Скачать презентацию XIX Symposium Neuroradiologicum Recent advent in DTI and Скачать презентацию XIX Symposium Neuroradiologicum Recent advent in DTI and

61454c76a425b29b1c61f52c0f45df4d.ppt

  • Количество слайдов: 46

XIX Symposium Neuroradiologicum Recent advent in DTI and tractography for neuro-oncology Kei Yamada M. XIX Symposium Neuroradiologicum Recent advent in DTI and tractography for neuro-oncology Kei Yamada M. D. [email protected] kpu-m. ac. jp Department of Radiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine

Overview Very basics of DTI Tractography Clinical application Future techniques Kei Yamada M. D. Overview Very basics of DTI Tractography Clinical application Future techniques Kei Yamada M. D. [email protected] kpu-m. ac. jp Department of Radiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine

Diffusion-weighted imaging • Characterizes diffusivity of water molecules • Using large bipolar gradients • Diffusion-weighted imaging • Characterizes diffusivity of water molecules • Using large bipolar gradients • Apparent diffusion coefficient (ADC) 90° 180° G Echo G d = 50 d D = 60 Kei Yamada

Diffusion of water molecules Size Shape Direction ADC Diffusion anisotropy Kei Yamada Diffusion of water molecules Size Shape Direction ADC Diffusion anisotropy Kei Yamada

Diffusion Tensor Image z x y Kei Yamada Diffusion Tensor Image z x y Kei Yamada

Diffusion Tensor Image Isotropic λ 2 Anisotropic λ 1 λ 3 Kei Yamada Diffusion Tensor Image Isotropic λ 2 Anisotropic λ 1 λ 3 Kei Yamada

Diffusion Tensor Image Isotropic λ 2 Anisotropic λ 1 λ 3 Fractional anisotropy Arbitrary Diffusion Tensor Image Isotropic λ 2 Anisotropic λ 1 λ 3 Fractional anisotropy Arbitrary Unit ( 0≦FA≦ 1) Apparent Diffusion Coefficient ( 2/s) mm Kei Yamada

ADC and FA of brain tumors • ADC correlates with cell density and prognosis ADC and FA of brain tumors • ADC correlates with cell density and prognosis – Guo et al. Radiology 2002 (11 pts lymphoma, 17 with astrocytoma) – Murakami et al. Radiology 2007 (79 pts, grade III and IV) • FA is better than ADC in assessing brain tumors – Stadlbauer et al. Radiology 2006 • FA correlation; mixed results – Beppu et al. Surg. Neurol 2005 – Kinoshita et al. Neuro. Image 2008 Kei Yamada

Stadlbauer et al. Radiology 2006 Kei Yamada Stadlbauer et al. Radiology 2006 Kei Yamada

Kinoshita et al. Neuro. Image 2008 Kei Yamada Kinoshita et al. Neuro. Image 2008 Kei Yamada

Neuro. Image 2008 Radiology 2006 Kei Yamada Neuro. Image 2008 Radiology 2006 Kei Yamada

Precontrast DWI (b = 1000)   b = 0 s/mm 2 Postcontrast   ADC Can Precontrast DWI (b = 1000)   b = 0 s/mm 2 Postcontrast   ADC Can we perform DWI after contrast administration? Kei Yamada 2005/2

Precontrast DWI (b = 1000)   b = 0 s/mm 2   ADC Postcontrast Kei Precontrast DWI (b = 1000)   b = 0 s/mm 2   ADC Postcontrast Kei Yamada et al. Stroke 2002; 33: 1799 Yamada 2005/2

DWI after contrast • Fırat et al. 2006 • Substantial ADC change after contrast DWI after contrast • Fırat et al. 2006 • Substantial ADC change after contrast admin. • Statistically significant difference between pre- and post-contrast ADC values – Normal brain tissue (1. 0%; p < 0. 049) – Lesions (2. 5%; p < 0. 008). Kei Yamada

Significant difference of FA at the tumor (p = 0. 0273). Bae et al. Significant difference of FA at the tumor (p = 0. 0273). Bae et al. Neuroradiology (2009) Kei Yamada

3 -D data set of diffusion anisotropy The software connects between pixels that are 3 -D data set of diffusion anisotropy The software connects between pixels that are similar in directions. Kei Yamada

Gd-T 1 WI T 2 WI Sensory Motor 64 y. o. M Metastasis (lung Gd-T 1 WI T 2 WI Sensory Motor 64 y. o. M Metastasis (lung ca. ) 術後 Sensory Motor  Kei Yamada

