Three-dimensional models have played important roles in medical simulation and education. Surface models can be manipulated in real time and even online; surface models have significant features for an interactive simulation system. The objective surface models are obtainable from accumulation of each structure’s outlines, followed by surface reconstruction. The aim of this research was to suggest the arranged methods of surface reconstruction, which might be applied to building surface models from serial images, such as computed tomographic scans and magnetic resonance images. We used recent state-of-the-art sectioned images of a cadaver head in which several structures were delineated. Four reconstruction methods were regulated according to the structure’s morphology: all outlines of a structure are overlapped and singular (method 1), overlapped and not singular (method 2), not overlapped but singular (method 3), and neither overlapped nor singular (method 4). From the trials with various kinds of head structures, we strongly suggested methods 1 and 2, in which volume reconstruction before surface reconstruction accelerated the processing speed on 3D-DOCTOR. So as to use methods 1 and 2, how to make the neighboring outlines overlapped in advance was discussed. The surface models of detailed head structures prepared in this investigation will hopefully contribute to various simulations for clinical practice. The value of the surface models are enhanced if they are placed over the original sectioned images, outlined images, and magnetic resonance images of the same cadaver.
From the *Department of Anatomy, Ajou University School of Medicine, Suwon; and †Department of Anatomy, Dongguk University College of Medicine, Gyeongju, Republic of Korea.
Received January 17, 2011.
Accepted for publication March 19, 2011.
Address correspondence and reprint requests to Jin Seo Park, PhD, Department of Anatomy, Dongguk University College of Medicine, 707 Seokjang-dong, Gyeongju 780-714, Republic of Korea; E-mail: email@example.com
This work was supported by a National Research Foundation of Korea grant funded by the Korean government (MEST; no. 2010-0015451).
The authors report no conflicts of interest.