David Minarik1, Olof Enqvist2,3, Elin Trägårdh4,5. 1. Radiation Physics, Skåne University Hospital, Malmö, Sweden david.minarik@med.lu.se. 2. Eigenvision AB, Malmö, Sweden. 3. Department of Electrical Engineering, Chalmers University of Technology, Göteborg, Sweden. 4. Clinical Physiology and Nuclear Medicine, Skåne University Hospital, Malmö, Sweden; and. 5. Wallenberg Centre for Molecular Medicine, Lund University, Lund, Sweden.
Abstract
Scintillation camera images contain a large amount of Poisson noise. We have investigated whether noise can be removed in whole-body bone scans using convolutional neural networks (CNNs) trained with sets of noisy and noiseless images obtained by Monte Carlo simulation. Methods: Three CNNs were generated using 3 different sets of training images: simulated bone scan images, images of a cylindric phantom with hot and cold spots, and a mix of the first two. Each training set consisted of 40,000 noiseless and noisy image pairs. The CNNs were evaluated with simulated images of a cylindric phantom and simulated bone scan images. The mean squared error between filtered and true images was used as difference metric, and the coefficient of variation was used to estimate noise reduction. The CNNs were compared with gaussian and median filters. A clinical evaluation was performed in which the ability to detect metastases for CNN- and gaussian-filtered bone scans with half the number of counts was compared with standard bone scans. Results: The best CNN reduced the coefficient of variation by, on average, 92%, and the best standard filter reduced the coefficient of variation by 88%. The best CNN gave a mean squared error that was on average 68% and 20% better than the best standard filters, for the cylindric and bone scan images, respectively. The best CNNs for the cylindric phantom and bone scans were the dedicated CNNs. No significant differences in the ability to detect metastases were found between standard, CNN-, and gaussian-filtered bone scans. Conclusion: Noise can be removed efficiently regardless of noise level with little or no resolution loss. The CNN filter enables reducing the scanning time by half and still obtaining good accuracy for bone metastasis assessment.
Scintillation camera images contain a large amount of Poisson noise. We have investigated whether noise can be removed in whole-body bone scans using convolutional neural networks (CNNs) trained with sets of noisy and noiseless images obtained by Monte Carlo simulation. Methods: Three CNNs were generated using 3 different sets of training images: simulated bone scan images, images of a cylindric phantom with hot and cold spots, and a mix of the first two. Each training set consisted of 40,000 noiseless and noisy image pairs. The CNNs were evaluated with simulated images of a cylindric phantom and simulated bone scan images. The mean squared error between filtered and true images was used as difference metric, and the coefficient of variation was used to estimate noise reduction. The CNNs were compared with gaussian and median filters. A clinical evaluation was performed in which the ability to detect metastases for CNN- and gaussian-filtered bone scans with half the number of counts was compared with standard bone scans. Results: The best CNN reduced the coefficient of variation by, on average, 92%, and the best standard filter reduced the coefficient of variation by 88%. The best CNN gave a mean squared error that was on average 68% and 20% better than the best standard filters, for the cylindric and bone scan images, respectively. The best CNNs for the cylindric phantom and bone scans were the dedicated CNNs. No significant differences in the ability to detect metastases were found between standard, CNN-, and gaussian-filtered bone scans. Conclusion: Noise can be removed efficiently regardless of noise level with little or no resolution loss. The CNN filter enables reducing the scanning time by half and still obtaining good accuracy for bone metastasis assessment.
Authors: Aseem Anand; Michael J Morris; Reza Kaboteh; Lena Båth; May Sadik; Peter Gjertsson; Milan Lomsky; Lars Edenbrandt; David Minarik; Anders Bjartell Journal: J Nucl Med Date: 2015-08-27 Impact factor: 10.057
Authors: Aseem Anand; Michael J Morris; Reza Kaboteh; Mariana Reza; Elin Trägårdh; Naofumi Matsunaga; Lars Edenbrandt; Anders Bjartell; Steven M Larson; David Minarik Journal: J Nucl Med Date: 2016-07-21 Impact factor: 10.057
Authors: T Van den Wyngaert; K Strobel; W U Kampen; T Kuwert; W van der Bruggen; H K Mohan; G Gnanasegaran; R Delgado-Bolton; W A Weber; M Beheshti; W Langsteger; F Giammarile; F M Mottaghy; F Paycha Journal: Eur J Nucl Med Mol Imaging Date: 2016-06-04 Impact factor: 9.236
Authors: Johannes Schwenck; Manfred Kneilling; Niels P Riksen; Christian la Fougère; Douwe J Mulder; Riemer J H A Slart; Erik H J G Aarntzen Journal: Eur J Hybrid Imaging Date: 2022-09-01