E-Books →Development and Evaluation of Coded Aperture Reconstruction Methods for Intraoperative Gamma Cameras

Free Download Development and Evaluation of Coded Aperture Reconstruction Methods for Intraoperative Gamma Cameras by Tobias Meißner
English | 2024 | ISBN: 3731513943 | 220 pages | PDF | 26 Mb
Female breast cancer has surpassed lung cancer as the most commonly diagnosed cancer. The biopsy of lymph nodes that receive drainage from the primary tumor are guided by Intraoperative Gamma Cameras (IGCs). Coded Aperture Imaging (CAI) has been proposed as an alternative collimation technique to produce an image, because it offers a better tradeoff between sensitivity and spatial resolution. However, it requires image reconstruction. Therefore, a Convolutional Encoder-Decoder Network (CED) was developed and trained on synthetic source images. It was quantitatively compared based on the reconstruction quality and the runtime with state-of-the-art methods, such as MURA Decoding and the iterative Maximum Likelihood Expectation Maximization (MLEM) algorithm. Afterwards the axial resolution and 3D localization accuracy of the most commonly used methods were analyzed. In conclusion, the CED outperforms all methods regarding the quality. From the analytical methods the MLEM algorithm yields a higher quality and a better axial resolution, but is not suitable for IGCs in its current form, due to its computational complexity. MURA Decoding provides robust reconstructions. Its fast computation enables 3D reconstruction which allows the localization of point-like sources with an accuracy comparable to that of stereoscopic cameras.