Archya Dasgupta1,2,3, Benjamin Geraghty1,2,3, Arjun Sahgal1,2,3, Gregory J Czarnota4,5,6,7, Pejman Jabehdar Maralani8,9, Nauman Malik1,2, Michael Sandhu3, Jay Detsky1,2,3, Chia-Lin Tseng1,2,3, Hany Soliman1,2,3, Sten Myrehaug1,2,3, Zain Husain1,2,3, James Perry10,11, Angus Lau3,12. 1. Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, Ontario, M4N3M5, Canada. 2. Department of Radiation Oncology, University of Toronto, Toronto, Canada. 3. Physical Sciences, Sunnybrook Research Institute, Toronto, Canada. 4. Department of Radiation Oncology, Sunnybrook Health Sciences Centre, 2075 Bayview Avenue, T2, Toronto, Ontario, M4N3M5, Canada. gregory.czarnota@sunnybrook.ca. 5. Department of Radiation Oncology, University of Toronto, Toronto, Canada. gregory.czarnota@sunnybrook.ca. 6. Physical Sciences, Sunnybrook Research Institute, Toronto, Canada. gregory.czarnota@sunnybrook.ca. 7. Department of Medical Biophysics, University of Toronto, Toronto, Canada. gregory.czarnota@sunnybrook.ca. 8. Department of Medical Imaging, Sunnybrook Health Sciences Centre, Toronto, Canada. 9. Department of Medical Imaging, University of Toronto, Toronto, Canada. 10. Department of Neurology, Sunnybrook Health Sciences Centre, Toronto, Canada. 11. Department of Medicine, University of Toronto, Toronto, Canada. 12. Department of Medical Biophysics, University of Toronto, Toronto, Canada.
Abstract
PURPOSE: The peritumoral region (PTR) in glioblastoma (GBM) represents a combination of infiltrative tumor and vasogenic edema, which are indistinguishable on magnetic resonance imaging (MRI). We developed a radiomic signature by using imaging data from low grade glioma (LGG) (marker of tumor) and PTR of brain metastasis (BM) (marker of edema) and applied it on the GBM PTR to generate probabilistic maps. METHODS: 270 features were extracted from T1-weighted, T2-weighted, and apparent diffusion coefficient maps in over 3.5 million voxels of LGG (36 segments) and BM (45 segments) scanned in a 1.5T MRI. A support vector machine classifier was used to develop the radiomics model from approximately 50% voxels (downsampled to 10%) and validated with the remaining. The model was applied to over 575,000 voxels of the PTR of 10 patients with GBM to generate a quantitative map using Platt scaling (infiltrative tumor vs. edema). RESULTS: The radiomics model had an accuracy of 0.92 and 0.79 in the training and test set, respectively (LGG vs. BM). When extrapolated on the GBM PTR, 9 of 10 patients had a higher percentage of voxels with a tumor-like signature over radiological recurrence areas. In 7 of 10 patients, the areas under curves (AUC) were > 0.50 confirming a positive correlation. Including all the voxels from the GBM patients, the infiltration signature had an AUC of 0.61 to predict recurrence. CONCLUSION: A radiomic signature can demarcate areas of microscopic tumors from edema in the PTR of GBM, which correlates with areas of future recurrence.
PURPOSE: The peritumoral region (PTR) in glioblastoma (GBM) represents a combination of infiltrative tumor and vasogenic edema, which are indistinguishable on magnetic resonance imaging (MRI). We developed a radiomic signature by using imaging data from low grade glioma (LGG) (marker of tumor) and PTR of brain metastasis (BM) (marker of edema) and applied it on the GBM PTR to generate probabilistic maps. METHODS: 270 features were extracted from T1-weighted, T2-weighted, and apparent diffusion coefficient maps in over 3.5 million voxels of LGG (36 segments) and BM (45 segments) scanned in a 1.5T MRI. A support vector machine classifier was used to develop the radiomics model from approximately 50% voxels (downsampled to 10%) and validated with the remaining. The model was applied to over 575,000 voxels of the PTR of 10 patients with GBM to generate a quantitative map using Platt scaling (infiltrative tumor vs. edema). RESULTS: The radiomics model had an accuracy of 0.92 and 0.79 in the training and test set, respectively (LGG vs. BM). When extrapolated on the GBM PTR, 9 of 10 patients had a higher percentage of voxels with a tumor-like signature over radiological recurrence areas. In 7 of 10 patients, the areas under curves (AUC) were > 0.50 confirming a positive correlation. Including all the voxels from the GBM patients, the infiltration signature had an AUC of 0.61 to predict recurrence. CONCLUSION: A radiomic signature can demarcate areas of microscopic tumors from edema in the PTR of GBM, which correlates with areas of future recurrence.
Authors: Ramon F Barajas; Joanna J Phillips; Rupa Parvataneni; Annette Molinaro; Emma Essock-Burns; Gabriela Bourne; Andrew T Parsa; Manish K Aghi; Michael W McDermott; Mitchel S Berger; Soonmee Cha; Susan M Chang; Sarah J Nelson Journal: Neuro Oncol Date: 2012-06-18 Impact factor: 12.300
Authors: Philippe Lambin; Emmanuel Rios-Velazquez; Ralph Leijenaar; Sara Carvalho; Ruud G P M van Stiphout; Patrick Granton; Catharina M L Zegers; Robert Gillies; Ronald Boellard; André Dekker; Hugo J W L Aerts Journal: Eur J Cancer Date: 2012-01-16 Impact factor: 9.162
Authors: Oliver Eidel; Sina Burth; Jan-Oliver Neumann; Pascal J Kieslich; Felix Sahm; Christine Jungk; Philipp Kickingereder; Sebastian Bickelhaupt; Sibu Mundiyanapurath; Philipp Bäumer; Wolfgang Wick; Heinz-Peter Schlemmer; Karl Kiening; Andreas Unterberg; Martin Bendszus; Alexander Radbruch Journal: PLoS One Date: 2017-01-19 Impact factor: 3.240
Authors: Ahmad Chaddad; Michael Jonathan Kucharczyk; Paul Daniel; Siham Sabri; Bertrand J Jean-Claude; Tamim Niazi; Bassam Abdulkarim Journal: Front Oncol Date: 2019-05-21 Impact factor: 6.244