• PDF / 774,590 Bytes
• 10 Pages / 595.224 x 790.955 pts Page_size
• 96 Downloads / 190 Views

DOWNLOAD

REPORT

ORIGINAL ARTICLE

A convolutional neural network for fully automated blood SUV determination to facilitate SUR computation in oncological FDG-PET ¨ 3,4,5 · Catharina Lange6 · Pavel Nikulin1 · Frank Hofheinz1 · Jens Maus1 · Yimin Li2 · Rebecca Butof 6 7,8 9 ¨ Kotzerke10 · Jorg ¨ van den Hoff1,10 Christian Furth · Sebastian Zschaeck · Michael C. Kreissl · Jorg Received: 22 May 2020 / Accepted: 5 August 2020 © The Author(s) 2020

Abstract Purpose The standardized uptake value (SUV) is widely used for quantitative evaluation in oncological FDG-PET but has well-known shortcomings as a measure of the tumor’s glucose consumption. The standard uptake ratio (SUR) of tumor SUV and arterial blood SUV (BSUV) possesses an increased prognostic value but requires image-based BSUV determination, typically in the aortic lumen. However, accurate manual ROI delineation requires care and imposes an additional workload, which makes the SUR approach less attractive for clinical routine. The goal of the present work was the development of a fully automated method for BSUV determination in whole-body PET/CT. Methods Automatic delineation of the aortic lumen was performed with a convolutional neural network (CNN), using the UNet architecture. A total of 946 FDG PET/CT scans from several sites were used for network training (N = 366) and testing (N = 580). For all scans, the aortic lumen was manually delineated, avoiding areas affected by motion-induced attenuation artifacts or potential spillover from adjacent FDG-avid regions. Performance of the network was assessed using the fractional deviations of automatically and manually derived BSUVs in the test data. Results The trained U-Net yields BSUVs in close agreement with those obtained from manual delineation. Comparison of manually and automatically derived BSUVs shows excellent concordance: the mean relative BSUV difference was (mean ± SD) = (– 0.5 ± 2.2)% with a 95% confidence interval of [−5.1, 3.8]% and a total range of [−10.0, 12.0]%. For four test cases, the derived ROIs were unusable (

Recommend Documents

Convolutional neural network for automated mass segmentation in mammography

212 7 5MB

A Novel Convolutional Neural Network for Continuous Blood Pressure Estimation

158 89 155MB

Automated Classification of Axial CT Slices Using Convolutional Neural Network

204 40 53MB

White Blood Cells Detection and Classification Using Convolutional Neural Network

174 19 114MB

Fully automated body composition analysis in routine CT imaging using 3D semantic segmentation convolutional neural netw

167 16 4MB

A Novel Keypoints-Based Image Registration Method with Fully Convolutional Neural Network

161 112 61MB

A Lightweight Fully Convolutional Neural Network of High Accuracy Surface Defect Detection

175 47 85MB

Convolutional Neural Network for Automatic MeSH Indexing

224 84 72MB

AutoGraph: Automated Graph Neural Network

176 64 95MB

A Shallow Convolutional Neural Network for Accurate Handwritten Digits Classification

184 73 1012KB

A Convolutional Neural Network Framework for Accurate Skin Cancer Detection

170 117 2MB

Transfer learning with deep convolutional neural network for automated detection of schizophrenia from EEG signals

182 95 844KB