Development of a microwave-based pre-clinical experimental orthopedical imaging system

Near-field microwave imaging (MI) is an emerging scanning method and has become a promising topic in both biomedical and industrial imaging in recent years. This is because conventional devices use high power and ionized signals and these signals have negative side effects on the target. In addition, these devices require massive measuring equipment and isolated rooms. Although the images obtained with MI do not have high resolution and contrast compared to conventional devices, this imaging technique provides very useful and valuable basic images for early diagnosis. Near-field MI uses non-ionizing and low-power electromagnetic signals to examine the target, unlike conventional devices. For this reason, there is no limit to their usage for biomedical purposes. In addition, microwave devices and equipment have portable, easy to install and has mobility features. Thus, the near field MI system enables the establishment of low-power, low-cost and practical imaging systems that do not emit ionizing radiation. Today, near-field MI is used for industrial and medical purposes. In recent years, imaging of body parts and tissues has attracted great interest and has become the focus of research around the world. MI is of great interest among researchers, especially in subjects such as breast tumour localization, skin diseases, brain stroke, tissue lesion detection, circulatory system diseases and skin cancer detection. Diagnosis in biomedical MI is made by utilizing different electrical properties of healthy and malignant tissues. Different electrical properties cause different electrical scattering events. That is, the signal scattered from the target has the electrical properties of itself. By using various inverse scattering algorithms, target specific electrical properties can be determined. Reconstruction of the image is the solution of an inverse scattering problem and different inverse scattering algorithms are used by the researchers. Although the quality of the images obtained with MI is similar, the data acquisition step, measurement campaign, calibration process and image reconstruction algorithms differ from each other in the studies conducted in the literature. In most studies in the literature, complex inverse scattering algorithms that involve mathematically intensive calculations have been used. These algorithms involve iterative and time-consuming steps that are difficult to implement. The inverse scattering algorithm used in this study uses circular analysis models instead of linear analysis models. This method is applied to raw data accurately, reliably and quickly. Since different tissue types have their own electrical properties, it is possible to distinguish them from each other. Proposed imaging technique consists of data acquisition, image reconstruction and image processing steps. In this study, firstly, cylindrical shaped metallic, dielectric and PVC materials are investigated for testing and validation of the near field MI system. In this study, main investigated biological tissue is buried and non-buried actual bone. This thesis has two main aims. The first purpose is to obtain microwave images using electromagnetic signals with low power and no ionizing effect. The second one is to automatically determine the orientation angle of the bone tissue obtained with the MI system. Because the existing bone curvature in orthopaedic images is calculated manually by the operator and varies from person to person. Miscalculations are possible in these measurements and the result is operator dependent. In the follow-up and treatment process of conditions that require constant observation, such as scoliosis, bone healing after fracture and monitoring of implanted body platinum, the patient is constantly exposed to ionized radiation found in conventional devices. For this reason, it is predicted that the presented imaging system can be used in the follow-up of bone healing and implants after union, and especially in the diagnosis and treatment of scoliosis. In this context, buried and unburied actual bone tissue was imaged with the near-field MI system. With the proposed imaging technique, the exact position of the target and its original shape except its background were obtained. The position and size of the target bone tissue were determined within 1% and 3% error in the x and y axes, respectively, based on real measurements. Finally, the orientation angle of the C-shaped curved bone fragment was calculated automatically with the algorithm running on the image. The orientation angle was determined with an error rate of approximately %3.5 compared to real measurements. The proposed imaging technique and orientation angle calculation give fast, accurate and reliable results and are not dependent on the user. The method does not involve iterative approach and intensive mathematical calculations.