Model basedHead & Neck injury criteria

Deck C, Meyer F, Bourdet N, Willinger R.Strasbourg University

Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie (Icube)

Equipe Matériaux multi-échelles et Biomécanique (MMB)

Rémy WILLINGERremy.willinger@unistra.fr

Menschmodelle- DynamoreStuttgart June 2016

INTRODUCTION

• Critical issue with current head injury criteria

• State of the Art head FE modelling and validation

• Focus on head trauma database and accident reconstruction

• Tissue level head injury criteria and risk assessment tool

• Neck FE modelling and validation

• Whiplash injury criteria based on modelling

2

HUMAN SEGMENTS

Gz

Gx

Gy

PROTECTIVE SYSTEMS

HEAD TOLERANCE LIMITS AND HEAD INJURY CRITERIA

Part I : tests on cadavers, skull failure considered as head injury.

Part II : intracranial pressure recorded on anatomical subjects and animals, head injury : commotion.

Part III : tests on human volunteers, no head impact, head kinematics recorded during sled tests.

Head tolerance curve proposed by Wayne State University given linear head accelerations versus time : WSUTC (1966). Head injuries occur in the part upper the curve.

HEAD INJURY CRITERION (1972) : HIC DEFINITION

0

10

20

30

40

50

60

70

80

90

100

0 500 1000 1500 2000 2500

HIC

AIS

3 inju

ry r

isk

(%)

Head mass = 4.58 kg; HIC = 1000

5.2

12

12),(

2

1

21

)(1

)(max

t

ttt

dttatt

ttHIC

CONTEXT OF HEAD PROTECTION STANDARDS

• Inside a car (1970)

- Dummy head; HIC 1000

• Outside – pedestrian (2005)

- Headform; V=11 m/s ;e = 7 cm ; HIC 1000 à 1700

• Motorcyclist (2002)

- Headform; V = 7.5 m/s ;

e = 5 cm ; HIC 2400 ; G= 275G

• Cyclist

- Headform; V = 5.42 m/s ;

e = 2.5 cm ; G= 250G

… for a same human head !

LIMITATIONS OF EXISTING STANDARDS

• Poor correlation with real world observation

• HIC was defined for a frontal impact…and is not direction

dependent

• Not injury mechanism related

• No consideration of rotational acceleration

• No criteria for children (6 YOC, 3 YOC…)

9

It is well known that brain is sensitive to rotational accelerationsince Holbourn (1943)

This phenomenon has essentially been addressed qualitatively with animal or physical models.

Ommaya et al. (1967, 1968), Unterharnscheidt (1971), Ono et al. (1980), Gennarelli et al. (1982), Newman et al. (1999,2000)…..

By using Finite Element Head Models it was expressed quantitatively how dramatic the influence of the rotational acceleration is on intra-

cerebral loading.Deck et al. (2007), Kleiven et al. (2007), Zhang et al. (2001)...

CRITICAL ASPECT OF ROTATIONAL LOADING

9

A number of experimental in vivo investigations emphasized that

axonal strain was the most realistic mechanism of DAI (Bain and Meaney, 2000, Meythaler et al., 2001, Morrison et al., 2003)

GLOBAL PARAMETERS (ROTATION)

Authors Global parameters

Gennarelli, Thibault, Ommaya

(1972)25 Monkeys alive

1800 rad/s² à 7500 rad/s²

60 rad/s à 70 rad/s

Pincemaille et al.

(1989)Boxers training

13600 rad/s² à 16000 rad/s²

28 rad/s à 48 rad/s

Gennarelli et al.

(1982)More than 100 primates alive

15000 rad/s²

150 rad/s

Margulies et al.

(1989)

Based on Gennarelli et al.

