Fredrik Langkilde1, Thiele Kobus2,3, Andriy Fedorov2, Ruth Dunne2, Clare Tempany2, Robert V Mulkern2,4, Stephan E Maier1,2. 1. Institute of Clinical Sciences, Sahlgrenska Academy, Gothenburg University, Gothenburg, Sweden. 2. Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA. 3. Department of Radiology and Nuclear Medicine, Radboud university medical center, Nijmegen, The Netherlands. 4. Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, Massachusetts, USA.
Abstract
PURPOSE: To compare the fitting and tissue discrimination performance of biexponential, kurtosis, stretched exponential, and gamma distribution models for high b-factor diffusion-weighted images in prostate cancer. METHODS: Diffusion-weighted images with 15 b-factors ranging from b = 0 to 3500 s/mm2 were obtained in 62 prostate cancer patients. Pixel-wise signal decay fits for each model were evaluated with the Akaike Information Criterion (AIC). Parameter values for each model were determined within normal prostate and the index lesion. Their potential to differentiate normal from cancerous tissue was investigated through receiver operating characteristic analysis and comparison with Gleason score. RESULTS: The biexponential slow diffusion fraction fslow , the apparent kurtosis diffusion coefficient ADCK , and the excess kurtosis factor K differ significantly among normal peripheral zone (PZ), normal transition zone (TZ), tumor PZ, and tumor TZ. Biexponential and gamma distribution models result in the lowest AIC, indicating a superior fit. Maximum areas under the curve (AUCs) of all models ranged from 0.93 to 0.96 for the PZ and from 0.95 to 0.97 for the TZ. Similar AUCs also result from the apparent diffusion coefficient (ADC) of a monoexponential fit to a b-factor sub-range up to 1250 s/mm2 . For kurtosis and stretched exponential models, single parameters yield the highest AUCs, whereas for the biexponential and gamma distribution models, linear combinations of parameters produce the highest AUCs. Parameters with high AUC show a trend in differentiating low from high Gleason score, whereas parameters with low AUC show no such ability. CONCLUSION: All models, including a monoexponential fit to a lower-b sub-range, achieve similar AUCs for discrimination of normal and cancer tissue. The biexponential model, which is favored statistically, also appears to provide insight into disease-related microstructural changes. Magn Reson Med 79:2346-2358, 2018.
PURPOSE: To compare the fitting and tissue discrimination performance of biexponential, kurtosis, stretched exponential, and gamma distribution models for high b-factor diffusion-weighted images in prostate cancer. METHODS: Diffusion-weighted images with 15 b-factors ranging from b = 0 to 3500 s/mm2 were obtained in 62 prostate cancerpatients. Pixel-wise signal decay fits for each model were evaluated with the Akaike Information Criterion (AIC). Parameter values for each model were determined within normal prostate and the index lesion. Their potential to differentiate normal from cancerous tissue was investigated through receiver operating characteristic analysis and comparison with Gleason score. RESULTS: The biexponential slow diffusion fraction fslow , the apparent kurtosis diffusion coefficient ADCK , and the excess kurtosis factor K differ significantly among normal peripheral zone (PZ), normal transition zone (TZ), tumor PZ, and tumorTZ. Biexponential and gamma distribution models result in the lowest AIC, indicating a superior fit. Maximum areas under the curve (AUCs) of all models ranged from 0.93 to 0.96 for the PZ and from 0.95 to 0.97 for the TZ. Similar AUCs also result from the apparent diffusion coefficient (ADC) of a monoexponential fit to a b-factor sub-range up to 1250 s/mm2 . For kurtosis and stretched exponential models, single parameters yield the highest AUCs, whereas for the biexponential and gamma distribution models, linear combinations of parameters produce the highest AUCs. Parameters with high AUC show a trend in differentiating low from high Gleason score, whereas parameters with low AUC show no such ability. CONCLUSION: All models, including a monoexponential fit to a lower-b sub-range, achieve similar AUCs for discrimination of normal and cancer tissue. The biexponential model, which is favored statistically, also appears to provide insight into disease-related microstructural changes. Magn Reson Med 79:2346-2358, 2018.