Lee SK, et al. 2005 Kei Yamada Lee SK, et al. 2005 Kei Yamada

Validation of tractography • Correlation with operative results and/or navigation – Kamada et al. Validation of tractography • Correlation with operative results and/or navigation – Kamada et al. J Neurosurgery 2005 – Nimsky et al. Neuro. Image 2006 – Okada et al. Radiology 2006 • Most of them report “successful results” – Indicating that it is helful for safer surgery • “Negative results” or warnings are scarce • A paper with scary title – Fiber-tracking does not accurately estimate size of fiber bundle in pathological condition – Neuro. Image 2005 Kei Yamada

axial b = 0 Navigation Color vector Red dots; tip of the sucker tube axial b = 0 Navigation Color vector Red dots; tip of the sucker tube Positive reaction by MEP right hemisphere Kinoshita M, Yamada K et al. Neuro. Image 2005 67 y. o. F Glioblastoma Kei Yamada

axial b = 0 Navigation right hemisphere Kinoshita M, Yamada K et al. Neuro. axial b = 0 Navigation right hemisphere Kinoshita M, Yamada K et al. Neuro. Image 2005 Color vector 67 y. o. F Glioblastoma Kei Yamada

Tractography MR acquisition Evolutions of DTI/tractography DTI (6 -axes) [Basser 94] HARDI (> 15 Tractography MR acquisition Evolutions of DTI/tractography DTI (6 -axes) [Basser 94] HARDI (> 15 -axes) Q-space (multi-axes, b) Q-ball (multi-axes) [Basser 02] [Tuch 02] [Mori 99] FACT Following e 1 TEND Following e 1 T Multi-tensor Probabilistic [Jones 03] Kei Yamada

Probabilistic tractography • Cones of uncertainty • Will allow tracking beyond voxels with low Probabilistic tractography • Cones of uncertainty • Will allow tracking beyond voxels with low anisotropy • Enables depiction of • crossing fibers • branching fibers • Resistant to noise Jones. MRM 2003 Kei Yamada

Downside of using probabilistic tractography • Takes a lot of time for postprocessing – Downside of using probabilistic tractography • Takes a lot of time for postprocessing – Cannot be used in interactively – Cannot be used for intraoperative tractography • Visual assessment is more difficult – You will find unexpected fibers Kei Yamada

Dual stream model for language processing Dorsal stream; Arcuate fasciculus and superior longitudinal fasciculus Dual stream model for language processing Dorsal stream; Arcuate fasciculus and superior longitudinal fasciculus Ventral stream; runs through extreme capsule (Em. C) Confirmed on autoradiographic tracing studies in monkeys Kei Yamada

b=0 Vector map EC EC Em. C FA map b=0 Vector map EC EC Em. C FA map

Kei Yamada Kei Yamada

Tractography MR acquisition Evolutions of DTI/tractography DTI (6 -axes) [Basser 94] HARDI (> 15 Tractography MR acquisition Evolutions of DTI/tractography DTI (6 -axes) [Basser 94] HARDI (> 15 -axes) [Mori 99] FACT Following e 1 TEND Following e 1 T Multi-tensor Probabilistic [Jones 03] Kei Yamada

Voxels with crossings * start FACT algorithm FA Angle-αbetween tensors < 30 ° Yamada Voxels with crossings * start FACT algorithm FA Angle-αbetween tensors < 30 ° Yamada et al. AJNR 2008 Yamada Kei

* start FACT algorithm FA Angle-αbetween tensors < 30 ° Yamada et al. AJNR * start FACT algorithm FA Angle-αbetween tensors < 30 ° Yamada et al. AJNR 2008 Yamada Kei

Tractography by multi-tensor ROI placement Tensor model Multi-tensor model 42 y. o. male with Tractography by multi-tensor ROI placement Tensor model Multi-tensor model 42 y. o. male with grade II glioma Kei Yamada

axial b = 0 Navigation right hemisphere Kinoshita M, Yamada K et al. Neuro. axial b = 0 Navigation right hemisphere Kinoshita M, Yamada K et al. Neuro. Image 2005 Color vector 67 y. o. F Glioblastoma Kei Yamada

axial b = 0 Multi-tensor Color vector right hemisphere 協力:大阪大学医学部脳神経外科 Drs. Kinoshita, Hashimoto, Yoshimine axial b = 0 Multi-tensor Color vector right hemisphere 協力:大阪大学医学部脳神経外科 Drs. Kinoshita, Hashimoto, Yoshimine Kinoshita M, Yamada K et al. Neuro. Image 2005 67 y. o. F Glioblastoma Kei Yamada

How high should our b-value be? • Previous research says “higher the better. ” How high should our b-value be? • Previous research says “higher the better. ” • When using high-angular resolution diffusion imaging (HARDI) techniques • But there is SNR limit • Practical b-value for clinical use will be around 3000. Kei Yamada

 • Visual grading of the crossing fibers • Two locations; Ø optic chiasm • Visual grading of the crossing fibers • Two locations; Ø optic chiasm Ø superior cerebellar decussation Kei Yamada

