卷积神经网络用于自动MR分类心肌铁过载在thalassemia主要患者
Vincenzo Positano1,2, Antonella Meloni3,4, Lisa Anita De Santi3,5
1Bioengineering Unit, Fondazione G. Monasterio CNR-Regione Toscana, Pisa, Italy. positano@ftgm.it.
European radiology
|December 10, 2024
概括
深度学习模型使用T2*MRI扫描准确地分类心肌铁过载 (MIO). 这些卷积神经网络 (CNN) 提供可靠的,自动化的MIO分期,与专家放射科医生相美.
科学领域:
- 心血管成像 - 心血管成像
- 人工智能在医学中的应用
- 血液学 血液学 血液学
背景情况:
- 心脏磁共振 (MR) T2*成像是评估心肌铁过载 (MIO) 的标准.
- 准确的MIO评估对于管理患有thalassemia major和其他血液学疾病的患者至关重要.
- 质量控制和MIO解释的一致性可能具有挑战性,特别是在经验较少的中心.
研究的目的:
- 开发和验证深度学习模型,用于使用心脏T2*多回声MRI图像自动分类MIO水平.
- 将这些模型的性能与已建立的临床评估方法和观察者间变异性进行比较.
主要方法:
- 开发二维卷积神经网络 (CNN):MS-HippoNet用于多切片和SS-HippoNet用于单切片分析.
- 培训和5倍交叉验证823个心脏T2*MRI图像来自496名患有thalassemia major的患者,标记为MIO严重程度 (正常,中度,严重).
- 使用多类准确度,灵敏度,特异性和相互协议的卡帕统计数据进行性能评估,与放射科医生观察者间协议相比较.
主要成果:
- 在测试组中,MS-HippoNet 实现了 0.885 的多类精度,MS-HippoNet 实现了 0.836.
- 外部验证证实了网络性能,多类精度为0.827 (MS) 和0.793 (SS).
- 自动分类和实地真相之间的一致性很好 (MS的κ=0.771,SS的κ=0.614),相当于观察者之间的一致性 (MS的κ=0.872,SS的κ=0.907).
结论:
- 深度学习模型在从T2*MR图像中分类MIO水平方面表现出强的表现.
- 使用CNN的自动MIO分期提供了一个可靠的替代方案,与标准临床程序相提并论.
- 这些CNN可以在患有血液学疾病的患者中促进非侵入性,自动评估心脏铁过载.
相关概念视频
Magnetic Resonance Imaging
4.9K
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
4.9K
Imaging Studies for Cardiovascular System IV: CMRI
5
Cardiovascular magnetic resonance imaging, or CMRI, is a non-invasive diagnostic test that employs a magnetic field and radiofrequency waves to create precise images of the heart and arteries. It provides comprehensive information about cardiac anatomy, function, perfusion, and tissue characterization without ionizing radiation.IndicationsCMRI diagnoses various heart conditions, including tissue damage from heart attacks, ischemic heart disease, myocarditis, aortic issues (tears, aneurysms,...
5


