Bo Li1, Changqing Jiang1, Luming Li1,2, Jianguo Zhang3, Dawei Meng3. 1. National Engineering Laboratory for Neuromodulation, School of Aerospace, Tsinghua University, Beijing, China. 2. Center of Epilepsy, Beijing Institute for Brain Disorders, Beijing, China. 3. Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China.
Abstract
OBJECTIVE: In the treatment of Parkinson's disease for deep brain stimulation (DBS), the subthalamic nucleus (STN) is the most important target on a specific brain nucleus. Although procedural details are well established, targeting STN remains problematic because of its variable location and relatively small size. MATERIALS AND METHODS: Data were collected from 10 patients with Parkinson's disease implanted with deep brain stimulation devices. This paper presents an automated algorithm for 3.0T magnetic resonance (MR) image segmentation using the level set method to reconstruct the STN based on automatic segmentation. Implicit polynomial surfaces are used for the reconstruction of the STN segmentation. RESULTS: The method was applied to 10 Parkinson's disease (PD) patients to automatically extract and rebuild the STN. A comparison of the Euclidean distances and dice overlap coefficient showed no significant differences with the segmentation-based method, with the present method having smaller prediction errors and being more robust than expert systems. CONCLUSIONS: This paper presents an automated algorithm to segment and reconstruct the small human STN using MR images. This method for STN should provide an effective method for advancing STN localization and direct visualization.
OBJECTIVE: In the treatment of Parkinson's disease for deep brain stimulation (DBS), the subthalamic nucleus (STN) is the most important target on a specific brain nucleus. Although procedural details are well established, targeting STN remains problematic because of its variable location and relatively small size. MATERIALS AND METHODS: Data were collected from 10 patients with Parkinson's disease implanted with deep brain stimulation devices. This paper presents an automated algorithm for 3.0T magnetic resonance (MR) image segmentation using the level set method to reconstruct the STN based on automatic segmentation. Implicit polynomial surfaces are used for the reconstruction of the STN segmentation. RESULTS: The method was applied to 10 Parkinson's disease (PD) patients to automatically extract and rebuild the STN. A comparison of the Euclidean distances and dice overlap coefficient showed no significant differences with the segmentation-based method, with the present method having smaller prediction errors and being more robust than expert systems. CONCLUSIONS: This paper presents an automated algorithm to segment and reconstruct the small human STN using MR images. This method for STN should provide an effective method for advancing STN localization and direct visualization.
Authors: Vincent Beliveau; Martin Nørgaard; Christoph Birkl; Klaus Seppi; Christoph Scherfler Journal: Hum Brain Mapp Date: 2021-07-29 Impact factor: 5.038