从低数PET生成临床前正子发射断层扫描 (PET) 图像的深度学习生成,并以基于任务的绩效评估为基础
Kaushik Dutta1,2, Richard Laforest1,2, Jingqin Luo3
1Mallinckrodt Institute of Radiology, Washington University School of Medicine, St Louis, Missouri, USA.
Medical physics
|May 6, 2024
概括
一种新的深度学习方法,即基于注意力的残余扩展网络 (ARD-Net),有效地从低数量的PET (LC-PET) 数据中生成标准数量的PET (SC-PET) 图像. 对于临床前成像,ARD-Net在图像保真度,细分和量化方面表现出卓越的性能.
科学领域:
- 医疗成像医学成像
- 人工智能在医学中的应用
- pozitron 发射断层扫描 (PET) 是一个技术.
背景情况:
- 临床前低数PET (LC-PET) 提供了改善后勤和吞吐量等优势.
- 然而,LC-PET的信号与噪声比较低,细分挑战和量化不确定性.
研究的目的:
- 开发和评估一种新的深度学习 (DL) 架构,即基于注意力的剩余扩展网络 (ARD-Net).
- 使用ARD-Net从LC-PET图像中生成标准计数PET (SC-PET) 图像.
- 使用定性,基于任务的细分和定量指标评估ARD-Net性能.
主要方法:
- 在临床前的[18F]-氧糖 (FDG) -PET/CT数据集上进行了ARD-Net的培训和验证.
- 通过将SC-PET图像的样本降低到原始计数的10%,5%,1.6%和0.8%来生成LC-PET图像.
- 性能与其他DL和非DL方法进行了基准测试,使用图像保真度,细分精度,SUV量化和放射学特征.
主要成果:
- 在各种计数级别中,ARD-Net在图像保真性 (SSIM,NRMSE) 方面显著优于基准方法.
- 与其他方法相比,ARD-Net在SUV平均和较低的变化中显示的绝对偏差中值低于5%.
- 来自ARD-Net图像的放射特征与真实SC-PET图像的一致性更高.
结论:
- ARD-Net提供了一个强大的框架,用于从LC-PET数据生成SC-PET图像.
- 与现有方法相比,ARD-Net在临床前PET成像的图像保真性,细分和量化方面表现出卓越的性能.
更多相关视频
相关概念视频
Positron Emission Tomography
4.2K
Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...
4.2K
Imaging Studies II: Positron Emission Tomography and Scintigraphy
111
Positron Emission Tomography (PET) is a medical imaging technique that provides crucial insights into the body's physiological functions at a molecular level. It is an indispensable resource for diagnosing, staging, and monitoring various illnesses, notably cancer, neurological disorders, and cardiovascular conditions.
Fundamental Principles of PET
Fundamental Principles of PET
111


