在Takotsubo综合征患者中,从心房平面移位进行深度学习:点亮黑子
Fahim Zaman1, Nicholas Isom2, Amanda Chang2
1Department of Electrical and Computer Engineering, University of Iowa, 103 S. Capitol St., 3318 SC, Iowa City, IA 52242, USA.
一个深度学习模型通过专注于心脏的底部,准确地识别了Takotsubo综合征 (TTS),与专注于心脏顶部的人类专家不同. 这种人工智能方法揭示了TTS病理生理学的新见解.
科学领域:
- 心脏病学 心脏病学
- 人工智能的人工智能
- 医疗成像医学成像
背景情况:
- 塔科苏博综合征 (TTS) 诊断可能具有挑战性,可能会被人类读者误解.
- 时空深卷积神经网络 (DCNNs) 是一个潜在的工具,可以帮助对TTS进行心声学评估.
研究的目的:
- 为了提高空间时空DCNN的解释性,用于识别TTS中的潜在成像特征.
- 将基于DCNN的可视化与人类专家的解释进行比较,以区分TTS与ST段升高心肌梗塞 (STEMI).
主要方法:
- 使用梯度加权类激活映射可视化时空DCNN,应用于150名TTS和150名STEMI患者的心声录像.
- 人类专家读者的解释与DCNN突出地图进行了比较.
- 进行了光学流量,心肌应变和多普勒心声图,以分析区域心肌力学.
主要成果:
- DCNN的突出焦点集中在心房 (AV) 平面上,与人类读者的顶点焦点形成鲜明对比.
- 与STEMI患者相比,TTS患者表现出显著较弱的基底内和前侧纵向位移和应变.
- 经过TTS的患者表现出缩力学受损,由改变的E'/A比率和E/E'值表明.
结论:
- 时空DCNN突出映射有助于理解心肌时间动力学和量化TTS中的区域力学.
- 降低TTS患者的AV平面位移与扩张功能受损有关,提供了一个新的诊断标志物.
更多相关视频
08:10Estimating Bilateral Atrial Function by Cardiovascular Magnetic Resonance Feature Tracking in Patients with Paroxysmal Atrial Fibrillation
Published on: July 20, 2022
08:19Transthoracic Echocardiography to Assess Post-Resuscitation Left Ventricular Dysfunction After Acute Myocardial Infarction and Cardiac Arrest in Pigs
Published on: July 12, 2022
相关概念视频
Electrophysiology of Normal Cardiac Rhythm
Disturbances in Heart Rhythm
Arrhythmias are categorized by their speed, rhythm, and origin. A slow heart...
Imbalances in Cardiac Output
CHF can occur due to the failure of either side of the heart. Left-side failure leads to pulmonary congestion—the right side continues to send blood...
Cardiac Action Potential
The cardiac action potential process involves a series of phases characterized by the movement of ions across the cardiac cell membranes, leading to the depolarization and repolarization of the cardiac myocytes.
Ionic Basis of Cardiac Action Potentials
Cardiomyopathy II: Dilated Cardiomyopathy
ECG Interpretation of Arrhythmias II: Atrial, Junctional and Ventricular Arrhythmias
