Joshua M Goldenberg1,2, Julio Cárdenas-Rodríguez3, Mark D Pagel2. 1. Department of Pharmaceutical Sciences, University of Arizona, Tucson, Arizona. 2. Department of Cancer Systems Imaging, University of Texas MD Anderson Cancer Center, Houston, Texas. 3. Banner University Medical Center, University of Arizona, Tucson, Arizona.
Abstract
PURPOSE: We sought to assess whether machine learning-based classification approaches can improve the classification of pancreatic tumor models relative to more simplistic analysis methods, using T1 relaxation, CEST, and DCE MRI. METHODS: The T1 relaxation time constants, % CEST at five saturation frequencies, and vascular permeability constants from DCE MRI were measured from Hs 766 T, MIA PaCa-2, and SU.86.86 pancreatic tumor models. We used each of these measurements as predictors for machine learning classifier algorithms. We also used principal component analysis to reduce the dimensionality of entire CEST spectra and DCE signal evolutions, which were then analyzed using classification methods. RESULTS: The T1 relaxation time constants, % CEST amplitudes at specific saturation frequencies, and the relative Ktrans and kep values from DCE MRI could not classify all three tumor types. However, the area under the curve from DCE signal evolutions could classify each tumor type. Principal component analysis was used to analyze the entire CEST spectrum and DCE signal evolutions, which predicted the correct tumor model with 87.5% and 85.1% accuracy, respectively. CONCLUSIONS: Machine learning applied to the entire CEST spectrum improved the classification of the three tumor models, relative to classifications that used % CEST values at single saturation frequencies. A similar improvement was not attained with machine learning applied to T1 relaxation times or DCE signal evolutions, relative to more simplistic analysis methods.
PURPOSE: We sought to assess whether machine learning-based classification approaches can improve the classification of pancreatic tumor models relative to more simplistic analysis methods, using T1 relaxation, CEST, and DCE MRI. METHODS: The T1 relaxation time constants, % CEST at five saturation frequencies, and vascular permeability constants from DCE MRI were measured from Hs 766 T, MIA PaCa-2, and SU.86.86 pancreatic tumor models. We used each of these measurements as predictors for machine learning classifier algorithms. We also used principal component analysis to reduce the dimensionality of entire CEST spectra and DCE signal evolutions, which were then analyzed using classification methods. RESULTS: The T1 relaxation time constants, % CEST amplitudes at specific saturation frequencies, and the relative Ktrans and kep values from DCE MRI could not classify all three tumor types. However, the area under the curve from DCE signal evolutions could classify each tumor type. Principal component analysis was used to analyze the entire CEST spectrum and DCE signal evolutions, which predicted the correct tumor model with 87.5% and 85.1% accuracy, respectively. CONCLUSIONS: Machine learning applied to the entire CEST spectrum improved the classification of the three tumor models, relative to classifications that used % CEST values at single saturation frequencies. A similar improvement was not attained with machine learning applied to T1 relaxation times or DCE signal evolutions, relative to more simplistic analysis methods.
Authors: Miles N Wernick; Yongyi Yang; Jovan G Brankov; Grigori Yourganov; Stephen C Strother Journal: IEEE Signal Process Mag Date: 2010-07 Impact factor: 12.551
Authors: Xiaolei Song; Jiadi Xu; Shuli Xia; Nirbhay N Yadav; Bachchu Lal; John Laterra; Jeff W M Bulte; Peter C M van Zijl; Michael T McMahon Journal: Magn Reson Med Date: 2014-12-16 Impact factor: 4.668
Authors: Emily L Deer; Jessica González-Hernández; Jill D Coursen; Jill E Shea; Josephat Ngatia; Courtney L Scaife; Matthew A Firpo; Sean J Mulvihill Journal: Pancreas Date: 2010-05 Impact factor: 3.327
Authors: Moritz Zaiss; Johannes Windschuh; Steffen Goerke; Daniel Paech; Jan-Eric Meissner; Sina Burth; Philipp Kickingereder; Wolfgang Wick; Martin Bendszus; Heinz-Peter Schlemmer; Mark E Ladd; Peter Bachert; Alexander Radbruch Journal: Magn Reson Med Date: 2016-01-27 Impact factor: 4.668
Authors: Chongxue Bie; Yuguo Li; Yang Zhou; Zaver M Bhujwalla; Xiaolei Song; Guanshu Liu; Peter C M van Zijl; Nirbhay N Yadav Journal: NMR Biomed Date: 2021-10-19 Impact factor: 4.044