An Edge Refined MAP Based Tomographic Image Reconstruction with Improved Computational Efficiency

Tomography has established itself as a reliable imaging technique utilized across diverse fields. X-ray tomography has been extensively used in medical imaging applications. It provides a highly precised method of visualizing internal structures of the body with a non-destructive imaging technique. While over time X-ray tomography has been widely utilized in biomedical applications, in recent times, microwave tomography has become an emerging technique in scientific research due to its radiation safety. However, there still exists a need for accurate image reconstruction with a much lower computational workload. In this paper we have proposed a novel iterative reconstruction method utilizing Maximum a posteriori estimation. The proposed method reduces the computational workload significantly as compared to conventional iterative methods. The performance of the proposed method for reconstruction has been evaluated using a publicly available Computed Tomography scan data-set. We have utilized some widely used metrics like Structural Similarity Index Measure, Mean Square Error and Peak Signal-to-Noise Ratio for the performance illustration of the reconstructed target. Experimental results suggest that the proposed method performs really efficient reconstruction, reducing the computational workload around 18 times compared to conventional iterative reconstruction method. In the future, we aim to extend our work to microwave tomography imaging experiments.