Objective Analysis of Neck Muscle Boundaries for Cervical Dystonia Using Ultrasound Imaging and Deep Learning

Abstract

To provide objective visualization and pattern analysis of neck muscle boundaries to inform and monitor treatment of cervical dystonia. Methods: We recorded transverse cervical ultrasound (US) images and whole-body motion analysis of sixty-one standing participants (35 cervical dystonia, 26 age matched controls). We manually annotated 3,272 US images sampling posture and the functional range of pitch, yaw, and roll head movements. Using previously validated methods, we used 60-fold cross validation to train, validate and test a deep neural network (U-net) to classify pixels to 13 categories (five paired neck muscles, skin, ligamentum nuchae, vertebra). For all participants for their normal standing posture, we segmented US images and classified condition (Dystonia/Control), sex and age (higher/lower) from segment boundaries. We performed an explanatory, visualization analysis of dystonia muscle-boundaries. Results: For all segments, agreement with manual labels was Dice Coefficient (64 ± 21%) and Hausdorff Distance (5.7 ± 4 mm). For deep muscle layers, boundaries predicted central injection sites with average precision 94 ± 3%. Using leave-one-out cross-validation, a support-vectormachine classified condition, sex, and age from predicted muscle boundaries at accuracy 70.5%, 67.2%, 52.4%

Will Artificial Intelligence Outperform the Clinical Neurologist in the Near Future? Yes

“…However, in some areas such as advanced neuroimaging or technology‐based monitoring, AI may advance clinical reasoning in the interpretation. The benefits of AI have recently been demonstrated in other fields of movement disorders beyond PD as shown in cervical dystonia when deep learning was used to classify it into 3 types and automatically segmenting individual muscles for targeted injections 26 . No matter whether it is AI, robotics, or augmented or virtual reality, we should accept that these advances are going to have a massive influence on the way we work.…”

The Role of Ultrasound for the Personalized Botulinum Toxin Treatment of Cervical Dystonia

“…Recently, a deep-learning algorithm has been applied to US images of neck muscles obtained in CD patients and controls. The authors describe distinct subtypes of CD based on ultrasonographic patterns of neck muscle shape [24]. US based real-time shear wave elastography has been demonstrated to identify the most hypertensive muscles in patients with CD [25].…”

A Review of the Challenges in Deep Learning for Skeletal and Smooth Muscle Ultrasound Images

“…Based on ten studies on deep learning on ultrasound muscle images, it was found that the U-Net and CNN architectures were the most used network architectures in segmentation [34,35,[37][38][39]. Meanwhile, the RAN was the latest development of the region-based convolutional neural network (RCNN) [36].…”

“…Additionally, it supported the identification of landmarks such as segmentation in the orientation of muscle fibers [33] and tracking the cross-sectional area of the rectus femoris [34]. In addition, deep learning methods have visualized the neck muscle pattern landmark and muscle-tendon landmark [35,36]. Besides, deep learning has assisted measurement and tracking in the urogenital hiatus and puborectalis muscle [37].…”