Today’s Agenda Today Today ’ ’ s Agenda s Agenda • Lunch • Sam Hund’s Computational Presentation • The Web Site • Subversion Repository • • Lunch Lunch • • Sam Sam Hund Hund ’ ’ s s Computational Presentation Computational Presentation • • The Web Site The Web Site • • Subversion Repository Subversion Repository Sangria Project
76
Embed
Today’s AgendaToday’s Agendasangria/publications/SHund_Sangria...James F. Antaki, PhD, CMU Biomedical Engineering June 4th, 2010 Samuel J. Hund James F. Antaki, PhD, CMU Biomedical
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Today’s AgendaTodayToday’’s Agendas Agenda• Lunch• Sam Hund’s Computational Presentation• The Web Site• Subversion Repository
•• LunchLunch•• Sam Sam HundHund’’ss Computational PresentationComputational Presentation•• The Web SiteThe Web Site•• Subversion RepositorySubversion Repository
Sangri
a Proj
ect
A Multi-Physics Approach for Predicting Platelet-Mediated
Thrombosisfor the Evaluation and Design of Medical Devices
A MultiA Multi--Physics Approach for Physics Approach for Predicting PlateletPredicting Platelet--Mediated Mediated
ThrombosisThrombosisfor the Evaluation and Design of Medical Devicesfor the Evaluation and Design of Medical Devices
Samuel J. HundJames F. Antaki, PhD,
CMU Biomedical EngineeringJune 4th, 2010
Samuel J. HundSamuel J. HundJames F. Antaki, PhD, James F. Antaki, PhD,
Intricacies of ThrombosisIntricacies of ThrombosisIntricacies of Thrombosis
Sangri
a Proj
ect
Brief Review of My WorkBrief Review of My WorkBrief Review of My Work• Computational Modeling of Blood
Rheological ModelingModeling of HemolysisModeling of Thrombosis
• Computational Optimization
•• Computational Modeling of BloodComputational Modeling of BloodRheological ModelingRheological ModelingModeling of HemolysisModeling of HemolysisModeling of ThrombosisModeling of Thrombosis
•• Is Is ββ+1 significantly +1 significantly different from 1?different from 1?
No pNo p--value = 0.42value = 0.42•• Is Is αα significantly different significantly different
from 2?from 2?No pNo p--value = 0.122value = 0.122
•• The overThe over--all powerall power--law law model is actually model is actually significantly better than significantly better than the Richardson model (p the Richardson model (p value 0.0065), but there value 0.0065), but there is no confidence in the is no confidence in the parameter parameter ββSan
gria P
rojec
t
Prediction of Hemolysis in a Blood Shearing Device
Prediction of Hemolysis in a Blood Prediction of Hemolysis in a Blood Shearing DeviceShearing Device
Inde
x of
Hem
olys
is (%
)
Shear Stress (dyn/cm2)
Exposure Time:
Sangri
a Proj
ect
ADP Release from RBCsADP Release from RBCsADP Release from RBCs
PROPORTIONAL MODELSlope: 0.071 +/- 0.03 μM/mg%
R2:0.89
0 5 10 15 20 25
2.5
2.0
1.5
1.0
0.5
0.0
AD
P (μ
M)
pfHb (mg%)
Data from Alkhamis et al. 1988Best Fit Proportional Line dt
pfHbddtADPd ][][ α=
•Hemolysis can directly lead to platelet activation through the release of ADP•ADP is an often neglected factor in shear induce platelet activation.
Sangri
a Proj
ect
Hemolysis in a NozzleHemolysis in a NozzleHemolysis in a Nozzle
Sangri
a Proj
ect
Early PredictionsEarly PredictionsEarly Predictions
Platelet Deposition to Surface Platelet Depositiononto Active platelets
Transport
47
Sangri
a Proj
ect
Weakness of Existing ModelWeakness of Existing ModelWeakness of Existing Model
1. Activation Kinetics2. No Shear Dependent Activation3. Active Cellular Transport4. Valid for Collagen Only5. Lacks Coagulation Cascade6. Lacks Protein Deposition7. Limited Anticoagulation/Plt8. Effect of Growing Thrombus on Flow
1.1. Activation KineticsActivation Kinetics2.2. No Shear Dependent ActivationNo Shear Dependent Activation3.3. Active Cellular TransportActive Cellular Transport4.4. Valid for Collagen OnlyValid for Collagen Only5.5. Lacks Coagulation CascadeLacks Coagulation Cascade6.6. Lacks Protein DepositionLacks Protein Deposition7.7. Limited Anticoagulation/PltLimited Anticoagulation/Plt8.8. EffectEffect of of Growing ThrombusGrowing Thrombus onon FlowFlow
Maximum surface coverageRate of Resting and Active Platelet Deposition onto the SurfaceRate of Active Platelets onto Other Plateletsθ, the “instantaneous effect” or rate at which a material by itself activates plateletsMass transfer rate
•• Unknowns:Unknowns:Maximum surface coverageMaximum surface coverageRate of Resting and Active Platelet Deposition Rate of Resting and Active Platelet Deposition onto the Surfaceonto the SurfaceRate of Active Platelets onto Other PlateletsRate of Active Platelets onto Other Plateletsθθ, the , the ““instantaneous effectinstantaneous effect”” or rate at which a or rate at which a material by itself activates plateletsmaterial by itself activates plateletsMass transfer rateMass transfer rateSan
ExperimentsExperimentsExperiments• Flow Between Parallel Plates (kaa, kas, krs)
(required) Direct measurement of platelets/areaPossibly platelet morphology
• Maximum surface coverageRocking or submersion testing
• Rate of platelet activation due to the surface
•• Flow Between Parallel Plates (kFlow Between Parallel Plates (kaaaa, k, kasas, , kkrsrs))(required) Direct measurement of (required) Direct measurement of platelets/areaplatelets/areaPossibly platelet morphologyPossibly platelet morphology
•• Maximum surface coverageMaximum surface coverageRocking or submersion testingRocking or submersion testing
•• Rate of platelet activation due to the Rate of platelet activation due to the surfacesurface San
gria P
rojec
t
0 1 2 3 4 5 6 7 8 9 100
200
400
600
800
1000
1200
Example Fit to Experimental Data of Wagner and Hubbell
Example Fit to Experimental Data Example Fit to Experimental Data of Wagner and Hubbellof Wagner and Hubbell
Sangri
a Proj
ect
Cellular Transport ModelsCellular Transport ModelsCellular Transport Models
Sangri
a Proj
ect
Transport ModelsTransport ModelsTransport Models• Challenge: Large Number of Parameters• Cannot be determined from Fully-
Developed, Steady-State flows
•• Challenge: Large Number of ParametersChallenge: Large Number of Parameters•• Cannot be determined from FullyCannot be determined from Fully--
Vary flow rate to help isolate lift forcesVary plasma viscosity to isolate drag forcesGet RBC concentration in a 3-Dimensional fieldGet RBC velocity and plasma velocity
•• Channel FlowChannel FlowVary flow rate to help isolate lift forcesVary flow rate to help isolate lift forcesVary plasma viscosity to isolate drag forcesVary plasma viscosity to isolate drag forcesGet RBC concentration in a 3Get RBC concentration in a 3--Dimensional Dimensional fieldfieldGet RBC velocity and plasma velocityGet RBC velocity and plasma velocity