(1982)

16000 rad/s²

46.5 rad/s

No agreement

Global parameters-Combined

HIP:

G(t) =a(t)

ac

æ

èç

ö

ø÷

m

+a(t)

ac

æ

èç

ö

ø÷

né

ë

êê

ù

û

úú

1

s

GAMBIT:

n = m = s = 2.5, ac=250g, c = 25.000 rad/s²

PRHIC:

HIP = max ax dt +ò may ay dt +ò maz az dt +ò

I xxax ax dt +ò I yyay ay dt +ò I zzaz az dtò

Newman et al 1986

Newman et al 2000

Kimpara et al. (2011)

Global parameters - Rotation

RIC:

BrIC:

Takhounts et al. 2013

Takhounts et al. 2011

Kimpara et al. (2011)

BRAIN INJURY CRITERIA

• There is no relevant combined, time and direction

dependent brain injury criteria in terms of global head

acceleration

• A number of tentatives exist

• There is a need to set properly :

• A tissue level brain injury criteria

• A measure of the quality of an injury criteria

STATE OF THE ART

FE HEAD MODELS AND

VALIDATION

14

HEAD FE MODELS AROUND THE WORLD

SUFEHMKang et al. 1997

SIMonTakhounst et al. 2003Takhounst et al. 2008

EindhovenClaessens et al. 1997

Brands 2002

StockholmKleiven et al. 2002Kleiven et al. 2007

TurinBelingardi et al. 2005

DublinHorgan et al. 2003Gilchrist et al. 2004

THUMSIwamoto et al. 2001

WSUBIMZhou et al. 1995

Zhang et al. 2001King et al. 2003

Nahum & Trosseille

(1977) (1992)Impact area : front

Impactor : Cylinder with padding

Impact velocity : 6.3 m/s

Duration : 6.2 ms

Yoganandan

(1994)Impact area : vertex

Impactor : Rigid sphere

Impact velocity : 7.3 m/s

Duration : 2 ms

Sarron

(1999)

Back face effect Under Balistic conditions

Hardy

(2001)Impact area : occipital

Impactor : Cylinder

Impact velocity : 2 m/s

Duration : 20 ms

Skull

validation

Intra-cranial

behaviour

validation

VALIDATION DATA

16

Input :

•A 5.6 kg cylindrical impactor (with padding).•An initial velocity about 6.3 m/s•Boundary conditions : Head free

Interaction force between the head and the impactor

BENCHMARK PROCEDURE : NAHUM INPUT

[Nahum et al., 1977]

0 2 4 6 8 10 12 14 16

0

1000

2000

3000

4000

5000

6000

7000

Experimental

Imp

act F

orc

e (

N)

Time (ms)

6 m/s

IMPACTOR

MASS 5.6Kg

E=13.6 Mpa

=0.16

NAHUM IMPACT NUMERICAL RESULTS

0 2 4 6 8 10 12 14

0

500

1000

1500

2000

2500

He

ad

Acce

lera

tio

n [

m/s

²]

Time (ms)

EXPERIMENTAL DATA

ULP

TURIN

THUMS

STOCKHOLM

TUE

0 2 4 6 8 10 12 14-1000

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

EXPERIMENTAL DATA

ULP

TURIN

THUMS

STOCKHOLM

TUE

Inte

r-a

ctio

n F

orc

e [

N]

Time (ms)

•Impact force, head acceleration

Some oscillations can appear in head acceleration results

NAHUM IMPACT NUMERICAL RESULTS

STATISTICAL ANALYSIS WITH ADVISER

STATISTICAL ANALYSIS : RESULTS

STATE OF THE ART NUMERICAL HEAD MODEL

Brain acceleration and pressure- THUMS, SUFEHM and KTH models provided a comparable level of accuracyfor brain acceleration

- Pressure prediction was at similar level of accuracy for all models

Brain displacement- THUMS, SUFEHM and KTH presented best accuracy

- NHTSA and TUE were less accurate

Skull deflection- Only THUMS and SUFEHM models predicted an accurate skull deflection aswell as skull rupture

22

Skull Model Improvement

• Refined meshing

• Skull thickness variation

• Inclusion of reinforced beams

• Improvement of non-linear

material characteristics

50th percentile

adult skull

Digitalisation

SUFEHM 98

Accident reconstructions

Tolerance limits

STRASBOURG UNIVERSITY FE HEAD MODEL

[Kang, 1997]

[Deck, 2004]

SUFEHM PRESENTATION

Membranes(Elastic E=31.5MPa, =0.23)

CSF(Elastic E=12kPa, =0.49)

Face(rigid)

Brain(Viscoelastic G0=49kPa, G=16.7kPa, β=145s-1)

Brainstem(Viscoelastic G0=49kPa, G=16.7kPa,

β=145s-1)

Skull(Shell elements, composite

law with failure criterion)

Scalp(Elastic E=16.7MPa, =0.42)

Identification of Skull mechanical parameters

25

Skull was modelled by a three layered composite shell and damage

mechanism based on Tsai and Wu criterion (Tsai and Wu ,1971).

