Chest CT automatic analysis for lung nodules detection implemented on a GPU computing system

作者: Niccolo Camarlinghi , Francesco Bagagli , Piergiorgio Cerello , Alessandra Retico , Maria Evelina Fantacci

DOI: 10.1109/NSSMIC.2012.6551464

关键词: Artificial intelligenceComputational scienceCUDAVoxelFilter (video)Computer visionTomographyHessian matrixGeneral-purpose computing on graphics processing unitsCADComputer scienceImage processing

摘要: The aim of this work is the efficient implementation Hessian based filters. These filters are commonly used in medical image analysis and employed Voxel Based Neural Approach (VBNA) lung CAD (Computer Aided Detection) system for nodule detection. This mainly focuses on optimization filter devoted to detection internal candidates, called Multi Scale Dot Enhancement (MSDE) algorithm. Two fast variants MSDE algorithm here proposed compared: first relies an analytical it implemented a standard CPU, whereas second consists implementing CUDA Graphical Processing Unit (GPU) framework. algorithms were tested with computed tomography images belonging Lung Image Database Consortium (LIDC) public research database using Intel Core i7 950 @ 3.07GHz NVIDIA GeForce GTX 580. Both approaches lead improvement performance respect original implementation, without any loss precision. initial realized Insight ToolKit open source framework (ITK), had average execution time 69 sec per CT five scales enhancement. analyticallyoptimized CPU leads computational speed gain 2.5× (28 CT), parallel speed-up 38x (1.8 CT) 15x approach. has been developed INFN-funded MAGIC-5 project.

参考文章(15)
Arnold M. R. Schilham, Bram van Ginneken, Marco Loog, Multi-scale Nodule Detection in Chest Radiographs medical image computing and computer assisted intervention. pp. 602- 609 ,(2003) , 10.1007/978-3-540-39899-8_74
Yoshinobu Sato, Shin Nakajima, Hideki Atsumi, Thomas Koller, Guido Gerig, Shigeyuki Yoshida, Ron Kikinis, 3D Multi-scale line filter for segmentation and visualization of curvilinear structures in medical images CVRMed-MRCAS '97 Proceedings of the First Joint Conference on Computer Vision, Virtual Reality and Robotics in Medicine and Medial Robotics and Computer-Assisted Surgery. pp. 213- 222 ,(1997) , 10.1007/BFB0029240
JOACHIM KOPP, Efficient numerical diagonalization of hermitian 3x3 matrices International Journal of Modern Physics C. ,vol. 19, pp. 523- 548 ,(2008) , 10.1142/S0129183108012303
Qiang Li, Shusuke Sone, Kunio Doi, Selective enhancement filters for nodules, vessels, and airway walls in two- and three-dimensional CT scans. Medical Physics. ,vol. 30, pp. 2040- 2051 ,(2003) , 10.1118/1.1581411
Niccolò Camarlinghi, Ilaria Gori, Alessandra Retico, Roberto Bellotti, Paolo Bosco, Piergiorgio Cerello, Gianfranco Gargano, Ernesto Lopez Torres, Rosario Megna, Marco Peccarisi, Maria Evelina Fantacci, Combination of computer-aided detection algorithms for automatic lung nodule identification computer assisted radiology and surgery. ,vol. 7, pp. 455- 464 ,(2012) , 10.1007/S11548-011-0637-6
I Gori, F Bagagli, M E Fantacci, A Preite Martinez, A Retico, I De Mitri, S Donadio, C Fulcheri, G Gargano, R Magro, M Santoro, S Stumbo, Multi-scale analysis of lung computed tomography images Journal of Instrumentation. ,vol. 2, pp. 09007- ,(2007) , 10.1088/1748-0221/2/09/P09007
Alessandra Retico, Francesco Bagagli, Niccolo Camarlinghi, Carmela Carpentieri, Maria Evelina Fantacci, Ilaria Gori, A voxel-based neural approach (VBNA) to identify lung nodules in the ANODE09 study Proceedings of SPIE. ,vol. 7260, ,(2009) , 10.1117/12.811721
Maria Evelina Fantacci, Niccolo Camarlinghi, Roberto Bellotti, Gianfranco Gargano, Rosario Megna, Piergiorgio Cerello, Ivan De Mitri, Giorgio De Nunzio, Ilaria Gori, Ernesto Lopez Torres, Cristiana Peroni, Alessandra Retico, Algorithms for automatic detection of lung nodules in CT scans ieee international symposium on medical measurements and applications. pp. 623- 627 ,(2011) , 10.1109/MEMEA.2011.5966737
Y. Sato, C. Westin, A. Bhalerao, S. Nakajima, N. Shiraga, S. Tamura, R. Kikinis, Tissue classification based on 3D local intensity structures for volume rendering IEEE Transactions on Visualization and Computer Graphics. ,vol. 6, pp. 160- 180 ,(2000) , 10.1109/2945.856997