Automatic Global Matching of Temporal Chest MDCT Scans for Computer-Aided Diagnosis

• Published in: Systems Modeling and Simulation: Theory and Applications pp. 652 - 662
• Main Authors: Hong, Helen, Lee, Jeongjin, Yim, Yeni, Shin, Yeong Gil
• Format: Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2005
• Series: Lecture Notes in Computer Science
• Subjects: Iterative Close Point; Iterative Close Point Algorithm; Lung Surface; Surface Registration; Target Volume
• ISBN: 3540244778; 9783540244776
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-540-30585-9_73

We propose a fast and robust global matching technique for detecting temporal changes of pulmonary nodules. For the registration of a pair of CT scans, a proper geometrical transformation is found through the following steps. First, an automatic segmentation is used for identifying lung surfaces in chest MDCT scans. Second, optimal cube registration is performed for the initial gross registration. Third, for allowing fast and robust convergence on the optimal value, a 3D distance map is generated by the narrow band distance propagation. Finally, the distance measure between surface boundary points is repeatedly evaluated by the selective distance measure to align lung surfaces. Experimental results show that the computation time and robustness of our registration method is very promising compared with conventional methods. Our method can be used for investigating temporal changes such as pulmonary infiltration, tumor masses, or pleural effusions.