Affordable Access

deepdyve-link
Publisher Website

Noise reduction using novel loss functions to compute tissue mineral density and trabecular bone volume fraction on low resolution QCT.

Authors
  • Thomsen, Felix S L1
  • Delrieux, Claudio A2
  • Pisula, Juan I3
  • Fuertes García, José M4
  • Lucena, Manuel5
  • de Luis García, Rodrigo6
  • Borggrefe, Jan7
  • 1 Departamento de Ingeniería Eléctrica y Computadoras, San Andrés 800, 8000 Bahía Blanca, Argentina. Electronic address: [email protected] , (Argentina)
  • 2 Departamento de Ingeniería Eléctrica y Computadoras, San Andrés 800, 8000 Bahía Blanca, Argentina. Electronic address: [email protected] , (Argentina)
  • 3 Departamento de Ingeniería Eléctrica y Computadoras, San Andrés 800, 8000 Bahía Blanca, Argentina. Electronic address: [email protected] , (Argentina)
  • 4 Campus Las Lagunillas, Edificio Centros de Investigación (C6), 23071 Jaén, Spain. Electronic address: [email protected] , (Spain)
  • 5 Campus Las Lagunillas, Edificio Centros de Investigación (C6), 23071 Jaén, Spain. Electronic address: [email protected] , (Spain)
  • 6 ETSI Telecomunicación, Campus Miguel Delibes, 47011 Valladolid, Spain. Electronic address: [email protected] , (Spain)
  • 7 Universitätsinstitut für Radiologie, Neuroradiologie und Nuklearmedizin, Hans-Nolte-Str. 1, 32429 Minden, Germany. Electronic address: [email protected] , (Germany)
Type
Published Article
Journal
Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society
Publication Date
Dec 01, 2020
Volume
86
Pages
101816–101816
Identifiers
DOI: 10.1016/j.compmedimag.2020.101816
PMID: 33221674
Source
Medline
Keywords
Language
English
License
Unknown

Abstract

Micro-structural parameters of the thoracic or lumbar spine generally carry insufficient accuracy and precision for clinical in vivo studies when assessed on quantitative computed tomography (QCT). We propose a 3D convolutional neural network with specific loss functions for QCT noise reduction to compute micro-structural parameters such as tissue mineral density (TMD) and bone volume ratio (BV/TV) with significantly higher accuracy than using no or standard noise reduction filters. The vertebra-phantom study contained high resolution peripheral and clinical CT scans with simulated in vivo CT noise and nine repetitions of three different tube currents (100, 250 and 360 mAs). Five-fold cross validation was performed on 20466 purely spongy pairs of noisy and ground-truth patches. Comparison of training and test errors revealed high robustness against over-fitting. While not showing effects for the assessment of BMD and voxel-wise densities, the filter improved thoroughly the computation of TMD and BV/TV with respect to the unfiltered data. Root-mean-square and accuracy errors of low resolution TMD and BV/TV decreased to less than 17% of the initial values. Furthermore filtered low resolution scans revealed still more TMD- and BV/TV-relevant information than high resolution CT scans, either unfiltered or filtered with two state-of-the-art standard denoising methods. The proposed architecture is threshold and rotational invariant, applicable on a wide range of image resolutions at once, and likely serves for an accurate computation of further micro-structural parameters. Furthermore, it is less prone for over-fitting than neural networks that compute structural parameters directly. In conclusion, the method is potentially important for the diagnosis of osteoporosis and other bone diseases since it allows to assess relevant 3D micro-structural information from standard low exposure CT protocols such as 100 mAs and 120 kVp. Copyright © 2020 Elsevier Ltd. All rights reserved.

Report this publication

Statistics

Seen <100 times