Comparison of human and automatic segmentations of kidneys from CT images.

PURPOSE A controlled observer study was conducted to compare a method for automatic image segmentation with conventional user-guided segmentation of right and left kidneys from planning computerized tomographic (CT) images. METHODS AND MATERIALS Deformable shape models called m-reps were used to automatically segment right and left kidneys from 12 target CT images, and the results were compared with careful manual segmentations performed by two human experts. M-rep models were trained based on manual segmentations from a collection of images that did not include the targets. Segmentation using m-reps began with interactive initialization to position the kidney model over the target kidney in the image data. Fully automatic segmentation proceeded through two stages at successively smaller spatial scales. At the first stage, a global similarity transformation of the kidney model was computed to position the model closer to the target kidney. The similarity transformation was followed by large-scale deformations based on principal geodesic analysis (PGA). During the second stage, the medial atoms comprising the m-rep model were deformed one by one. This procedure was iterated until no changes were observed. The transformations and deformations at both stages were driven by optimizing an objective function with two terms. One term penalized the currently deformed m-rep by an amount proportional to its deviation from the mean m-rep derived from PGA of the training segmentations. The second term computed a model-to-image match term based on the goodness of match of the trained intensity template for the currently deformed m-rep with the corresponding intensity data in the target image. Human and m-rep segmentations were compared using quantitative metrics provided in a toolset called Valmet. Metrics reported in this article include (1) percent volume overlap; (2) mean surface distance between two segmentations; and (3) maximum surface separation (Hausdorff distance). RESULTS Averaged over all kidneys the mean surface separation was 0.12 cm, the mean Hausdorff distance was 0.99 cm, and the mean volume overlap for human segmentations was 88.8%. Between human and m-rep segmentations the mean surface separation was 0.18-0.19 cm, the mean Hausdorff distance was 1.14-1.25 cm, and the mean volume overlap was 82-83%. CONCLUSIONS Overall in this study, the best m-rep kidney segmentations were at least as good as careful manual slice-by-slice segmentations performed by two experienced humans, and the worst performance was no worse than typical segmentations from our clinical setting. The mean surface separations for human-m-rep segmentations were slightly larger than for human-human segmentations but still in the subvoxel range, and volume overlap and maximum surface separation were slightly better for human-human comparisons. These results were expected because of experimental factors that favored comparison of the human-human segmentations. In particular, m-rep agreement with humans appears to have been limited largely by fundamental differences between manual slice-by-slice and true three-dimensional segmentation, imaging artifacts, image voxel dimensions, and the use of an m-rep model that produced a smooth surface across the renal pelvis.

[1]  R J Dowsett,et al.  Contouring structures for 3-dimensional treatment planning. , 1992, International journal of radiation oncology, biology, physics.

[2]  E van der Schueren,et al.  Quality assessment of medical decision making in radiation oncology: variability in target volume delineation for brain tumours. , 1993, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[3]  Johan Montagnat,et al.  Volumetric medical images segmentation using shape constrained deformable models , 1997, CVRMed.

[4]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[5]  Paul A. Yushkevich,et al.  Deformable M-Reps for 3D Medical Image Segmentation , 2003, International Journal of Computer Vision.

[6]  R. Mirimanoff,et al.  Comparison of treatment techniques for lung cancer. , 1993, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[7]  M A Viergever,et al.  Model-based analysis of medical images. , 1999, IEEE transactions on medical imaging.

[8]  Andrew Thall,et al.  Fast C 2 Interpolating Subdivision Surfaces using Iterative Inversion of Stationary Subdivision Rules , 2003 .

[9]  C H Ketting,et al.  Consistency of three-dimensional planning target volumes across physicians and institutions. , 1997, International journal of radiation oncology, biology, physics.

[10]  T. Pan,et al.  4D-CT imaging of a volume influenced by respiratory motion on multi-slice CT. , 2004, Medical physics.

[11]  Bart M. ter Haar Romeny,et al.  Geometry-Driven Diffusion in Computer Vision , 1994, Computational Imaging and Vision.

[12]  G E Hanks,et al.  Localization of the prostatic apex for radiation treatment planning. , 1995, International journal of radiation oncology, biology, physics.

[13]  Conglin Lu,et al.  A Markov Random Field Approach to Multi-scale Shape Analysis , 2003, Scale-Space.

[14]  Edward L. Chaney,et al.  Medical anatomy segmentation kit: combining 2D and 3D segmentation methods to enhance functionality , 1994, Optics & Photonics.

[15]  G E Hanks,et al.  Initial clinical assessment of CT-MRI image fusion software in localization of the prostate for 3D conformal radiation therapy. , 1997, International journal of radiation oncology, biology, physics.

[16]  Lewis D. Griffin,et al.  Scale Space Methods in Computer Vision , 2003, Lecture Notes in Computer Science.

[17]  Stephen M. Pizer,et al.  Guest Editorial—Medial & Medical: A Good Match for Image Analysis , 2003, International Journal of Computer Vision.

[18]  William E. Lorensen,et al.  Marching cubes: A high resolution 3D surface construction algorithm , 1987, SIGGRAPH.

[19]  H. Hricak,et al.  Prostate volumes defined by magnetic resonance imaging and computerized tomographic scans for three-dimensional conformal radiotherapy. , 1996, International journal of radiation oncology, biology, physics.

[20]  George T. Y. Chen,et al.  Artifacts in computed tomography scanning of moving objects. , 2004, Seminars in radiation oncology.

[21]  P. Thomas Fletcher,et al.  Principal geodesic analysis for the study of nonlinear statistics of shape , 2004, IEEE Transactions on Medical Imaging.

[22]  Edward L. Chaney,et al.  Clustering on image boundary regions for deformable model segmentation , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).

[23]  Demetri Terzopoulos,et al.  Deformable models in medical image analysis: a survey , 1996, Medical Image Anal..

[24]  James W. Moore,et al.  Institute of Electrical and Electronics Engineers (IEEE) , 2002 .

[25]  Guido Gerig,et al.  A user-guided tool for efficient segmentation of medical image data , 1997, CVRMed.

[26]  M van Herk,et al.  Definition of the prostate in CT and MRI: a multi-observer study. , 1999, International journal of radiation oncology, biology, physics.

[27]  Andrew Thall,et al.  A method and software for segmentation of anatomic object ensembles by deformable m-reps. , 2005, Medical physics.

[28]  Lawrence M. Fagan,et al.  Automated planning target volume generation: an evaluation pitting a computer-based tool against human experts. , 1997, International journal of radiation oncology, biology, physics.

[29]  Grigore C. Burdea,et al.  Virtual reality and robotics in medicine , 1996, Proceedings 5th IEEE International Workshop on Robot and Human Communication. RO-MAN'96 TSUKUBA.

[30]  Akshay K. Singh,et al.  Deformable models in medical image analysis , 1996, Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis.

[31]  Martin Styner,et al.  Automatic and Robust Computation of 3D Medial Models Incorporating Object Variability , 2003, International Journal of Computer Vision.

[32]  Guido Gerig,et al.  Valmet: A New Validation Tool for Assessing and Improving 3D Object Segmentation , 2001, MICCAI.

[33]  James S. Duncan,et al.  Mathematical Methods in Medical Imaging , 1992 .