Andrew T Grainger1, Arun Krishnaraj2, Michael H Quinones2, Nicholas J Tustison2, Samantha Epstein2, Daniela Fuller3, Aakash Jha3, Kevin L Allman3, Weibin Shi4. 1. Departments of Biochemistry & Molecular Genetics, Richmond, Virginia. 2. Radiology & Medical Imaging, School of Medicine, Virginia. 3. School of Engineering and Applied Science, University of Virginia, 480 Ray C. Hunt Drive, Charlottesville, VA 22908. 4. Departments of Biochemistry & Molecular Genetics, Richmond, Virginia; Radiology & Medical Imaging, School of Medicine, Virginia. Electronic address: ws4v@virginia.edu.
Abstract
RATIONALE AND OBJECTIVES: Develop a deep learning-based algorithm using the U-Net architecture to measure abdominal fat on computed tomography (CT) images. MATERIALS AND METHODS: Sequential CT images spanning the abdominal region of seven subjects were manually segmented to calculate subcutaneous fat (SAT) and visceral fat (VAT). The resulting segmentation maps of SAT and VAT were augmented using a template-based data augmentation approach to create a large dataset for neural network training. Neural network performance was evaluated on both sequential CT slices from three subjects and randomly selected CT images from the upper, central, and lower abdominal regions of 100 subjects. RESULTS: Both subcutaneous and abdominal cavity segmentation images created by the two methods were highly comparable with an overall Dice similarity coefficient of 0.94. Pearson's correlation coefficients between the subcutaneous and visceral fat volumes quantified using the two methods were 0.99 and 0.99 and the overall percent residual squared error were 5.5% and 8.5%. Manual segmentation of SAT and VAT on the 555 CT slices used for testing took approximately 46 hours while automated segmentation took approximately 1 minute. CONCLUSION: Our data demonstrates that deep learning methods utilizing a template-based data augmentation strategy can be employed to accurately and rapidly quantify total abdominal SAT and VAT with a small number of training images.
RATIONALE AND OBJECTIVES: Develop a deep learning-based algorithm using the U-Net architecture to measure abdominal fat on computed tomography (CT) images. MATERIALS AND METHODS: Sequential CT images spanning the abdominal region of seven subjects were manually segmented to calculate subcutaneous fat (SAT) and visceral fat (VAT). The resulting segmentation maps of SAT and VAT were augmented using a template-based data augmentation approach to create a large dataset for neural network training. Neural network performance was evaluated on both sequential CT slices from three subjects and randomly selected CT images from the upper, central, and lower abdominal regions of 100 subjects. RESULTS: Both subcutaneous and abdominal cavity segmentation images created by the two methods were highly comparable with an overall Dice similarity coefficient of 0.94. Pearson's correlation coefficients between the subcutaneous and visceral fat volumes quantified using the two methods were 0.99 and 0.99 and the overall percent residual squared error were 5.5% and 8.5%. Manual segmentation of SAT and VAT on the 555 CT slices used for testing took approximately 46 hours while automated segmentation took approximately 1 minute. CONCLUSION: Our data demonstrates that deep learning methods utilizing a template-based data augmentation strategy can be employed to accurately and rapidly quantify total abdominal SAT and VAT with a small number of training images.
Authors: Alexander D Weston; Panagiotis Korfiatis; Timothy L Kline; Kenneth A Philbrick; Petro Kostandy; Tomas Sakinis; Motokazu Sugimoto; Naoki Takahashi; Bradley J Erickson Journal: Radiology Date: 2018-12-11 Impact factor: 11.105
Authors: Johannes Schindelin; Ignacio Arganda-Carreras; Erwin Frise; Verena Kaynig; Mark Longair; Tobias Pietzsch; Stephan Preibisch; Curtis Rueden; Stephan Saalfeld; Benjamin Schmid; Jean-Yves Tinevez; Daniel James White; Volker Hartenstein; Kevin Eliceiri; Pavel Tomancak; Albert Cardona Journal: Nat Methods Date: 2012-06-28 Impact factor: 28.547
Authors: J K Snell-Bergeon; J E Hokanson; G L Kinney; D Dabelea; J Ehrlich; R H Eckel; L Ogden; M Rewers Journal: Int J Obes Relat Metab Disord Date: 2004-12