Brain mechanical properties

26

Rheometry

• 0.004 to 8000Hz

MRE

• Mean values for the whole brain• 20 to 200Hz

Indentation

• 1 to 1000Hz

High discrepancy of values for shear modulus Confirms the stiffest in vitro results (shear modulus ~10KPa at 100Hz)

27

RECENT DEVLOPMENTS

• Marjoux, D., Bourdet, N., and Willinger, R. 2009 . Computation of axonal elongations: towards a new brain injury criterion. International Journal of Vehicle Safety, Vol.4 No 4, 271

• Chatelin S., Deck C., Renard F., Kremer S., Heinrich C., Armspach JP, Willinger R : 2011Computation of axonal elongation in head trauma finite element simulation. J of Mech. Behavior of Biomed Material, V4, 1905-1919.

• Cloots, R.J.H., van Dommelen, J.A., Nyberg, T., Kleiven, S., Geers, M.G., 2011. Micromechanics of diffuse axonal injury: influence of axonal orientation and anisotropy. Biomechanics and Modeling in Mechanobiology. 10, 3, 413-422.

• Wright R, K Ramesh : 2011, An axonal strain injury criterion for traumatic brain injury, Biomechanics and Modeling in Mechanobiology, , 1-16.

• Cloots RJH, van Dommelen JAW, Kleiven S, Geers M, 2013. Multi-scale mechanics of traumatic brain injury: predicting axonal strains from head loads. Biomechanics and Modeling in Mechanobiology, 12(1):137-150.

• Giordano, C, Kleiven, S, 2014. Evaluation of Axonal Strain as a Predictor for Mild TraumaticBrain Injuries Using Finite Element Modeling. Stapp Car Crash Journal, Vol. 58

• Sahoo D., Deck C., Willinger R.:2014 Development and validation of an advanced anisotropic visco-hyperelastic human brain FE model. Journal of the Mechanical Behavior of Biomedical Materials, 2014, vol.33, 24-42

• Sahoo D., Deck C., Willinger R.:2015 Axonal strain as brain injury predictor based on real world head trauma simulation. IRCOBI 2015 and AAP 2016

DTI OF THE BRAIN

28FRACTIONAL ANISOTROPY

DIFFUSION TENSOR IMAGING

ISOTROPIC DIFFUSION

ANISOTROPIC DIFFUSION

12

HEALTHY PERSONS

COUPLING OF DTI DATA

1( ( ))²3 3

2 ( )²

ii

ii

tr DFA

Fibers orientationFractional Anisotropy

| max( )j ijl FA e

[Papadakis et al., 1999; LeBihan et al. , 2001]

Diffusion Parameters

Anisotropy vector

Fractional Anisotropy

1

1

i e

i e

ND L

i

i

NelD L

i

FAe

FA

e

1

1

i e

i e

ND L

i

i

Nel

D L

i

l e

l

e

Rigid transformation between mask of in vivo diffusion data (in red) and brain FEM (in blue)

Voxel selection

Assign Weighting function

NEW ENHANCED BRAIN MODEL

30

l

Longitudinal superior

fascius

Cingulum

Cerebellum

Anisotropic visco-hyperelastic brain model (Chatelin et al. 2012)Matrix stress Fibers stress

, ,e vS C t S C S C t

0

, 2 ( )( )

tv W

S C t G t s dsC s

2 0110

12e C

C

4 1

3

0 0 1

1 1C

C e

DTI

Coupling between anisotropy information and brain FEM meshing

Heterogeneous and anisotropicbrain model

Sagittal view

Axial view Frontal view

MODEL BASED HEAD

INJURY CRITERIA

REAL WORLD HEAD TRAUMA

SIMULATION

31

PUBLISHED

TISSUE LEVEL

INJURY CRITERIA

32

Local Parameters (FE)

