Reconstructing Big Data Acquired from Radioisotope Distribution in Medical Scanner Detectors

In the last few years, CAD systems has been optimized significantly medical applications. Digital systems have been employed recently in diagnosing procedures and facilitate the process in determining illness in patients. Different scanner system have been used to acquire medical images, which are different the quality output. The main problem associated to image acquisition is the amount of information in the acquired images, and what are the exact time and bed displacement in each scan. This work tests different factors in PET scanners leading to standards for optimizing the scanner variables in medical diagnosis area. Simulated body phantom is experimented here due to the closure properties regarding to real patient data. The phantom out put images were acquired under different circumstances with different scanning time at each bed scan. The segmentation process is then applied on the best slice evaluating the actual spheres size. We propose an efficient way to set the best scanner variables during the scanning process, which lead to the most accurate segmentation result.