A random forest-based framework for 3D kidney segmentation from dynamic contrast-enhanced CT images

A framework for 3D kidney segmentation from abdominal computed tomography (CT) images is proposed. Accurate kidney segmentation from CT images is a challenging task due to the large inhomogeneity of the kidney (e.g., cortex and medulla), inter-patient anatomical differences, etc. To account for these challenges, a novel framework utilizing random forest (RF) classification that has the ability to cluster complex data is proposed. To build a robust classification model, discriminative features are needed for better separation of data classes. In this work, regional features from the CT appearance, a kidney shape prior model, and higher-order spatial interactions are extracted and are used for tissue classification. The shape model is constructed using a set of training images and is updated during segmentation using an appearance-based method taking into account both voxels' locations and appearances. The spatial interactions between CT data voxels are modeled using a higher-order spatial model that adds to the pairwise cliques the families of the triple- and quad cliques. The proposed framework has been tested on CT data that has been collected from 20 subjects and consist of multiple 3D CT scans acquired at the pre-and post-contrast agent administration. Evaluation results, using both volumetric and distance-based metrics, between manually drawn and automatically segmented contours confirm the high accuracy of the proposed technique.

[1]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[2]  Jia Gu,et al.  Segmentation of Kidneys from Computed Tomography Using 3D Fast GrowCut Algorithm , 2013 .

[3]  L.-K. Shark,et al.  Medical Image Segmentation Using New Hybrid Level-Set Method , 2008, 2008 Fifth International Conference BioMedical Visualization: Information Visualization in Medical and Biomedical Informatics.

[4]  Laurent D. Cohen,et al.  Automatic Detection and Segmentation of Kidneys in 3D CT Images Using Random Forests , 2012, MICCAI.

[5]  Jakob Wasza,et al.  Article in Press G Model Computerized Medical Imaging and Graphics Segmentation of Kidneys Using a New Active Shape Model Generation Technique Based on Non-rigid Image Registration , 2022 .

[6]  R. Pohle,et al.  Self-learning model-based segmentation of medical images , 2001 .

[7]  Ayman El-Baz,et al.  Dynamic Contrast-Enhanced MRI-Based Early Detection of Acute Renal Transplant Rejection , 2013, IEEE Transactions on Medical Imaging.

[8]  Georgy L. Gimel'farb,et al.  A new deformable model-based segmentation approach for accurate extraction of the kidney from abdominal CT images , 2011, 2011 18th IEEE International Conference on Image Processing.

[9]  Xinjian Chen,et al.  Medical Image Segmentation by Combining Graph Cuts and Oriented Active Appearance Models , 2012, IEEE Transactions on Image Processing.

[10]  Jong-Won Park,et al.  Organ segmentation by comparing of gray value portion on abdominal CT image , 2000, WCC 2000 - ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000.

[11]  Xinjian Chen,et al.  Hierarchical Scale-Based Multiobject Recognition of 3-D Anatomical Structures , 2012, IEEE Transactions on Medical Imaging.

[12]  Leo Joskowicz,et al.  Non-parametric Iterative Model Constraint Graph min-cut for Automatic Kidney Segmentation , 2010, MICCAI.

[13]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[14]  D. Mahapatra,et al.  Analyzing Training Information From Random Forests for Improved Image Segmentation , 2014, IEEE Transactions on Image Processing.

[15]  Yanmei Liang,et al.  Contextual information-aided kidney segmentation in CT sequences , 2013 .

[16]  Ayman El-Baz,et al.  3D Kidney Segmentation from CT Images Using a Level Set Approach Guided by a Novel Stochastic Speed Function , 2011, MICCAI.

[17]  Aly A. Farag,et al.  Precise segmentation of multimodal images , 2006, IEEE Transactions on Image Processing.

[18]  Ben Glocker,et al.  Deformable medical image registration: setting the state of the art with discrete methods. , 2011, Annual review of biomedical engineering.

[19]  Daw-Tung Lin,et al.  Computer-aided kidney segmentation on abdominal CT images , 2006, IEEE Transactions on Information Technology in Biomedicine.

[20]  Ayman El-Baz,et al.  Kidney segmentation from CT images using a 3D NMF-guided active contour model , 2016, 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI).

[21]  Ronald M. Summers,et al.  Automatic 3D kidney segmentation based on shape constrained GC-OAAM , 2011, Medical Imaging.

[22]  Jong-Won Park,et al.  Segmentation of kidney without using contrast medium on abdominal CT image , 2000, WCC 2000 - ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000.

[23]  Huazhong Shu,et al.  Automatic kidney segmentation in CT images based on multi-atlas image registration , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[24]  Pau-Choo Chung,et al.  Multiphase level set with multi dynamic shape models on kidney segmentation of CT image , 2009, 2009 IEEE Biomedical Circuits and Systems Conference.

[25]  Ronald M. Summers,et al.  Computer-aided detection of exophytic renal lesions on non-contrast CT images , 2015, Medical Image Anal..

[26]  Klaus D. Tönnies,et al.  A New Approach for Model-Based Adaptive Region Growing in Medical Image Analysis , 2001, CAIP.

[27]  Chengwen Chu,et al.  Multi-organ Segmentation Based on Spatially-Divided Probabilistic Atlas from 3D Abdominal CT Images , 2013, MICCAI.