The aim of study is to determine if deep learning enables MR Neuropathy of peripheral nerves in shorter scan times without compromising data quality.

研究的目的是确定深度学习是否能够在更短的扫描时间内实现周围神经的 MR 神经病而不影响数据质量。

• 批准号: 2888464
• 金额: --
• 依托单位: University of Sheffield
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2888464
• 关键词: aim; study; determine; deep; learning

The overall area is development of imaging methodology and software in relation to advanced PET/MRI. It falls within the AI technologies and healthcare technologies areas of EPSRC remit. The application area is detection of neuropathy of the peripheral nerves, but the work proposed in this project is method development and optimization with respect to imaging, including use of deep learning techniques. The aim is to use new deep learning methods to push the spatial resolution boundaries for model-based diffusion-tensor imaging, a technique that can give fairly specific and non-invasive information on the microstructure of individual fibre bundles. There is currently a technical barrier with respect to low signal levels as used in established approaches. In other application areas such as spinal cord & brain there is emerging evidence that incorporating prior knowledge in the image reconstruction process using deep learning techniques can help to overcome this barrier. If successful, it will be possible to test the clinical potential of the advances in a patient cohort.

整个领域是与先进的PET/MRI相关的成像方法和软件的开发。它福尔斯EPSRC职权范围内的AI技术和医疗保健技术领域。应用领域是周围神经病变的检测，但该项目中提出的工作是关于成像的方法开发和优化，包括使用深度学习技术。其目的是使用新的深度学习方法来推动基于模型的扩散张量成像的空间分辨率边界，这种技术可以提供关于单个纤维束微观结构的相当具体和非侵入性的信息。目前，在已建立的方法中使用的低信号电平方面存在技术障碍。在其他应用领域，如脊髓和大脑，有新的证据表明，使用深度学习技术在图像重建过程中结合先验知识可以帮助克服这一障碍。如果成功，将有可能在患者队列中测试这些进展的临床潜力。