Deep learning techniques such as convolutional neural networks (cnns) can potentially provide powerful tools for classifying, identifying, and predicting patterns in climate and environmental data. However, because of the inherent complexities of such data, which are often spatio-temporal, chaotic, and non-stationary, the CNN algorithms must be designed/evaluated for each specific dataset and application. Yet cnn, being a supervised technique, requires a large labeled dataset to start. Labeling demands (human) expert time which, combined with the limited number of relevant examples in this area, can discourage using CNNs for new problems. To address these challenges, here we (1) Propose an effective auto-labeling strategy based on using an unsupervised clustering algorithm and evaluating the performance of CNNs in re-identifying and predicting these clusters up to 5 days ahead of time;(2) Use this approach to label thousands of daily large-scale weather patterns over North America in the outputs of a fully-coupled climate model and show the capabilities of cnns in re-identifying and predicting the 4 clustered regimes up to 5 days ahead of time. The deep CNN trained with 1000 samples or more per cluster has an accuracy of 90% or better for both identification and prediction while prediction accuracy scales weakly with the number of lead days. Accuracy scales monotonically but nonlinearly with the size of the training set, e.g. reaching 94% with 3000 training samples per cluster for identification and 93-76% for prediction at lead day 1-5, outperforming logistic regression, a simpler machine learning algorithm, by ~ 25%. Effects of architecture and hyperparameters on the performance of cnns are examined and discussed. open Scientific RepoRtS | (2020) 10:1317 | https://doi.org/10.1038/s41598-020-57897-9www.nature.com/scientificreports www.nature.com/scientificreports/ transformed pattern recognition and image processing in various domains of business and science 44,45 and can potentially become a powerful tool for classifying and identifying patterns in climate and environmental data 43 . In fact, in their pioneering work, Liu et al. 46 and Racah et al. 47 have shown the promising capabilities of CNNs in identifying tropical cyclones, weather fronts, and atmospheric rivers in large, labeled climate datasets.Despite the success in applying CNNs in these few studies, there are some challenges that should be addressed to further expand the applications and usefulness of CNNs (and similar deep learning techniques) in climate and environmental sciences 48 . One major challenge is that unlike the data traditionally used to develop and assess CNN algorithms such as the static images in ImageNet 49 , climate and environmental data, from model simulations or observations are often spatio-temporal, highly nonlinear, chaotic, high-dimensional, non-stationary, multi-scale, and correlated. For example, the large-scale atmospheric circulation, whose variability strongly affects day-to-day weather and extreme events, is a high-dim...

This study aimed to identify the differentiating parameters of the spinal curves’ 2D projections through a hierarchical classification of the 3D spinal curve in adolescent idiopathic scoliosis (AIS). A total number of 103 right thoracic left lumbar pre-operative AIS patients were included retrospectively and consecutively. A total number of 20 non-scoliotic adolescents were included as the control group. All patients had biplanar X-rays and 3D reconstructions of the spine. The 3D spinal curve was calculated by interpolating the center of vertebrae and was isotropically normalized. A hierarchical classification of the normalized spinal curves was developed to group the patients based on the similarity of their 3D spinal curve. The spinal curves’ 2D projections and clinical spinal measurements in the three anatomical planes were then statistically compared between these groups and between the scoliotic subtypes and the non-scoliotic controls. A total of 5 patient groups of right thoracic left lumbar AIS patients were identified. The characteristics of the posterior-anterior and sagittal views of the spines were: Type 1: Normal sagittal profile and S shape axial view. T1 is leveled or tilted to the right in the posterior view. Type 2: Hypokyphotic and a V shape axial view. T1 is tilted to the left in the posterior view. Type 3: Hypokyphotic (only T5-T10) and frontal imbalance, S shape axial view. T1 is leveled or tilted to the right, and 3 frontal curves. Type 4: Flat sagittal profile (T1-L2), slight frontal imbalance with a V shape axial view, T1 tilted to the left. Type 5: flat sagittal profile and forward trunk shift with a proximal kyphosis and S shape axial view. T1 is leveled or tilted to the right. In conclusion, a hierarchical classification of the 3D scoliotic spine allowed identifying various distinguishing features of the spinal curves in patients with a right thoracic curve in an orderly fashion. The subtypes’ characteristics resulting from this 3D classification can be identified from the pairs of the frontal and sagittal spinal curves i.e. X-rays in right thoracic AIS patients.

Novel pelvic parameters were introduced to characterize the spinopelvic relative alignment in scoliotic subgroups. The proposed method related the orientation of the pelvis in the coronal and transverse planes to both thoracic and lumbar spinal deformities.

Adolescent idiopathic scoliosis (AIS) is a three-dimensional (3D) deformity of the spinal column in pediatric population. The primary cause of scoliosis remains unknown. The lack of such understanding has hampered development of effective preventive methods for management of this disease. A long-held assumption in pathogenesis of AIS is that the upright spine in human plays an important role in induction of scoliosis. Here, the variations in the sagittal curve of the scoliotic and non-scoliotic pediatric spines were used to study whether specific sagittal curves, under physiological loadings, are prone to 3D deformation leading to scoliosis. To this end, finite element models of the S shaped elastic rods, which their curves were derived from the radiographs of 129 sagittal spinal curves of adolescents with and without scoliosis, were generated. Using the mechanics of deformation in elastic rods, this study showed that the 3D deformation patterns of the two-dimensional S shaped slender elastic rods mimics the 3D patterns of the spinal deformity in AIS patients with the same S shaped sagittal spinal curve. On the other hand, the rods representing the non-scoliotic sagittal spinal curves, under the same mechanical loading, did not twist thus did not lead to a 3D deformation. This study provided strong evidence that the shape of the sagittal profile in individuals can be a leading cause of the 3D spinal deformity as is observed in the AIS population.