Literature DB >> 33889658

Deep-learning-based direct synthesis of low-energy virtual monoenergetic images with multi-energy CT.

Hao Gong1, Jeffrey F Marsh1, Karen N D'Souza1, Nathan R Huber1, Kishore Rajendran1, Joel G Fletcher1, Cynthia H McCollough1, Shuai Leng1.   

Abstract

Purpose: We developed a deep learning method to reduce noise and beam-hardening artifact in virtual monoenergetic image (VMI) at low x-ray energy levels. Approach: An encoder-decoder type convolutional neural network was implemented with customized inception modules and in-house-designed training loss (denoted as Incept-net), to directly estimate VMI from multi-energy CT images. Images of an abdomen-sized water phantom with varying insert materials were acquired from a research photon-counting-detector CT. The Incept-net was trained with image patches ( 64 × 64    pixels ) extracted from the phantom data, as well as synthesized, random-shaped numerical insert materials. The whole CT images ( 512 × 512    pixels ) with the remaining real insert materials that were unseen in network training were used for testing. Seven contrast-enhanced abdominal CT exams were used for preliminary evaluation of Incept-net generalizability over anatomical background. Mean absolute percentage error (MAPE) was used to evaluate CT number accuracy.
Results: Compared to commercial VMI software, Incept-net largely suppressed beam-hardening artifact and reduced noise (53%) in phantom study. Incept-net presented comparable CT number accuracy at higher-density ( P -value [0.0625, 0.999]) and improved it at lower-density inserts ( P - value = 0.0313 ) with overall MAPE: Incept-net [2.9%, 4.6%]; commercial-VMI [6.7%, 10.9%]. In patient images, Incept-net suppressed beam-hardening artifact and reduced noise (up to 50%, P - value = 0.0156 ).
Conclusion: In this preliminary study, Incept-net presented the potential to improve low-energy VMI quality.
© 2021 Society of Photo-Optical Instrumentation Engineers (SPIE).

Entities:  

Keywords:  artifact reduction; convolutional neural network; deep learning; dual-energy CT; noise reduction; photon counting detector; virtual monoenergetic image

Year:  2021        PMID: 33889658      PMCID: PMC8054272          DOI: 10.1117/1.JMI.8.5.052104

Source DB:  PubMed          Journal:  J Med Imaging (Bellingham)        ISSN: 2329-4302


  31 in total

1.  Monochromatic image reconstruction by dual energy imaging allows half iodine load computed tomography coronary angiography.

Authors:  Patricia Carrascosa; Jonathon A Leipsic; Carlos Capunay; Alejandro Deviggiano; Javier Vallejos; Alejandro Goldsmit; Gaston A Rodriguez-Granillo
Journal:  Eur J Radiol       Date:  2015-06-24       Impact factor: 3.528

2.  Prior image constrained compressed sensing (PICCS): a method to accurately reconstruct dynamic CT images from highly undersampled projection data sets.

Authors:  Guang-Hong Chen; Jie Tang; Shuai Leng
Journal:  Med Phys       Date:  2008-02       Impact factor: 4.071

Review 3.  Dual-energy CT-based monochromatic imaging.

Authors:  Lifeng Yu; Shuai Leng; Cynthia H McCollough
Journal:  AJR Am J Roentgenol       Date:  2012-11       Impact factor: 3.959

4.  Initial experience with single-source dual-energy CT abdominal angiography and comparison with single-energy CT angiography: image quality, enhancement, diagnosis and radiation dose.

Authors:  Daniella F Pinho; Naveen M Kulkarni; Arun Krishnaraj; Sanjeeva P Kalva; Dushyant V Sahani
Journal:  Eur Radiol       Date:  2012-08-25       Impact factor: 5.315

Review 5.  Review of Clinical Applications for Virtual Monoenergetic Dual-Energy CT.

Authors:  Moritz H Albrecht; Thomas J Vogl; Simon S Martin; John W Nance; Taylor M Duguay; Julian L Wichmann; Carlo N De Cecco; Akos Varga-Szemes; Marly van Assen; Christian Tesche; U Joseph Schoepf
Journal:  Radiology       Date:  2019-09-10       Impact factor: 11.105

6.  Maximizing Iodine Contrast-to-Noise Ratios in Abdominal CT Imaging through Use of Energy Domain Noise Reduction and Virtual Monoenergetic Dual-Energy CT.

Authors:  Shuai Leng; Lifeng Yu; Joel G Fletcher; Cynthia H McCollough
Journal:  Radiology       Date:  2015-04-10       Impact factor: 11.105

7.  Reduced iodine load at CT pulmonary angiography with dual-energy monochromatic imaging: comparison with standard CT pulmonary angiography--a prospective randomized trial.

Authors:  Ren Yuan; William P Shuman; James P Earls; Cameron J Hague; Hina A Mumtaz; Andrew Scott-Moncrieff; Jennifer D Ellis; John R Mayo; Jonathon A Leipsic
Journal:  Radiology       Date:  2011-11-14       Impact factor: 11.105

8.  Spectral performance of a whole-body research photon counting detector CT: quantitative accuracy in derived image sets.

Authors:  Shuai Leng; Wei Zhou; Zhicong Yu; Ahmed Halaweish; Bernhard Krauss; Bernhard Schmidt; Lifeng Yu; Steffen Kappler; Cynthia McCollough
Journal:  Phys Med Biol       Date:  2017-08-21       Impact factor: 3.609

9.  Dual-energy CT of the pancreas: improved carcinoma-to-pancreas contrast with a noise-optimized monoenergetic reconstruction algorithm.

Authors:  Claudia Frellesen; Freia Fessler; Andrew D Hardie; Julian L Wichmann; Carlo N De Cecco; U Joseph Schoepf; J Matthias Kerl; Boris Schulz; Renate Hammerstingl; Thomas J Vogl; Ralf W Bauer
Journal:  Eur J Radiol       Date:  2015-07-19       Impact factor: 3.528

10.  Deep-learning-based direct inversion for material decomposition.

Authors:  Hao Gong; Shengzhen Tao; Kishore Rajendran; Wei Zhou; Cynthia H McCollough; Shuai Leng
Journal:  Med Phys       Date:  2020-10-30       Impact factor: 4.071

View more
  1 in total

1.  Improving coronary artery imaging in single source CT with cardiac motion correction using attention and spatial transformer based neural networks.

Authors:  Hao Gong; Zaki Ahmed; Thorne E Jamison; Joel G Fletcher; Cynthia H McCollough; Shuai Leng
Journal:  Proc SPIE Int Soc Opt Eng       Date:  2022-04-04
  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.