Real-time tumor ablation simulation based on the dynamic mode decomposition method

Handle URI:
http://hdl.handle.net/10754/563534
Title:
Real-time tumor ablation simulation based on the dynamic mode decomposition method
Authors:
Bourantas, George C.; Ghommem, Mehdi; Kagadis, George C.; Katsanos, Konstantinos H.; Loukopoulos, Vassilios C.; Burganos, Vasilis N.; Nikiforidis, George C.
Abstract:
Purpose: The dynamic mode decomposition (DMD) method is used to provide a reliable forecasting of tumor ablation treatment simulation in real time, which is quite needed in medical practice. To achieve this, an extended Pennes bioheat model must be employed, taking into account both the water evaporation phenomenon and the tissue damage during tumor ablation. Methods: A meshless point collocation solver is used for the numerical solution of the governing equations. The results obtained are used by the DMD method for forecasting the numerical solution faster than the meshless solver. The procedure is first validated against analytical and numerical predictions for simple problems. The DMD method is then applied to three-dimensional simulations that involve modeling of tumor ablation and account for metabolic heat generation, blood perfusion, and heat ablation using realistic values for the various parameters. Results: The present method offers very fast numerical solution to bioheat transfer, which is of clinical significance in medical practice. It also sidesteps the mathematical treatment of boundaries between tumor and healthy tissue, which is usually a tedious procedure with some inevitable degree of approximation. The DMD method provides excellent predictions of the temperature profile in tumors and in the healthy parts of the tissue, for linear and nonlinear thermal properties of the tissue. Conclusions: The low computational cost renders the use of DMD suitable forin situ real time tumor ablation simulations without sacrificing accuracy. In such a way, the tumor ablation treatment planning is feasible using just a personal computer thanks to the simplicity of the numerical procedure used. The geometrical data can be provided directly by medical image modalities used in everyday practice. © 2014 American Association of Physicists in Medicine.
KAUST Department:
Numerical Porous Media SRI Center (NumPor); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Publisher:
American Association of Physicists in Medicine (AAPM)
Journal:
Medical Physics
Issue Date:
May-2014
DOI:
10.1118/1.4870976
PubMed ID:
24784405
Type:
Article
ISSN:
00942405
Appears in Collections:
Articles; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorBourantas, George C.en
dc.contributor.authorGhommem, Mehdien
dc.contributor.authorKagadis, George C.en
dc.contributor.authorKatsanos, Konstantinos H.en
dc.contributor.authorLoukopoulos, Vassilios C.en
dc.contributor.authorBurganos, Vasilis N.en
dc.contributor.authorNikiforidis, George C.en
dc.date.accessioned2015-08-03T11:53:51Zen
dc.date.available2015-08-03T11:53:51Zen
dc.date.issued2014-05en
dc.identifier.issn00942405en
dc.identifier.pmid24784405en
dc.identifier.doi10.1118/1.4870976en
dc.identifier.urihttp://hdl.handle.net/10754/563534en
dc.description.abstractPurpose: The dynamic mode decomposition (DMD) method is used to provide a reliable forecasting of tumor ablation treatment simulation in real time, which is quite needed in medical practice. To achieve this, an extended Pennes bioheat model must be employed, taking into account both the water evaporation phenomenon and the tissue damage during tumor ablation. Methods: A meshless point collocation solver is used for the numerical solution of the governing equations. The results obtained are used by the DMD method for forecasting the numerical solution faster than the meshless solver. The procedure is first validated against analytical and numerical predictions for simple problems. The DMD method is then applied to three-dimensional simulations that involve modeling of tumor ablation and account for metabolic heat generation, blood perfusion, and heat ablation using realistic values for the various parameters. Results: The present method offers very fast numerical solution to bioheat transfer, which is of clinical significance in medical practice. It also sidesteps the mathematical treatment of boundaries between tumor and healthy tissue, which is usually a tedious procedure with some inevitable degree of approximation. The DMD method provides excellent predictions of the temperature profile in tumors and in the healthy parts of the tissue, for linear and nonlinear thermal properties of the tissue. Conclusions: The low computational cost renders the use of DMD suitable forin situ real time tumor ablation simulations without sacrificing accuracy. In such a way, the tumor ablation treatment planning is feasible using just a personal computer thanks to the simplicity of the numerical procedure used. The geometrical data can be provided directly by medical image modalities used in everyday practice. © 2014 American Association of Physicists in Medicine.en
dc.publisherAmerican Association of Physicists in Medicine (AAPM)en
dc.subjectbioheat equationen
dc.subjectEulerianen
dc.subjectmeshless methoden
dc.subjectmoving least squaresen
dc.subjectthermal ablationen
dc.titleReal-time tumor ablation simulation based on the dynamic mode decomposition methoden
dc.typeArticleen
dc.contributor.departmentNumerical Porous Media SRI Center (NumPor)en
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalMedical Physicsen
dc.contributor.institutionMax Planck Inst Mol Cell Biol & Genet, MOSAIC Grp, D-01307 Dresden, Germanyen
dc.contributor.institutionUniv Patras, Sch Med, Dept Med Phys, GR-26504 Rion, Greeceen
dc.contributor.institutionUniv Texas MD Anderson Canc Ctr, Dept Imaging Phys, Houston, TX 77030 USAen
dc.contributor.institutionSt Thomas Hosp, Kings Coll London, Div Endovascular Spine & Intervent Oncol, London SE1 7EH, Englanden
dc.contributor.institutionUniv Patras, Dept Phys, Rion 26500, Greeceen
dc.contributor.institutionFdn Res & Technol, Inst Chem Engn Sci, Patras 26504, Greeceen
kaust.authorGhommem, Mehdien

Related articles on PubMed

All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.