Improved curvature estimation for computer-aided detection of colonic polyps in CT colonography.

RATIONALE AND OBJECTIVES Current schemes for computer-aided detection (CAD) of colon polyps usually use kernel methods to perform curvature-based shape analysis. However, kernel methods may deliver spurious curvature estimations if the kernel contains two surfaces, because of the vanished gradient magnitudes. The aim of this study was to use the Knutsson mapping method to deal with the difficulty of providing better curvature estimations and to assess the impact of improved curvature estimation on the performance of CAD schemes. MATERIALS AND METHODS The new method was compared to two widely used kernel methods in terms of the performance of two stages of CAD: initial detection and true-positive and false-positive classification. The evaluation was conducted on a database of 130 computed tomographic scans from 67 patients. In these patient scans, there were 104 clinically significant polyps and masses >5 mm. RESULTS In the initial detection stage, the detection sensitivity of the three methods was comparable. In the classification stage, at a 90% sensitivity level on the basis of the input of this step, the new technique yielded 3.15 false-positive results per scan, demonstrating reductions in false-positive findings of 30.2% (P < .01) and 27.9% (P < .01) compared to the two kernel methods. CONCLUSIONS The new method can benefit CAD schemes with reduced false-positive rates, without sacrificing detection sensitivity.

[1]  Frans Vos,et al.  Detection and Segmentation of Colonic Polyps on Implicit Isosurfaces by Second Principal Curvature Flow , 2010, IEEE Transactions on Medical Imaging.

[2]  Ender Konukoglu,et al.  Polyp Enhancing Level Set Evolution of Colon Wall: Method and Pilot Study , 2007, IEEE Transactions on Medical Imaging.

[3]  Zhengrong Liang,et al.  Improving initial polyp candidate extraction for CT colonography , 2010, Physics in medicine and biology.

[4]  Alexis Gourdon,et al.  Computing the Differential Characteristics of Isointensity Surfaces , 1995, Comput. Vis. Image Underst..

[5]  Hiroyuki Yoshida,et al.  Three-dimensional computer-aided diagnosis scheme for detection of colonic polyps , 2001, IEEE Transactions on Medical Imaging.

[6]  Afra Zomorodian,et al.  Colon polyp detection using smoothed shape operators: Preliminary results , 2008, Medical Image Anal..

[7]  G. Kiss,et al.  Computer-aided detection for CT colonography: update 2007 , 2007, Abdominal Imaging.

[8]  Zhengrong Liang,et al.  Volume-based feature analysis of mucosa for automatic initial polyp detection in virtual colonoscopy , 2008, International Journal of Computer Assisted Radiology and Surgery.

[9]  C. Johnson,et al.  Colorectal cancer screening with CT colonography, colonoscopy, and double-contrast barium enema examination: prospective assessment of patient perceptions and preferences. , 2003, Radiology.

[10]  Dev P Chakraborty,et al.  Observer studies involving detection and localization: modeling, analysis, and validation. , 2004, Medical physics.

[11]  Lucas J. van Vliet,et al.  Representing Orientation in n-Dimensional Spaces , 2003, CAIP.

[12]  Zhengrong Liang,et al.  An EM approach to MAP solution of segmenting tissue mixture percentages with application to CT-based virtual colonoscopy. , 2008, Medical physics.

[13]  P. Pickhardt,et al.  Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. , 2003, The New England journal of medicine.

[14]  Zhengrong Liang,et al.  Computer-aided detection of colonic polyps with level set-based adaptive convolution in volumetric mucosa to advance CT colonography toward a screening modality , 2009, Cancer management and research.

[15]  D. Eddy,et al.  Screening for colorectal cancer , 1992, The Lancet.

[16]  B. Levin,et al.  Emerging Technologies in Screening for Colorectal Cancer: CT Colonography, Immunochemical Fecal Occult Blood Tests, and Stool Screening Using Molecular Markers , 2003, CA: a cancer journal for clinicians.

[17]  Nicholas Ayache,et al.  From voxel to intrinsic surface features , 1992, Image Vis. Comput..

[18]  Zhengrong Liang,et al.  Increasing computer-aided detection specificity by projection features for CT colonography. , 2010, Medical physics.

[19]  Tianhu Lei,et al.  Statistical approach to X-ray CT imaging and its applications in image analysis. I. Statistical analysis of X-ray CT imaging , 1992, IEEE Trans. Medical Imaging.

[20]  Zhengrong Liang,et al.  An EM Approach to MAP Solution of Segmenting Tissue Mixtures: A Numerical Analysis , 2009, IEEE Transactions on Medical Imaging.

[21]  Lucas J. van Vliet,et al.  On curvature estimation of ISO surfaces in 3D gray-value images and the computation of shape descriptors , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  Zhengrong Liang,et al.  Reduction of false positives by internal features for polyp detection in CT-based virtual colonoscopy. , 2005, Medical physics.

[23]  I. Bitter,et al.  Computed tomographic virtual colonoscopy computer-aided polyp detection in a screening population. , 2005, Gastroenterology.

[24]  Ronald M. Summers,et al.  Analysis of kernel method for surface curvature estimation , 2004, CARS.

[25]  R. Jeffrey,et al.  Automated polyp detector for CT colonography: feasibility study. , 2000, Radiology.

[26]  Stuart A. Taylor,et al.  Computer-assisted reader software versus expert reviewers for polyp detection on CT colonography. , 2006, AJR. American journal of roentgenology.

[27]  Guy Marchal,et al.  Computer-aided diagnosis in virtual colonography via combination of surface normal and sphere fitting methods , 2002, European Radiology.