Enhancing clinical brain MRI scans with deep learning for improved deep brain stimulation

1)简要描述研究背景，包括潜在影响脑深部电刺激（DBS）是一种用于治疗多种神经系统疾病的外科手术，如帕金森病和癫痫。这个过程包括在大脑中植入电极，以刺激大脑的某些区域。要做到这一点，外科医生首先需要准确定位要刺激的区域。超高场核磁共振成像产生高质量的脑部扫描，大大减轻了这项任务。然而，超高场扫描仪并没有广泛使用。因此，如果能够从标准MRI扫描仪获得更高质量的扫描，则可以改善对DBS的访问。2)目的和目标该项目的目的是：-开发一种人工智能系统，用于增强标准临床MRI图像，使用超高场扫描仪的数据-提高使用标准临床MRI扫描定位DBS大脑区域时的准确性3)研究方法的新颖性该项目方法学的新颖性在于利用最先进的深度学习方法从高质量的MRI数据中增强低质量的临床MRI数据。4)与EPSRC的战略和研究领域保持一致该项目与EPSRC的“医疗技术”主题保持一致，特别是解决其挑战3 -“发现和加速新干预措施的发展”-因为它寻求开发一种新方法来改善图像引导治疗的结果。5)任何参与的公司或合作者该项目由英国大脑研究中心共同资助

1) Brief description of the context of the research including potential impactDeep brain stimulation (DBS) is a surgical procedure used to treat several neurological conditions, such as Parkinson's disease and epilepsy. The procedure involves implanting electrodes in the brain in order to stimulate certain brain regions. To do this, surgeons first need to accurately locate the regions to be stimulated. Ultra-high field MRI produces high-quality brain scans which significantly ease this task. However, ultra-high field scanners are not widely available. Access to DBS could thus be improved if it was possible to obtain higher quality scans from standard MRI scanners.2) Aims and ObjectivesThe aims of the project are:- To develop an AI system for enhancing standard clinical MRI images, using data from the ultra-high field scanners- To improve accuracy when locating brain regions for DBS using standard clinical MRI scans3) Novelty of Research MethodologyThe methodological novelty of this project lies in exploiting state-of-the-art deep learning approaches to enhance low-quality clinical MRI data from high-quality MRI data.4) Alignment to EPSRC's strategies and research areasThe project aligns with EPSRC's "Healthcare technologies" theme, addressing in particular its Challenge three - "Discovering and accelerating the development of new interventions" - as it seeks to develop a novel method to improve outcomes of image-guided treatments.5) Any companies or collaborators involvedThe project is co-funded by Brain Research UK