Zabihollahy Fatemeh1, Schieda Nicola2, Krishna Satheesh3, Ukwatta Eranga4. 1. Department of Systems and Computer Engineering, Carleton University, Ottawa, ON, Canada. 2. Department of Radiology, University of Ottawa, Ottawa, ON, Canada. 3. Department of Medical Imaging, University of Toronto, Toronto, ON, Canada. 4. School of Engineering, University of Guelph, Guelph, ON, Canada.
Abstract
PURPOSE: Detection and accurate localization of renal masses (RM) are important steps toward future potential classification of benign vs malignant RM. A fully automated algorithm for detection and localization of RM may eliminate the observer variability in the clinical workflow. METHOD: In this paper, we describe a fully automated methodology for accurate detection and segmentation of RM from contrast-enhanced computed tomography (CECT) images. We first determine the boundaries of the kidneys on the CECT images utilizing a convolutional neural network-based method to be used as a region of interest to search for RM. We then employ a homogenous U-Net-based ensemble learning model to identify and delineate RM. We used an institutional dataset comprised of CECT images in 315 patients to train and evaluate the proposed method. We compared results of our method to those of three-dimensional (3D) U-Net for RM localization and further evaluated our algorithm using the kidney tumor segmentation (KiTS19) challenge dataset. RESULTS: The developed algorithm reported a Dice similarity coefficient (DSC) of 95.79% ± 5.16% and 96.25 ± 3.37 (mean ± standard deviation) for segmentation accuracy of kidney boundary from 125 and 60 test images from institutional and KiTS19 datasets, respectively. Using our method, RM were detected in 125 and 52 test cases, which corresponds to 100% and 86.67% sensitivity at patient level in institutional and KiTS19 test images. Our ensemble method for RM localization yielded a mean DSC of 88.65% ± 7.31% and 87.91% ± 6.82% on the institutional and KiTS19 test datasets, respectively. The mean DSC for RM delineation from CECT institutional test images using 3D U-Net was 85.95% ± 1.46%. CONCLUSION: We describe a method for automated localization of RM using CECT images. Our results are important in terms of clinical perspective as fully automated detection of RM is a fundamental step for further diagnosis of cystic vs solid RM and eventually benign vs malignant solid RM, that has not been reported previously.
PURPOSE: Detection and accurate localization of renal masses (RM) are important steps toward future potential classification of benign vs malignant RM. A fully automated algorithm for detection and localization of RM may eliminate the observer variability in the clinical workflow. METHOD: In this paper, we describe a fully automated methodology for accurate detection and segmentation of RM from contrast-enhanced computed tomography (CECT) images. We first determine the boundaries of the kidneys on the CECT images utilizing a convolutional neural network-based method to be used as a region of interest to search for RM. We then employ a homogenous U-Net-based ensemble learning model to identify and delineate RM. We used an institutional dataset comprised of CECT images in 315 patients to train and evaluate the proposed method. We compared results of our method to those of three-dimensional (3D) U-Net for RM localization and further evaluated our algorithm using the kidney tumor segmentation (KiTS19) challenge dataset. RESULTS: The developed algorithm reported a Dice similarity coefficient (DSC) of 95.79% ± 5.16% and 96.25 ± 3.37 (mean ± standard deviation) for segmentation accuracy of kidney boundary from 125 and 60 test images from institutional and KiTS19 datasets, respectively. Using our method, RM were detected in 125 and 52 test cases, which corresponds to 100% and 86.67% sensitivity at patient level in institutional and KiTS19 test images. Our ensemble method for RM localization yielded a mean DSC of 88.65% ± 7.31% and 87.91% ± 6.82% on the institutional and KiTS19 test datasets, respectively. The mean DSC for RM delineation from CECT institutional test images using 3D U-Net was 85.95% ± 1.46%. CONCLUSION: We describe a method for automated localization of RM using CECT images. Our results are important in terms of clinical perspective as fully automated detection of RM is a fundamental step for further diagnosis of cystic vs solid RM and eventually benign vs malignant solid RM, that has not been reported previously.
Authors: Narendra N Khanna; Mahesh Maindarkar; Anudeep Puvvula; Sudip Paul; Mrinalini Bhagawati; Puneet Ahluwalia; Zoltan Ruzsa; Aditya Sharma; Smiksha Munjral; Raghu Kolluri; Padukone R Krishnan; Inder M Singh; John R Laird; Mostafa Fatemi; Azra Alizad; Surinder K Dhanjil; Luca Saba; Antonella Balestrieri; Gavino Faa; Kosmas I Paraskevas; Durga Prasanna Misra; Vikas Agarwal; Aman Sharma; Jagjit Teji; Mustafa Al-Maini; Andrew Nicolaides; Vijay Rathore; Subbaram Naidu; Kiera Liblik; Amer M Johri; Monika Turk; David W Sobel; Gyan Pareek; Martin Miner; Klaudija Viskovic; George Tsoulfas; Athanasios D Protogerou; Sophie Mavrogeni; George D Kitas; Mostafa M Fouda; Manudeep K Kalra; Jasjit S Suri Journal: J Cardiovasc Dev Dis Date: 2022-08-15
Authors: Danju Huang; Han Bai; Li Wang; Yu Hou; Lan Li; Yaoxiong Xia; Zhirui Yan; Wenrui Chen; Li Chang; Wenhui Li Journal: Technol Cancer Res Treat Date: 2021 Jan-Dec