相关实验视频

Updated: May 1, 2026

Genetic Analysis of Hereditary Transthyretin Ala97Ser Related Amyloidosis

Genetic Analysis of Hereditary Transthyretin Ala97Ser Related Amyloidosis

Published on: June 9, 2018

评估预测模型的性能和潜在偏差,以检测心脏氨基粉症的Transthyretin

Jonathan Hourmozdi¹, Nicholas Easton¹, Simon Benigeri¹

¹Division of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA; Bluhm Cardiovascular Institute Center for AI, Northwestern Medicine, Chicago, Illinois, USA.

|July 5, 2025

概括

与其他算法相比,深度学习回声模型在检测跨氨基酸性心肌病 (ATTR-CM) 中表现优越. 这些人工智能工具为心力衰竭患者早期ATTR-CM诊断提供了一个有希望的,低偏差的方法.

科学领域:

心脏病学心脏病学
人工智能的人工智能
医学诊断医学诊断医学诊断

背景情况:

延迟诊断转氨胺类心肌病变 (ATTR-CM) 会增加患者的发病率,特别是在新的疾病修饰疗法时.
选算法,包括人工智能,可能加速ATTR-CM诊断,但直接比较是缺乏的.

研究的目的:

为了比较四个不同的ATTR-CM检测算法的诊断性能.
在心力衰竭人群中评估潜在的模型偏差和相关风险.

主要方法:

利用综合卫生系统数据库 (2010-2022) 识别ATTR-CM病例并匹配心力衰竭控制.
对比了基于索赔的随机森林模型,基于回归的得分 (Mayo ATTR-CM) 和两个深度学习回声心脏学模型 (EchoNet-LVH,EchoGo Amyloidosis).
使用标准公平度指标评估算法偏差.

主要成果:

这项研究包括176例ATTR-CM病例和3192例对照;9.0%被确定为黑人.
深度学习的回声模型 (EchoNet-LVH AUC:0.88,EchoGo Amyloidosis AUC:0.92) 的表现优于梅奥的ATTR-CM评分 (AUC:0.79).
胡达人和其他人. 模型表现不佳 (AUC:0.49);偏见审计符合黑人患者机会平等的公平标准.

结论:

深度学习,基于心声回声的模型在外部验证中证明了ATTR-CM检测的优越歧视.
这些由人工智能驱动的模型由于种族偏见而具有较低的伤害风险,从而促进及时诊断.

关键词:

氨基粉症是什么?人工智能的人工智能是人工智能.医疗保健的差异,医疗保健的差异心脏衰竭是因为心脏衰竭.机器学习是机器学习.

更多相关视频

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Identification and Classification of Position-specific GABAA Receptor Subunit Missense Variants for Their Role In Hippocampal Pyramidal Neurons

Identification and Classification of Position-specific GABA_A Receptor Subunit Missense Variants for Their Role In Hippocampal Pyramidal Neurons

Published on: June 6, 2025

相关概念视频

Receiver Operating Characteristic Plot

Receiver Operating Characteristic Plot

A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...