Local tissue level brain injury criteria are based on SIMon, KTH, WSU,

THUMS and SUFEHM finite element head models:

• MPS Max principal strain

• SCC Strain in Corpus Callosum

• VM strain Max VM strain

• SSR Strain*Strain rate

• Pmax Max pressure

• VM stress Max VM stress

• CSDM Cumulative Strain Damage Measure

• MAS Maximum axonal strain

INJURY CRITERIA FROM THE LITERATURE

ACCIDENTS RECONSTRUCTIONS

35

• METHODOLOGY

Database (125 cases)

36

HEAD TRAUMA DATABASE

Experimental Skull fracture tests

38

86 drop tests

were conducted

from 17 PMHS

specimens

17 PMHS’ heads are tested.

Accelerometer packages are attached to the skull using screws.

Drop techniques for impact with successively increasing input energies until fracture.

Identification of skull constitutive law

39

Parameters Cortical bone Diploe Bone

Mass density (Kg/m3) 1900 1500

Young’s Modulus (Mpa) 15000 4665

Poisson’s ratio 0.21 0.05

Longitudinal and transverse compressive strength (Mpa) 132 24.8

Longitudinal and transverse Tensile strength (Mpa) 90 34.8

0 2 4 6 8 100

2000

4000

6000

8000

10000 Force UC

Force mean

Force LC

Simulation

V=6.47m/sF

orc

e (

N)

Time (ms)

15 impact

conditions

[Gent et al., 1958][Gray et al., 1991][Pampush et al., 2011]

Parameters 40D Flat 90D Flat 90D Cylindrical

Mass density (Kg/m3) 4230 4930 4930Young’s Modulus (MPa) 9 12 12

Poisson’s ratio 0.43 0.43 0.43

Xc

Xt

Sc

Yc

Yt

DETAILED ACCIDENT

RECONSTRUCTION

MODELING OF THE ACCIDENTS

Unistra modeling

0 20 40 60 80 100 120 140 1600

5

10

15

20

25

30

Fo

rce

[kN

]

Deflection [mm]

(a)

0 20 40 60 80 1000.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

(b)

Fo

rce

[kN

]

Deflection [mm]

0 20 40 60 80 1000

2

4

6

8

10

(d)

Fo

rce

[kN

]

Deflection [mm]

The scalable TNO Pedestrian

model implemented in

Madymo® package

The contact force functions used on

each part of the car are extracted from

the study of Martinez et al.,2007

The Unistra bicycle model

EXAMPLE : DESCRIPTION OF ACCIDENT CASE

Impact Conditions

Car velocity 45 km/h

Cycle Velocity 5.5 km/h

Cycle/Car angle 6°

Vehicle deceleration 6,5 m/s²

Victim

Man, 91 years old,

Failure parieto-occipito-temporal

Coma with a Glasgow score of 5

Unistra modeling

EXAMPLE : KINEMATICS RECONSTRUCTION

Vresultant = 10.9 m/s

Vnormal = 10.0 m/s

Vtangential = 4.4 m/s

Two impacts

• on windshield with the left shoulder,

• on pillar with head area occipito-parieto-temporal.

Projection distance of 16.3 m

WAD of 2.10 m

Unistra modeling

Ped. moving direction

Impact point

Rest points

From IVAC database

- Victim information: 49-year-old female, 158cm and 58kg

- Vehicle information: BMW 318

- Impact speed: about 62.9 km/h

Injury details:- Cerebral contusion (AIS3), Hematoma (AIS2), Fatal

head injuries (AIS6)

- Right tibia (AIS3) and fibula (AIS3) fracture

Exemple pedestrian case (2)

ACCIDENT DATA COLLECTION AND RECONSTRUCTION

Reconstruction results

Example 1 Example 2

Example 1 Example 2

Accident Simulation Accident Simulation

Throw distance (m) 12.4 11.3 18 17.5

WAD (mm) 2000 2030 1980 1940

Velocity (km/h) 60 54 60 62.9

ACCIDENT DATA COLLECTION AND RECONSTRUCTION

Windscreen FEM

46

Perpendicular to the windshield at 40 km/h [Lex van Rooij et al, 2001]

