The lumbar spine plays a very important role in our load transfer and mobility. Vertebrae localization and segmentation are useful in detecting spinal deformities and fractures. Understanding of automated medical imagery is of main importance to help doctors in handling the time-consuming manual or semi-manual diagnosis. Our paper presents the methods that will help clinicians to grade the severity of the disease with confidence, as the current manual diagnosis by different doctors has dissimilarity and variations in the analysis of diseases. In this paper we discuss the lumbar spine localization and segmentation which help for the analysis of lumbar spine deformities. The lumber spine is localized using YOLOv5 which is the fifth variant of the YOLO family. It is the fastest and the lightest object detector. Mean average precision (mAP) of 0.975 is achieved by YOLOv5. To diagnose the lumbar lordosis, we correlated the angles with region area that is computed from the YOLOv5 centroids and obtained 74.5% accuracy. Cropped images from YOLOv5 bounding boxes are passed through HED U-Net, which is a combination of segmentation and edge detection frameworks, to obtain the segmented vertebrae and its edges. Lumbar lordortic angles (LLAs) and lumbosacral angles (LSAs) are found after detecting the corners of vertebrae using a Harris corner detector with very small mean errors of 0.29° and 0.38°, respectively. This paper compares the different object detectors used to localize the vertebrae, the results of two methods used to diagnose the lumbar deformity, and the results with other researchers.

The modern computer-aided screening systems re-quire a large amount of well-annotated training data to produce robust and consistent diagnostic performance. Furthermore, the public datasets designed to evaluate automated spinal disorders screening frameworks lack quantitative labels, which are marked by expert radiologists and clinically validated by spinal surgeons. This paper presents a dataset containing high-resolution (and well-labeled) mid-sagittal views of lumbar spine magnetic resonance imaging (MRI) scans. These scans also contain vertebral body masks along with clinically significant spinal measurements, including lumbar height, intervertebral body distances, vertebral body sidewall dimensions, vertebral body superior and inferior end-plates dimensions, lumbar lordotic angles, and lumbosacral angles. The mid-sagittal view MRI scans within the proposed dataset were first procured, and then they were manually marked by the expert radiologists and validated by the expert spinal surgeons. Afterward, different spinal measurements were recorded, which serves as a benchmark to evaluate the autonomous frameworks for predicting spinal misalignments. In addition to this, the proposed dataset is, to the best of our knowledge, the first composite database that contains lumbar spine mid-sagittal images along with spinal attributes and detailed markings of radiologists duly verified by the spinal surgeons. The proposed dataset, unlike its competitors, also introduces a quantitative vote to the clinicians and researchers in the assessment process of lumbar spine disorders. Apart from this, the dataset is publicly available at https://data.mendeley.com/datasets/k3b363f3vz/2.

Analysis of light curves emanating from various celestial bodies is of paramount importance in order to enable ourselves with quantify the variability in sky and discover time-varying objects. Large Synoptic Survey Telescope (LSST)gather voluminous time-series data. However, classifying these events from large-scale surveys is a challenging task that requires efficient and robust machine learning methods. In this paper, we present a novel approach for astronomical time series classification using gradient boost, a powerful ensemble technique that combines weak learners into a strong classifier. We apply our method to two datasets from the Catalina and Zwicky TransientFacility surveys, which contain light curves of various types of transients and variables. We compare our results with state-of-the-art methods that use different features and models. We show that our method achieves superior performance in terms of accuracy with comparable computational complexity. We also discuss the advantages and limitations of our method and suggest possible directions for future work.

The authors have proposed a new classification system for lateral lumbar interbody fusion (LLIF) at the L4-L5 level which is based on identification of easily recognizable key structures i.e., the position of the psoas muscle, the lumbar plexus and t...