 • Visual grading of the crossing fibers • Two locations; Ø optic chiasm • Visual grading of the crossing fibers • Two locations; Ø optic chiasm Ø superior cerebellar decussation Kei Yamada

Evolutions of DTI/tractography Tractography MR acquisition Q-space (multi-axes, b) [Callaghan 88] DTI (6 -axes) Evolutions of DTI/tractography Tractography MR acquisition Q-space (multi-axes, b) [Callaghan 88] DTI (6 -axes) [Basser 94] HARDI (> 15 -axes) Q-space (multi-axes, b) Q-ball (multi-axes) [Basser 02] [Tuch 02] [Mori 99] FACT Following e 1 TEND Following e 1 T Multi-tensor Probabilistic [Jones 03] Kei Yamada

Q-space imaging • Designed to analyze porous material – Callaghan 1988 • Introduced into Q-space imaging • Designed to analyze porous material – Callaghan 1988 • Introduced into medical field – Basser 2002 • Stepwise increment of the MPG • Measures displacement distribution of H 2 O Kei Yamada

90° 180° G Echo G d = 50 d D = 60 Assaf, et 90° 180° G Echo G d = 50 d D = 60 Assaf, et al. 2000 Mixing time G d=1 D = 100 d << D Kei Yamada

Q-space imaging on a pineapple ⊿=142. 4 ms,δ=17 ms ROI-2 (core) ROI-1 (fruit) Core Q-space imaging on a pineapple ⊿=142. 4 ms,δ=17 ms ROI-2 (core) ROI-1 (fruit) Core Fruite 0. 16 Si/S 0 0. 14 0. 12 0. 1 0. 08 0. 06 0. 04 0. 02 0 0 2000 4000 6000 8000 10000 b-value [s/mm 2] 12000 FWHM = 6. 07 RMSD = 2. 58 μm MD = 1. 42 E-7 FWHM = 2. 01 RMSD = 0. 85 μm MD = 4. 70 E-7 Kei Yamada

Brain temperature • Glucose metabolism → heat generation • Heat does not radiate through Brain temperature • Glucose metabolism → heat generation • Heat does not radiate through the thick skull • “Radiator” of the brain is cerebral blood flow (CBF) Kei Yamada

Brain temperature • Brain temperature changes with imbalance between r. CBF and metabolism • Brain temperature • Brain temperature changes with imbalance between r. CBF and metabolism • Can be measured by MRS • What about using DWI? – Nagy Z, et al. ISMRM 2009 • Direct relationship between the – absolute temperature (T) – diffusion coefficient (D) Kei Yamada

Temperature [℃] Temperature vs. age; normal volunteers 19 -24 25 -29 30 -34 35 Temperature [℃] Temperature vs. age; normal volunteers 19 -24 25 -29 30 -34 35 -39 40 -44 45 -56 • age Brain metabolism declines with age P < 0. 001 • Blood flow does not decline substantially • Leads to decrease in temperature • “Cool-brain phenomenon” of middle Sakai K, et al. under review age ; -) Kei Yamada

Summary • • ADC correlates with cell density and prognosis Clinical value of FA Summary • • ADC correlates with cell density and prognosis Clinical value of FA measurement → controversial Tractography may underestimate the size of the tract. Always compare the tractography results with known classical anatomy • Q-space imaging could allow us assessing white matter • Ventricular temperature measurement is now feasible • DTI for stroke imaging (tomorrow morning 09: 45) Kei Yamada

Acknowledgements Kizu O Matsushima S Oouchi H Akazawa K Goto M Kubota H Yuen Acknowledgements Kizu O Matsushima S Oouchi H Akazawa K Goto M Kubota H Yuen S Nishida K Ito H Tanaka O Sano Y Kakoi N Nagakane Y Tokuda T Ohwada K Kawai A Yamamoto T Hosomi A Mizuno T Tatsuzawa K Nishimura A Kinoshita S Shiga K Nakagawa M Sasajima H Morimoto M Kuriyama N Dept. of Neurology Mineura K Hosoi H Dept. of Neurosurgery Dept. of Pediatrics Dept. of Ophthalmology Miki Y Togashi K Sakai K Konishi J Kinoshita M Nakamura H Sasaki M Osaka City University Dept. of Radiology Kyoto University Department of HHS Sugimura K Hashimoto N Kato A Yoshimine T Nakaji S Nakajima Y Kudo K and all the ASIST-J members Kobe University Dept. of Radiology Osaka University Dept. of Neurosurgery St. Marianna's University Dept. of Radiology ASIST-Japan Obara M Hoogenraad F Okuaki T Philips Medical Systems van Cauteren M Kyoto Prefectural University of Medicine; Dept. of Radiology Nishimura T Mori S Johns Hopkins University Numaguchi Y St. Luke’s International Hospital Sorensen AG Massachusetts General Hospital NMR-center Leif Ostergaard Kei Yamada

Kei Yamada Kei Yamada