Fast and accurate segmentation of musculoskeletal ultrasound images is an on-going challenge. Two principal factors make this task difficult: firstly, the presence of speckle noise arising from the interference that accompanies all coherent imaging approaches; secondly, the sometimes subtle interaction between musculoskeletal components that leads to non-uniformity of the image intensity. Our work presents an investigation of the potential of Convolutional Neural Networks (CNNs) to address both of these problems. CNNs are an effective tool that has previously been used in image processing of several biomedical imaging modalities. However, there is little published material addressing the processing of musculoskeletal ultrasound images, particularly using a panoramic technique. In our work we explore the effectiveness of CNNs when trained to act as a pre-segmentation pixel classifier that determines whether a pixel is an edge or non-edge pixel. Our CNNs are trained using two different ground truth interpretations. The first one uses an automatic Canny edge detector to provide the ground truth image; in the second interpretation, the ground truth was obtained using the same image marked-up by an expert anatomist. In this initial study the CNNs have been trained using half of the prepared data from one image, using the other half for testing; validation was also carried out using three unseen ultrasound images. CNN performance was assessed using Mathew's Correlation Coefficient, Sensitivity, Specificity and Accuracy. The results show that CNN performance when using expert ground truth image is better than in the case of using Canny ground truth image. Our technique is promising and has the potential to speed-up the image processing pipeline using appropriately trained CNNs.

One of the significant parameters that helps in the reporting the highest risk areas, which have COVID 19 pandemic is case fatality rate (CFR). In this work, automated analysis was carried out to evaluate fatality rate (CFR) across different countries. Furthermore, a state of art algorithm is proposed to estimate CFR and it is possible to make it applicable in the mobile phone. This application will enable us to monitor the status level of the patients (suspected, exposed and infected) to save time , efforts and get a high quailty of the recordings. All data were obtained from ( https://www.worldometers.info/coronavirus/ ) and pointed at the period between 27th March and 27th May, 2020. Results present Spain and Egypt have a highest score of the fatality rate (approximately 24%) compared with previous research, which Italy was the highest score of the case fatality rate (CFR). On the other hand, Australia has had the lowest of the (CFR) in the current and previous researches. Furthermore, Spain has the highest percentage score of the total active cases and death rate: 0.41% and 0.00073% respectively. Documentation and comparison fatality rate of COVID 19 pandemic across different countries could assist in illustrating the strength of this pandemic, speed spreading and risk area which infected of this disease.

Edge detection in Musculoskeletal Ultrasound Imaging readily allows an ultrasound image to be rendered as a binary image. This facilitates automated measurement of geometric parameters, such as muscle thickness, circumference and cross-sectional area of the tendon. In this work, we introduced a new method of edge detection based on a fuzzy inference system and apply it to the ultrasound image. An anisotropic diffusion filter was used to reduce speckle noise before implementation of the edge detection method, which consists of three characteristic steps. The first step entailed fuzzification, for which three fuzzy membership functions were applied to the image. The parameters of these functions were selected based on an analysis of the standard deviation of grey level intensities in the image. Secondly, 12 fuzzy rules for identifying edges were constructed. Thirdly, defuzzification was carried out using the Takagi-Sugeno method. Furthermore, a reference-based edge measurement was quantitatively determined by comparing edge characteristics with a standard reference. We made two inferences from our observations. Firstly, the ability to automatically identify the important details of a musculoskeletal ultrasound image in a very short time is possible. Secondly, this method is effective compared with other methods.

Capturing accurate representations of musculoskeletal system morphology is a core aspect of musculoskeletal modelling of the upper limb. Measurements of important geometric parameters such as the thickness of muscles and tendons are key descriptors of the underlying morphology. Though the measurement of those parameters can be estimated manually using cadaveric measurements, this is not an appropriate technique for constructing a personalised musculoskeletal model for an individual. Therefore, this work proposes and applies a novel method for evaluating the geometric parameters of the upper extremity based on automated ultrasound image analysis. The proposed algorithm involves advanced techniques from artificial intelligence and image processing to outline the necessary details of the musculoskeletal morphology from appropriately enhanced ultrasound images. The ultrasound images were collected from 25 healthy volunteers from different parts of upper limb. The results were compared with measurements of a manual evaluation. Our results showed that the average discrepancy between the manual and automatic measures of triceps thickness is 0.115 mm. This represents improved accuracy compared to several current approaches.

INTRODUCTION Mini-incision focused parathyroidectomy (MI-FP) is advocated as an alternative to bilateral neck exploration (BNE), owing to its reduced morbidity. The site and side of the affected gland is identified preoperatively using a combination of ultrasound and sestamibi scans. However, the acceptable degree of inter-scan concordance required to prompt MI-FP without compromising accuracy is undetermined. METHODS Accuracy of preoperative imaging was determined both individually and in combination for all parathyroidectomies (2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014). A grading system (excellent, good, poor) was devised to describe the interscan concordance, which was validated by the operative and histological findings. RESULTS Eighty-nine patients (17 male, 68 female) underwent parathyroidectomy (MI-FP 44, BNE 45). The accuracy of scans interpreted individually was 53% for ultrasound and 60% for sestamibi, with no difference according to surgical technique (P = 0.43, P = 1, respectively). The proportion of interscan concordance was: excellent -35%, good -40%, poor 25%. Combined accuracy was 100% for both excellent and good grades but only 13% for those graded poor. Similar rates of normocalcaemia were observed for MI-FP and BNE, while postoperative hypocalcaemia was five times higher in those undergoing BNE. CONCLUSIONS Reduction in the inter-scan concordance from excellent to good does not compromise accuracy. MI-FP could be successfully performed in up to 75% of patients -25% higher than recommended in national guidelines. Focused parathyroidectomy does not compromise surgical and endocrinological outcomes but boasts a far superior complication rate.