Authors: Kevin M Bennett; Kathleen M Schmainda; Raoqiong Tong Bennett; Daniel B Rowe; Hanbing Lu; James S Hyde Journal: Magn Reson Med Date: 2003-10 Impact factor: 4.668
Authors: Thomas L Chenevert; Craig J Galbán; Marko K Ivancevic; Susan E Rohrer; Frank J Londy; Thomas C Kwee; Charles R Meyer; Timothy D Johnson; Alnawaz Rehemtulla; Brian D Ross Journal: J Magn Reson Imaging Date: 2011-10 Impact factor: 4.813
Authors: Robert V Mulkern; Kelsey I Ricci; Sridhar Vajapeyam; Thomas L Chenevert; Dariya I Malyarenko; Mehmet Kocak; Tina Young Poussaint Journal: Acad Radiol Date: 2014-11-27 Impact factor: 3.173
Authors: Andrew B Rosenkrantz; Eric E Sigmund; Glyn Johnson; James S Babb; Thais C Mussi; Jonathan Melamed; Samir S Taneja; Vivian S Lee; Jens H Jensen Journal: Radiology Date: 2012-05-01 Impact factor: 11.105
Authors: Yuxi Pang; Baris Turkbey; Marcelino Bernardo; Jochen Kruecker; Samuel Kadoury; Maria J Merino; Bradford J Wood; Peter A Pinto; Peter L Choyke Journal: Magn Reson Med Date: 2012-04-09 Impact factor: 4.668
Authors: Baris Turkbey; Anna M Brown; Sandeep Sankineni; Bradford J Wood; Peter A Pinto; Peter L Choyke Journal: CA Cancer J Clin Date: 2015-11-23 Impact factor: 508.702
Authors: John V Hegde; Robert V Mulkern; Lawrence P Panych; Fiona M Fennessy; Andriy Fedorov; Stephan E Maier; Clare M C Tempany Journal: J Magn Reson Imaging Date: 2013-05 Impact factor: 4.813
Authors: Pelin Aksit Ciris; Jr-Yuan George Chiou; Daniel I Glazer; Tzu-Cheng Chao; Clare M Tempany-Afdhal; Bruno Madore; Stephan E Maier Journal: Invest Radiol Date: 2019-04 Impact factor: 6.016
Authors: Sean D McGarry; Michael Brehler; John D Bukowy; Allison K Lowman; Samuel A Bobholz; Savannah R Duenweg; Anjishnu Banerjee; Sarah L Hurrell; Dariya Malyarenko; Thomas L Chenevert; Yue Cao; Yuan Li; Daekeun You; Andrey Fedorov; Laura C Bell; C Chad Quarles; Melissa A Prah; Kathleen M Schmainda; Bachir Taouli; Eve LoCastro; Yousef Mazaheri; Amita Shukla-Dave; Thomas E Yankeelov; David A Hormuth; Ananth J Madhuranthakam; Keith Hulsey; Kurt Li; Wei Huang; Wei Huang; Mark Muzi; Michael A Jacobs; Meiyappan Solaiyappan; Stefanie Hectors; Tatjana Antic; Gladell P Paner; Watchareepohn Palangmonthip; Kenneth Jacobsohn; Mark Hohenwalter; Petar Duvnjak; Michael Griffin; William See; Marja T Nevalainen; Kenneth A Iczkowski; Peter S LaViolette Journal: J Magn Reson Imaging Date: 2021-11-12 Impact factor: 5.119
Authors: Björn J Langbein; Filip Szczepankiewicz; Carl-Fredrik Westin; Camden Bay; Stephan E Maier; Adam S Kibel; Clare M Tempany; Fiona M Fennessy Journal: Invest Radiol Date: 2021-12-01 Impact factor: 6.016
Authors: Andrey Fedorov; Reinhard Beichel; Jayashree Kalpathy-Cramer; David Clunie; Michael Onken; Jörg Riesmeier; Christian Herz; Christian Bauer; Andrew Beers; Jean-Christophe Fillion-Robin; Andras Lasso; Csaba Pinter; Steve Pieper; Marco Nolden; Klaus Maier-Hein; Markus D Herrmann; Joel Saltz; Fred Prior; Fiona Fennessy; John Buatti; Ron Kikinis Journal: JCO Clin Cancer Inform Date: 2020-05