人工智能建议的正确性对临床决策中的诊断准确性的影响
Florian Kücking1, Dorothee A Busch2, Mareike Przysucha1
1Research Center of Health and Social Informatics, Osnabrück University of Applied Sciences, Osnabrück, Germany.
人工智能建议显著影响临床决策,在正确时提高准确性,但在错误时降低准确性. 这凸显了需要高质量的AI系统和临床医师培训,以减轻与自动化偏差相关的风险.
科学领域:
- 医疗信息学 医疗信息学
- 医疗保健中的人工智能
- 临床决策支持系统 临床决策支持系统
背景情况:
- 临床决策依赖于医疗保健提供者的专业知识,但基于AI的临床决策支持系统 (CDSS) 引入了自动化偏见.
- 准确/不准确的人工智能建议与人类因素对诊断准确性的比较影响尚不清楚.
研究的目的:
- 研究AI建议对模拟临床场景中的诊断准确度的影响.
- 为了比较人工智能建议对医疗保健提供者相关因素的影响.
主要方法:
- 一项模拟诊断研究涉及223名医疗保健专业人员,他们评估了与AI建议 (正确/不正确) 的伤口蚀图像.
- 通用线性混合模型 (GLMM) 分析了决策准确性,考虑了AI正确性和提供商因素 (性能,资格,经验,信任,人口统计学).
主要成果:
- 正确的AI建议提高了决策准确度的十倍 (OR=10.0,p<0.001),而不正确的建议则降低了它.
- 高基线表现 (OR=2.44),资格 (OR=1.40),经验 (OR=1.89),女性性别 (OR=1.55) 与更高的准确性相关.
- 人工智能的整体效果是模两可的,但它的影响严重依赖于建议的正确性.
结论:
- 人工智能建议可以取代医疗保健提供者的判断,在正确的情况下显著提高准确性,但由于过度依赖而导致不正确时存在安全风险.
- 确保高质量的AI系统和培训临床医生批判性地评估AI输出对于安全的临床实践至关重要.
更多相关视频
12:18A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
06:11High-definition Transcranial Direct Current Stimulation over Right Dorsolateral Prefrontal Cortex to Enhance Metacognitive Sensitivity
Published on: September 26, 2025
相关概念视频
Documentation of Nursing Diagnosis
In some settings, data-driven computerized decision support systems are in place, allowing for more accurate nursing diagnoses. The database within one of these systems includes diagnostic labels defining characteristics, activities, and indicators for nursing. A nurse enters...
Receiver Operating Characteristic Plot
Sensitivity, Specificity, and Predicted Value
Sensitivity is the...
Errors occurring during blood pressure monitoring
Several factors...
Accuracy and Errors in Hypothesis Testing
In hypothesis testing, the probability of making a Type I error, denoted as α, is commonly set at 0.05. This significance level indicates a 5%...
Issues And Trends In Healthcare Delivery System
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
