Soudabeh Kargar1, Eric A Borisch2, Adam T Froemming2, Akira Kawashima3, Lance A Mynderse4, Eric G Stinson2, Joshua D Trzasko2, Stephen J Riederer5. 1. Biomedical Engineering and Physiology Program, Mayo Graduate School, Rochester, MN, United States; Department of Radiology, Mayo Clinic, Rochester, MN, United States. 2. Department of Radiology, Mayo Clinic, Rochester, MN, United States. 3. Department of Radiology, Mayo Clinic, Scottsdale, AZ, United States. 4. Department of Urology, Mayo Clinic, Rochester, MN, United States. 5. Biomedical Engineering and Physiology Program, Mayo Graduate School, Rochester, MN, United States; Department of Radiology, Mayo Clinic, Rochester, MN, United States. Electronic address: riederer@mayo.edu.
Abstract
PURPOSE: To describe an efficient numerical optimization technique using non-linear least squares to estimate perfusion parameters for the Tofts and extended Tofts models from dynamic contrast enhanced (DCE) MRI data and apply the technique to prostate cancer. METHODS: Parameters were estimated by fitting the two Tofts-based perfusion models to the acquired data via non-linear least squares. We apply Variable Projection (VP) to convert the fitting problem from a multi-dimensional to a one-dimensional line search to improve computational efficiency and robustness. Using simulation and DCE-MRI studies in twenty patients with suspected prostate cancer, the VP-based solver was compared against the traditional Levenberg-Marquardt (LM) strategy for accuracy, noise amplification, robustness to converge, and computation time. RESULTS: The simulation demonstrated that VP and LM were both accurate in that the medians closely matched assumed values across typical signal to noise ratio (SNR) levels for both Tofts models. VP and LM showed similar noise sensitivity. Studies using the patient data showed that the VP method reliably converged and matched results from LM with approximate 3× and 2× reductions in computation time for the standard (two-parameter) and extended (three-parameter) Tofts models. While LM failed to converge in 14% of the patient data, VP converged in the ideal 100%. CONCLUSION: The VP-based method for non-linear least squares estimation of perfusion parameters for prostate MRI is equivalent in accuracy and robustness to noise, while being more reliably (100%) convergent and computationally about 3× (TM) and 2× (ETM) faster than the LM-based method.
PURPOSE: To describe an efficient numerical optimization technique using non-linear least squares to estimate perfusion parameters for the Tofts and extended Tofts models from dynamic contrast enhanced (DCE) MRI data and apply the technique to prostate cancer. METHODS: Parameters were estimated by fitting the two Tofts-based perfusion models to the acquired data via non-linear least squares. We apply Variable Projection (VP) to convert the fitting problem from a multi-dimensional to a one-dimensional line search to improve computational efficiency and robustness. Using simulation and DCE-MRI studies in twenty patients with suspected prostate cancer, the VP-based solver was compared against the traditional Levenberg-Marquardt (LM) strategy for accuracy, noise amplification, robustness to converge, and computation time. RESULTS: The simulation demonstrated that VP and LM were both accurate in that the medians closely matched assumed values across typical signal to noise ratio (SNR) levels for both Tofts models. VP and LM showed similar noise sensitivity. Studies using the patient data showed that the VP method reliably converged and matched results from LM with approximate 3× and 2× reductions in computation time for the standard (two-parameter) and extended (three-parameter) Tofts models. While LM failed to converge in 14% of the patient data, VP converged in the ideal 100%. CONCLUSION: The VP-based method for non-linear least squares estimation of perfusion parameters for prostate MRI is equivalent in accuracy and robustness to noise, while being more reliably (100%) convergent and computationally about 3× (TM) and 2× (ETM) faster than the LM-based method.
Keywords:
Dynamic-contrast-enhanced magnetic resonance imaging; Multi-parametric magnetic resonance imaging; Perfusion; Pharmacokinetic modeling; Prostate cancer
Authors: Tobias Heye; Matthew S Davenport; Jeffrey J Horvath; Sebastian Feuerlein; Steven R Breault; Mustafa R Bashir; Elmar M Merkle; Daniel T Boll Journal: Radiology Date: 2012-12-06 Impact factor: 11.105
Authors: David Bonekamp; Michael A Jacobs; Riham El-Khouli; Dan Stoianovici; Katarzyna J Macura Journal: Radiographics Date: 2011 May-Jun Impact factor: 5.333
Authors: Tobias Heye; Elmar M Merkle; Caecilia S Reiner; Matthew S Davenport; Jeffrey J Horvath; Sebastian Feuerlein; Steven R Breault; Peter Gall; Mustafa R Bashir; Brian M Dale; Atilla P Kiraly; Daniel T Boll Journal: Radiology Date: 2012-12-06 Impact factor: 11.105
Authors: Urban Simoncic; Sara Leibfarth; Stefan Welz; Nina Schwenzer; Holger Schmidt; Gerald Reischl; Christina Pfannenberg; Christian la Fougère; Konstantin Nikolaou; Daniel Zips; Daniela Thorwarth Journal: Med Phys Date: 2017-04-20 Impact factor: 4.071
Authors: Fatemeh Adelnia; Michelle Shardell; Christopher M Bergeron; Kenneth W Fishbein; Richard G Spencer; Luigi Ferrucci; David A Reiter Journal: NMR Biomed Date: 2019-03-12 Impact factor: 4.044
Authors: Kay Pepin; Roger Grimm; Soudabeh Kargar; B Matthew Howe; Karen Fritchie; Matthew Frick; Doris Wenger; Scott Okuno; Richard Ehman; Kiaran McGee; Sarah James; Nadia Laack; Michael Herman; Deanna Pafundi Journal: Biomed Phys Eng Express Date: 2019-05-22
Authors: Amanina Ahmad Safri; Che Mohd Nasril Che Mohd Nassir; Ismail Nurul Iman; Nur Hartini Mohd Taib; Anusha Achuthan; Muzaimi Mustapha Journal: World J Clin Cases Date: 2022-08-26 Impact factor: 1.534