, the tooth germ differentiates into enamel organ, dental papilla, and dental sac, which will later develop into the tooth's enamel, dentin and...
01:21Tooth Anatomy
The human tooth enables us to eat a variety of foods, speak clearly, and even aid in shaping our faces. Teeth are composed of various elements that work together. Here's a detailed look at the anatomy of a human tooth.
The Crown, Neck, and Root
The visible part of the tooth is referred to as the crown. It's covered by enamel, the hardest substance in the human body. The crown is uniquely shaped for each type of tooth, allowing for different functions such as cutting, tearing, or grinding food.
The Crown, Neck, and Root
The visible part of the tooth is referred to as the crown. It's covered by enamel, the hardest substance in the human body. The crown is uniquely shaped for each type of tooth, allowing for different functions such as cutting, tearing, or grinding food.
01:26Assessment of the Mouth
A thorough mouth assessment, including inspection and palpation of the lips, gums, tongue, tonsils, uvula, and pharynx, is crucial in detecting potential health issues. Diseases ranging from oral cancer to systemic conditions like diabetes could be identified early through careful oral examination. This article provides a detailed guide on conducting a comprehensive mouth assessment.
Mouth Inspection
The inspection begins with visually examining the mouth for symmetry, color, and size.
Mouth Inspection
The inspection begins with visually examining the mouth for symmetry, color, and size.