No abstract

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.

Medical Image Analysis is an ongoing field of research nowadays. Diabetic Retinopathy (DR) is one of the major diseases being workaround using different techniques of image analysis. Advanced stage of DR is commonly treated with laser at the present time which is a major tool to safe further vision loss which leaves marks on the surface of the retina. We present an automated system for detection of laser marks from colored retinal images to facilitate automated diagnosis of retinal diseases. The proposed system performs preprocessing on the image in order to extract all possible candidate laser mark regions. This is followed by a post processing stage to remove false pixels from candidate regions. The method extracts a number of features for proper representation of all candidate regions. These extracted features are used to facilitate the later classification stage for accurate detection of laser marks from all candidate regions. The validity of a proposed system is performed on a locally gathered retinal image database and results show the significance of proposed system.

The retinal recognition technique is yet another step in biometrics, as a very rare but the most accurate authentication method. It contains the distinct property of exceptionally very low false acceptance and false rejection rates. The performance of any biometrics based identification system is enormously affected by its correct matching rate. In this paper, we present a new windowing technique for feature point validation in order to improve the accuracy of retinal vascular pattern recognition. The proposed method applies a windowing technique on skeleton vascular pattern to eliminate false feature points which appear due to vessel breakages, short vessels and spurs. The testing of the proposed methodology is performed using two retinal image databases i.e. VARIA and RIDB. Experimental analysis of given approach shows a significant decrease of false feature points.

Summary Spinal cord is the one of the most important organs in the central nervous system (CNS). It acts as the main processing hub which serves as the main passage line for information transfer from brain to the rest of the body. It supports the whole skeleton structure along with mobility, bending, turning, twisting and so forth. Several factors may result in the deformity of spine such as a major injury, fracture or a defect by birth. In this research, we have discussed two modules: one is for vertebrae localization and spine segmentation and the second one is for analysis of spine dis‐proportionality. A recent approach of YOLOv5 is used for the localization of vertebrae in combination with Mask RCNN for segmentation of spinal column. The combined results from both these modules are used for feature extraction which supports our classification‐based shape analysis module. The AASCE 2019 challenge dataset is used to evaluate the experimental results and the value of mAP achieved is 0.94 at 0.5 IOU threshold of YOLOv5 model. The proposed technique with novel feature set achieved an average classification accuracy of 94.69%.