Journal of Intelligent Systems and Internet of Things

For businesses that depend on reliable and secure IT systems, choosing the best location for a data center is of paramount importance. Data center accessibility, operational efficiency, cost, and security are all affected by their physical location. The procedure entails considering a wide range of elements to guarantee that the final site meets the needs of the business. This paper investigated the multi-criteria decision-making (MCDM) model to select the best data center location based on a set of criteria. The MCDM method is integrated with the single-valued neutrosophic set (SVNS) to deal with vague and inaccurate information. A neutrosophic set with truth, indeterminacy, and falsity membership functions all in the range [0, 1] is called a SVNS. This paper used SVNS with three MCDM methods such as entropy, TOPSIS, and MABAC techniques. The entropy technique is used to compute the weights of criteria, then the TOPSIS and MABAC methods are used to rank the locations. The case study is investigated in Egypt. This paper used ten criteria and eight alternatives.

In the present era, there are lots of advancements and initiatives that have been undertaken through image processing techniques and IoT (Internet of Things). Image processing has proven its valuable insights in various applications such as GIS, biomedical, security, satellite imaging, medicine, and personal image analysis. In the context of fracture detection, image improvements, feature segmentation, and feature extraction techniques are commonly implemented including in the IoT Environment. The lower long bone, hand bone, and elbow bones are the particular interest due to their high incidence of fractures. X-ray diagnosis is a common method of detecting bone fractures due to its rapid and widespread availability. X-ray imaging involves a small amount of ionizing radiation in each part of the body, which is then captured on a particular film or digital detector. X-ray images, though they may have limitations compared to other imaging modalities, provide sufficient quality for fracture detection. There are three points of motivation for this research i.e. First- ease of use of software for patients and reduce the time for doctors and patients by screening out straight forward, Second- to decrease human mistakes that can also occur from manually inspecting a massive dataset of X-ray images to become aware of fractured sections of bones in hospitals, third- use of IoT infrastructure to collecting images of X-Rays and performing processing on received data by which we can send some accurate information back to the patients. The research aims to develop an automated environment i.e IoT emulation Framework consisting of image pre-processing such as attainment of images, pre-post-processing, segment methods, feature extraction, fracture detection, and visualization. Feature Extraction algorithm includes, CLAHE object with the preferred clip limit 2.0, CLAHE to the grayscale image, Gaussian blur to overcome more noise, Canny side detection, Hough Transform for line detection, and the gradient magnitude to acquire binary edges varied out through IoT. The framework utilizes the Canny edge detection methodology and Sobel operator for image segmentation. In this heat maps of images are also observed, which provide accurate information from bone images through IoT. The proposed system illustrates extreme accuracy and effectiveness, as proved by the results acquired from numerous experiments. The automated labeling and detection of bone fractures through photo processing by way of IoT offer great potential for fast and correct diagnosis, contributing to successful treatment outcomes.