AI, Planning and Control Algorithms for IoRT Systems

Introduction: Hamstring avulsion injury can be defined as improper functioning of the hamstring muscles due to external stress or strain. The detachment of hamstring muscular region from the tendons occurs under severe conditions. Magnetic Resonance Imaging (MRI) is used as a diagnostic tool to study the region of injury. Histogram analysis is a special type of image processing technique used to describe the changes in the gray levels of the images based on the abnormalities. Aim: To analyse the region of Hamstring avulsion injury in MRI slices using histogram. Materials and Methods: Hamstring avulsion injury MRI slices were obtained from an authorised database, Radiopeadia for analysis. Histogram of each slice of the MRI image was determined. Statistical features were derived from the histogram of each slice which was used to identify specific slice of MRI with high intensity avulsion injury. MRI slice with high intensity was determined for further processing and analysis of the injury. For further accurate identification of the high intensity slice in the MRI slices, statistical features of the slice histogram were obtained. These features were used to define the exact high intensity slice. Results: Histograms of 10 slices of an image were compared to obtain the slice with high intensity region. After comparison, slice four described the high intensity which defines the affected region of hamstring muscle. For further detailed study, statistical values were derived from the histogram. These values were significant for all the 10 slices but highly significant for the slice four. Slice four was further analysed to confirm all the details about the nature of the affected region of injury in hamstring muscle. Conclusion: Histogram defines the intensity variations of the pixels which mainly illustrate the nature of abnormalities, injuries or recognition of any region in the human body. Histogram based statistical values were used to analyse the abnormality or injury. These features were used in different ways for diagnosis and analysis of various abnormalities or disorders. Thus, Histogram and statistical features derived from the histogram from the MRI slices of hamstring avulsion injury were compared to identify the slice that possess high intensity region for the identification of the injury and its characteristics.

Robotics involves design, construction, operation, and use of intelligent machines that possess the ability to sense, compute, manipulate, and navigate environments to monitor events and execute an appropriate course of action. Internet of Things (IoT) on the other hand is a fast-developing novel technology consisting of group of uniquely addressable heterogeneous smart objects or tiny devices (things) interconnected via the Internet to share and process data from different sources. IoT is designed with the goal to “connect everything and everyone everywhere to everything and everyone else.” The two technologies, IoT and robotics, have evolved into Internet of Robotic Things (IoRT) by the creation of a synergy between the two. IoRT aims at enhancing the current IoT with active sensing and actuation from robotics. This idea opened a novel opportunity for collaboration between IoT and robotics applications and research communities. However, most application domains of IoT and robotics have not fully explored the use of IoRT. This chapter discusses the (potential) applications of IoT-aided robotics in different domains, explaining how robots can extend the capabilities of existing IoT architectures to make them more knowledgeable and smarter; discuss some of the challenges in the full realization and application of IoRT; and lastly proposes an IoRT architecture for smart library management, an area that has not received much attention in the research community.

The primary indication of breast cancer is the presence of calcification clusters. It is challenging and lengthy process for radiologists to identify and classify micro calcifications as non-cancerous or cancerous. In this proposed work, a novel method for the detection of micro calcification clusters in mammograms is explained that consists of two main sections. First, mammogram preprocessing is done. Second, micro calcification are segmented out. In preprocessing noise and label are removed as well as contrast is enhanced. Then various segmentation methods are used for comparison of calcification region. Watershed segmentation, Marker controlled watershed segmentation (MCWS), Texture segmentation and Level set segmentation methods are applied to Digital Database for Screening Mammography (DDSM) database. Results show that the MCWS provides quite acceptable detection performance. The major advantage of this method is its capability to detect micro calcifications perfectly even in case of very dense mammograms. The performance of different methods is evaluated by comparing the obtained segmented image with expert radiologist data. The comparison study aptly shows that the micro calcifications can be exactly segment and can avoid over segmentation problem of existing method.