E D H Gates1,2, J S Weinberg3, S S Prabhu3, J S Lin1,4,5, J Hamilton6,7, J D Hazle1, G N Fuller8, V Baladandayuthapani9, D T Fuentes1, D Schellingerhout10,11. 1. From the Departments of Imaging Physics (E.D.H.G., J.S.L., J.D.H., D.T.F.). 2. The University of Texas MD Anderson Cancer Center UTHealth Graduate School of Biomedical Sciences (E.D.H.G.), Houston, Texas. 3. Neurosurgery (J.S.W., S.S.P.). 4. Baylor College of Medicine (J.S.L.), Houston, Texas. 5. Department of Bioengineering (J.S.L.), Rice University, Houston, Texas. 6. Neuroradiology (J.H., D.S.). 7. Radiology Partners (J.H.), Houston, Texas. 8. Pathology (G.N.F.). 9. Department of Computational Medicine and Bioinformatics (V.B.), University of Michigan School of Public Health, Ann Arbor, Michigan. 10. Neuroradiology (J.H., D.S.) Dawid.Schellingerhout@mdanderson.org. 11. Cancer Systems Imaging (D.S.), University of Texas MD Anderson Cancer Center, Houston, Texas.
Abstract
BACKGROUND AND PURPOSE: Increased cellular density is a hallmark of gliomas, both in the bulk of the tumor and in areas of tumor infiltration into surrounding brain. Altered cellular density causes altered imaging findings, but the degree to which cellular density can be quantitatively estimated from imaging is unknown. The purpose of this study was to discover the best MR imaging and processing techniques to make quantitative and spatially specific estimates of cellular density. MATERIALS AND METHODS: We collected stereotactic biopsies in a prospective imaging clinical trial targeting untreated patients with gliomas at our institution undergoing their first resection. The data included preoperative MR imaging with conventional anatomic, diffusion, perfusion, and permeability sequences and quantitative histopathology on biopsy samples. We then used multiple machine learning methodologies to estimate cellular density using local intensity information from the MR images and quantitative cellular density measurements at the biopsy coordinates as the criterion standard. RESULTS: The random forest methodology estimated cellular density with R 2 = 0.59 between predicted and observed values using 4 input imaging sequences chosen from our full set of imaging data (T2, fractional anisotropy, CBF, and area under the curve from permeability imaging). Limiting input to conventional MR images (T1 pre- and postcontrast, T2, and FLAIR) yielded slightly degraded performance (R2 = 0.52). Outputs were also reported as graphic maps. CONCLUSIONS: Cellular density can be estimated with moderate-to-strong correlations using MR imaging inputs. The random forest machine learning model provided the best estimates. These spatially specific estimates of cellular density will likely be useful in guiding both diagnosis and treatment.
BACKGROUND AND PURPOSE: Increased cellular density is a hallmark of gliomas, both in the bulk of the tumor and in areas of tumor infiltration into surrounding brain. Altered cellular density causes altered imaging findings, but the degree to which cellular density can be quantitatively estimated from imaging is unknown. The purpose of this study was to discover the best MR imaging and processing techniques to make quantitative and spatially specific estimates of cellular density. MATERIALS AND METHODS: We collected stereotactic biopsies in a prospective imaging clinical trial targeting untreated patients with gliomas at our institution undergoing their first resection. The data included preoperative MR imaging with conventional anatomic, diffusion, perfusion, and permeability sequences and quantitative histopathology on biopsy samples. We then used multiple machine learning methodologies to estimate cellular density using local intensity information from the MR images and quantitative cellular density measurements at the biopsy coordinates as the criterion standard. RESULTS: The random forest methodology estimated cellular density with R 2 = 0.59 between predicted and observed values using 4 input imaging sequences chosen from our full set of imaging data (T2, fractional anisotropy, CBF, and area under the curve from permeability imaging). Limiting input to conventional MR images (T1 pre- and postcontrast, T2, and FLAIR) yielded slightly degraded performance (R2 = 0.52). Outputs were also reported as graphic maps. CONCLUSIONS: Cellular density can be estimated with moderate-to-strong correlations using MR imaging inputs. The random forest machine learning model provided the best estimates. These spatially specific estimates of cellular density will likely be useful in guiding both diagnosis and treatment.
Authors: E D H Gates; J S Lin; J S Weinberg; S S Prabhu; J Hamilton; J D Hazle; G N Fuller; V Baladandayuthapani; D T Fuentes; D Schellingerhout Journal: AJNR Am J Neuroradiol Date: 2020-02-06 Impact factor: 3.825
Authors: Christopher R Durst; Prashant Raghavan; Mark E Shaffrey; David Schiff; M Beatriz Lopes; Jason P Sheehan; Nicholas J Tustison; James T Patrie; Wenjun Xin; W Jeff Elias; Kenneth C Liu; Greg A Helm; A Cupino; Max Wintermark Journal: Neuroradiology Date: 2013-12-15 Impact factor: 2.804
Authors: Evan D H Gates; Jonathan S Lin; Jeffrey S Weinberg; Jackson Hamilton; Sujit S Prabhu; John D Hazle; Gregory N Fuller; Veera Baladandayuthapani; David Fuentes; Dawid Schellingerhout Journal: Neuro Oncol Date: 2019-03-18 Impact factor: 12.300
Authors: Virendra Kumar; Yuhua Gu; Satrajit Basu; Anders Berglund; Steven A Eschrich; Matthew B Schabath; Kenneth Forster; Hugo J W L Aerts; Andre Dekker; David Fenstermacher; Dmitry B Goldgof; Lawrence O Hall; Philippe Lambin; Yoganand Balagurunathan; Robert A Gatenby; Robert J Gillies Journal: Magn Reson Imaging Date: 2012-08-13 Impact factor: 2.546
Authors: E D H Gates; D Suki; A Celaya; J S Weinberg; S S Prabhu; R Sawaya; J T Huse; J P Long; D Fuentes; D Schellingerhout Journal: AJNR Am J Neuroradiol Date: 2022-09-15 Impact factor: 4.966