EdgeRunner: a novel shape-based pipeline for tumours analysis and characterisation

Abstract Characterisation of tumours on imaging is usually performed qualitatively by visual analysis by radiologists. However, incorporating quantitative imaging features would potentially increase diagnostic accuracy and would improve patient care and management. The goal is to eventually be able to distinguish certain quantifiable features, which along with qualitative ones could differentiate between benign and malignant tumours. This may obviate the need for invasive procedures such as a biopsy, and may also allow for earlier detection as well as better follow-up. The shape of a tumour is often qualitatively described to help differentiate between tumor types. A smooth round outline is generally thought to be indicative of a benign or slowly growing lesion. A malignant neoplasm, on the other hand, tends to have disorganised growth and has a lobulated or spiculated margin. In this manuscript, we introduce a new computed tomography-based pipeline for tumour analysis and characterisation. This method yields an easily interpretable histogram-based index of lobularity. The EdgeRunner Pipeline performs equally well in studies of individual subjects, or in population-based assessments. Importantly, the methods can be easily translated to clinical use.

[1]  Matthew S Davenport,et al.  Histogram analysis of small solid renal masses: differentiating minimal fat angiomyolipoma from renal cell carcinoma. , 2012, AJR. American journal of roentgenology.

[2]  R. Rangayyan,et al.  Boundary modelling and shape analysis methods for classification of mammographic masses , 2000, Medical and Biological Engineering and Computing.

[3]  R Kloeckner,et al.  Effect of kernels used for the reconstruction of MDCT datasets on the semi-automated segmentation and volumetry of liver lesions. , 2014, RoFo : Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedizin.

[4]  Chih-Yen Chen,et al.  Computer-aided diagnosis of soft-tissue tumors using sonographic morphologic and texture features. , 2009, Academic radiology.

[5]  A. Giaccia,et al.  The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. , 1998, Cancer research.

[6]  Yu Guo,et al.  Automatic Lung Tumor Segmentation on PET/CT Images Using Fuzzy Markov Random Field Model , 2014, Comput. Math. Methods Medicine.

[7]  D. Ambrosi,et al.  On the mechanics of a growing tumor , 2002 .

[8]  J. Bruzzi,et al.  Liver biopsy. , 2001, The New England journal of medicine.

[9]  K. Han,et al.  Breast lesions on sonograms: computer-aided diagnosis with nearly setting-independent features and artificial neural networks. , 2003, Radiology.

[10]  V. Roggli,et al.  Solitary pulmonary nodules: Part I. Morphologic evaluation for differentiation of benign and malignant lesions. , 2000, Radiographics : a review publication of the Radiological Society of North America, Inc.

[11]  W. Peng,et al.  Intra-abdominal fibromatosis: differentiation from gastrointestinal stromal tumour based on biphasic contrast-enhanced CT findings. , 2013, Clinical radiology.

[12]  A. Burroughs,et al.  Seeding following percutaneous diagnostic and therapeutic approaches for hepatocellular carcinoma. What is the risk and the outcome? Seeding risk for percutaneous approach of HCC. , 2007, Cancer treatment reviews.

[13]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[14]  G. Ram Mohana Reddy,et al.  A Novel CBIR Approach to Differential Diagnosis of Liver Tumor on Computed Tomography Images , 2012 .

[15]  Mitsuru Ikeda,et al.  Computer-aided differentiation of malignant from benign solitary pulmonary nodules imaged by high-resolution CT , 2008, Comput. Medical Imaging Graph..

[16]  Maciej A Mazurowski,et al.  A fully automatic extraction of magnetic resonance image features in glioblastoma patients. , 2014, Medical physics.

[17]  Rangaraj M. Rangayyan,et al.  Polygonal Modeling of Contours of Breast Tumors With the Preservation of Spicules , 2008, IEEE Transactions on Biomedical Engineering.

[18]  Steven S Raman,et al.  Qualitative and quantitative MDCT features for differentiating clear cell renal cell carcinoma from other solid renal cortical masses. , 2014, AJR. American journal of roentgenology.

[19]  Stein Olav Skrøvseth,et al.  A semiautomatic tool for prostate segmentation in radiotherapy treatment planning , 2014, BMC Medical Imaging.

[20]  Turner Whitted,et al.  An improved illumination model for shaded display , 1979, CACM.

[21]  Michael D. Abràmoff,et al.  Image processing with ImageJ , 2004 .

[22]  Michalis A. Savelonas,et al.  A computer-aided system for malignancy risk assessment of nodules in thyroid US images based on boundary features , 2009, Comput. Methods Programs Biomed..

[23]  Anthony E. Samir,et al.  Pixel distribution analysis: can it be used to distinguish clear cell carcinomas from angiomyolipomas with minimal fat? , 2008, Radiology.

[24]  Hamid Soltanian-Zadeh,et al.  Atlas-based fiber bundle segmentation using principal diffusion directions and spherical harmonic coefficients , 2011, NeuroImage.

[25]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Chokri Ben Amar,et al.  Alzheimer's disease diagnosis on structural MR images using circular harmonic functions descriptors on hippocampus and posterior cingulate cortex , 2015, Comput. Medical Imaging Graph..