用视觉变压器和CNN进行肺质子和高极化气体MRI的强有力的细分:人工噪声下的性能比较分析
Ramtin Babaeipour1, Matthew S Fox2,3,4, Grace Parraga1,4,5
1School of Biomedical Engineering, Faculty of Engineering, The University of Western Ontario, London, ON N6A 3K7, Canada.
Bioengineering (Basel, Switzerland)
|August 28, 2025
概括
与传统的卷积神经网络 (CNN) 相比,视觉转换器 (ViT) 模型在与噪音相关的医疗图像细分方面表现出卓越的性能. 这种强度对于高极化气体MRI等应用至关重要,在低信号噪声比环境中提高诊断准确性.
科学领域:
- 医学成像和诊断
- 医疗保健中的人工智能
- 生物医学工程
背景情况:
- 精确的医学图像细分对于疾病诊断和监测至关重要,特别是在使用质子和高极化气体MRI的肺部成像中.
- 在超极化气体MRI中常见的图像噪声和文物,由于呼吸阻断,挑战了传统的细分算法.
- 深度学习模型提供了潜在的解决方案,但它们对不同噪音水平的稳定性需要彻底评估.
研究的目的:
- 评估深度学习细分模型的稳定性,特别是卷积神经网络 (CNN) 和视觉变换器 (ViT),在不同的高斯噪声水平下.
- 将基于CNN和基于ViT的架构在分段质子和超极化气体MRI数据中的性能进行比较.
- 在低信号噪声比 (SNR) 医学成像环境中确定最适合的深度学习模型.
主要方法:
- 使用CNN (VGG16,VGG19,ResNet152) 和ViT (MiT-B0,B3,B5) 骨架进行训练和测试的特征金字塔网络 (FPN) 和U-Net细分架构.
- 使用了56名参与者的质子和高极化气体MRI片.
- 通过使用子得分和边界误差等指标,评估不同级别的高斯噪声的模型性能.
主要成果:
- 在所有测试的指标和噪音条件下,基于ViT的模型,特别是具有SegFormer骨干的模型,始终优于基于CNN的模型.
- 在高噪音水平下,ViT模型的性能优势最为显著,表现出卓越的Dice分数和减少的边界误差.
- 与传统的CNN架构相比,变压器模型对图像退化具有更强的弹性.
结论:
- 基于ViT的深度学习架构在噪音的情况下显示出卓越的稳定性和准确性.
- 这些发现支持在临床上相关的低SNR环境中部署ViT模型,如高极化气体MRI,提高分段可靠性.
- 这项研究强调了变压器模型在具有挑战性的医学成像场景中克服传统CNN的局限性.
相关概念视频
01:24Magnetic Resonance Imaging
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...
01:30Radiological Investigation II: MRI and Ventilation Perfusion Scan
Description
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...
Magnetic Resonance Imaging (MRI) and Ventilation Perfusion Scans are two radiological investigations that offer detailed diagnostic images of the body, particularly lung structures.
MRI
MRI uses magnetic fields and radiofrequency signals to distinguish between normal and abnormal tissues. This technology provides a more detailed diagnostic image than CT scans, enabling it to characterize pulmonary nodules, stage bronchogenic carcinoma, and evaluate inflammatory activity in...
01:21Imaging Studies for Cardiovascular System IV: CMRI
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,...