Search

Purpose: Volume estimation of organs or fluid is one of the most important roles in autopsy. The aim of this study is to develop a volume estimation program for CT images of Digital Imaging and Communication in Medicine (DICOM) which runs on a personal computer, and to confirm the usefulness and accuracy of it. Materials and methods: We developed a program which extracted correct boundaries of the organs in spite of setting of rough region of interest (ROI). We retrospectively evaluated five postmortem CT (PMCT) examinations which were performed with a 64-row detector CT. Volume of five livers, five spleens, and 10 kidneys were evaluated by means of a general image processing program and our developed program. The averages of obtained volume and required time to set ROIs were analyzed with paired t-test.Results: There were no significant differences for the estimated volume of all organs between with the general image processing program and with our program. However, there were significant differences for the required time to set ROIs between with the two methods in all organs. The mean evaluation time was shortened by 52.0%, 67.2% and 74.0% in liver, spleen, and kidney, respectively. Conclusion: Our developed program provided correct volume value, and was able to reduce labor and time to set ROIs compared with using the general image processing program. Our program is useful to evaluate the volume on CT images of DICOM data especially in PMCT.

A system has been developed for calculating volume from a sequence of multiplanar two dimensional ultrasound images. Ultrasound image capture is via a video digitising card installed in a personal computer. Regions of interest are transformed from 2D image space to 3D space using position and orientation data obtained from an electromagnetic device (Fastrak, Polhemus Inc, VT) attached to the ultrasound probe. The accuracy of the system was assessed by scanning 10 water filled balloons (13 - 141 ml), 10 kidneys (147 - 200 ml) and 16 fetal livers (8 - 37 ml) immersed in water using an Acuson 128XP/10 (5 MHz curvilinear probe). Volume was calculated using the ellipsoid and planimetry methods, two tetrahedral methods, and two integral methods - a ray tracing algorithm and one based on Gauss’ theorem. Actual volume was estimated by weighing (balloons) and water displacement (kidneys and livers). The mean percentage error (± one standard deviation) for the ray tracing algorithm was 0.9 ± 2.4%, 2.7 ± 2.3%, 6.6 ± 5.4% for balloons, kidneys and livers respectively. Four sets of 10 kidneys were scanned using three scan techniques on four different ultrasound machines of varying image quality. There was no significant difference between scan techniques or machines. Twelve of the 16 fetal livers were scanned by computed tomography and magnetic resonance. The mean percentage errors were 5.3 ± 4.7%, -3.1 ± 9.6%, -0.03 ± 9.7 for ultrasound (radial scans, ray volumes), magnetic resonance and X-ray computed tomography (voxel counting) respectively. Tests carried out on the Fastrak showed that it is suitable for use in a clinical environment if care is exercised. These in vitro studies suggest that the system has the potential to be a useful tool in measuring organ volume in vivo.