Original Research ARTICLE
Deep learning-based deep brain stimulation targeting and clinical applications
- 1Department of Neurosurgery, Seoul Metropolitan Government Seoul National University Boramae Medical Center, South Korea
- 2Department of Neurosurgery, Seoul National University College of Medicine, South Korea
- 3School of Medicine, Inha University, South Korea
- 4Department of Bio-convergence Engineering, College of Health Science, Korea University, South Korea
- 5Department of Neurology, Gangneung Asan Hospital, South Korea
- 6Department of Neurology, Asan Medical Center, South Korea
- 7Department of Neurosurgery, Asan Medical Center, South Korea
Background: The purpose of the present study was to evaluate deep learning-based image-guided surgical planning for deep brain stimulation. We developed deep learning semantic segmentation-based deep brain stimulation targeting and prospectively applied the method clinically.
Methods: T2* fast gradient-echo images from 102 patients were used for training and validation. Manually drawn ground truth information was prepared for the subthalamic and red nuclei with an axial cut ~4 mm below the anterior–posterior commissure line. A fully convolutional neural network (FCN-VGG-16) was used to ensure margin identification by semantic segmentation. Image contrast augmentation was performed nine times. Up to 102 original images and 918 augmented images were used for training and validation. The accuracy of semantic segmentation was measured in terms of mean accuracy and mean intersection over the union. Targets were calculated based on their relative distance from these segmented anatomical structures considering the Bejjani target.
Results: Mean accuracies and mean intersection over the union values were high: 0.904 and 0.813, respectively, for the 62 training images, and 0.911 and 0.821, respectively, for the 558 augmented training images when 360 augmented validation images were used. The dice coefficient converted from the intersection over the union was 0.902 when 720 training and 198 validation images were used. Semantic segmentation was adaptive to high anatomical variations of size, shapes, and asymmetry. Concerning clinical applications, two patients were assessed: one with essential tremor and another with bradykinesia and gait disturbance due to Parkinson’s disease. Both improved without complications after surgery, and microelectrode recordings showed subthalamic nuclei signals in the latter patient.
Conclusions: Deep learning-based semantic segmentation accuracy may surpass previous methods. Deep brain stimulation targeting and its clinical application were possible using accurate, deep learning-based semantic segmentation, which is adaptive to anatomical variations.
Keywords: deep learning, Deep Brain Stimulation, Convolutional Neural Network, Semantic segmentation, clinical applications
Received: 27 Apr 2019;
Accepted: 04 Oct 2019.
Copyright: © 2019 Park, Cha, Lee, Jang, Lee and Lee. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Prof. Seong-Cheol Park, Department of Neurosurgery, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul, South Korea, email@example.com