Material Parameters

Glass E=74GPa; ρ=2500kg/m3; μ=0.227; EFG=0.001

PVB E=2.6GPa; ρ=1100kg/m3; μ=0.435

Windscreen Mechanical properties

Adult headform

Masse : 4.8 kg

Constraint

A

B

A

B

Case 2

ACCIDENT CASE

MODEL BASED HEAD

INJURY CRITERIA

48

HEAD TRAUMA SIMULATIONS

EXTRACTION OF CRITERIA

Binary logistic regression (SPSS v14.0)

we compared

the Nagelkerke R-sq statistics

)(1

1bxae

0

10

20

30

40

50

60

70

80

90

100

0 500 1000 1500Pro

bab

ility

of

sku

ll fr

actu

re (

%)

Skull internal energy (mJ)

Skull fracture criteria

51

0200400600800

1000120014001600

SG1

0050

6

SG1

0045

6

P38

SG1

0047

5

VIR

GIN

IA33

SG1

0045

7

P36

SG1

0046

3

VIR

GIN

IA27

GID

AS1

3

GID

AS1

4

GID

AS7

IVA

C7

GID

AS2

IVA

C3

P37

IVA

C4

case

A1

VIR

GIN

IA30

P34

GID

AS8

GID

AS3

GID

AS1

1

case

A3

GID

AS1

6

IVA

C1

0

GID

AS2

3

GID

AS9

case

B1

GID

AS2

4

case

B2

SG1

1019

7

IVA

C1

2

P31

GID

AS2

8

VIR

GIN

IA 2

3

IVA

C1

5

case

B3

IVA

C1

4

case

B4

GID

AS2

6

case

B5Sk

ull

inte

rnal

en

erg

y (m

J)

448

RobustnessR2=0.633

Brain Injury criteria DAI (AIS 2+)

52

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0,45

GID

AS4

GID

AS1

1

GID

AS1

SG1

0045

6

GID

AS2

6

S00

9

GID

AS1

6

GID

AS2

1

IVA

C3

GID

AS1

5

GID

AS1

0

IVA

C7

GID

AS1

8

GID

AS1

3

VIR

GIN

IA30

S015

GID

AS1

4

S01

7

VIR

GIN

IA 2

3

S01

1

S00

1

FIA

4

M0

10

S014

M0

08

P32

S013

M0

01

P34

P37

P31

M0

11

S01

6

S02

2

S00

2

IVA

C8

Axo

nal

elo

nga

tio

n

0.1465

RobustnessR2=0.876

0

10

20

30

40

50

60

70

80

90

100

0 0,05 0,1 0,15 0,2 0,25 0,3

Pro

bab

ility

of

DA

I (%

)

Axonal strain

Evaluation of existing Head Injury Criteria

53

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

HIC FORCE SKULLINTERNALENERGY

HIC FirstPrincipal

Strain

Von MisesStrain

FirstPrincipal

Stress

Von MisesStress

AxonalStrain

Nag

elke

rke

R²

valu

e

Skull Fracture

DAI

AXON STRAIN IN THE LITTERATURE

0

5

10

15

20

25

30

Haftek et al.,(1970)(rabbit)

Galbraith etal., (1993)

(squid)

Bain andMeaney(2000)

(guinea pig)

Morrison etal., (2003)

(cell culture)

Singh et al.,(2006) (cell

culture)

Nakadate etal., (2014)

(cell culture)

Present Study

Thre

sho

ld S

trai

n %

15%

Proposed tolerance limit is in accordance with various studied reported in literature.

HEAD INJURY

PREDICTION TOOL FOR

END USERS

HEAD INJURY PREDICTION TOOL

• COUPLED EXPERIMENTAL VS NUMERICAL TEST METHODS

• FULL FE APPROACH

FROM RESEARCH TO END USERS

57

• PRE-POST-PROCESSING USER INTERFACES :

3

4

SUFEHM IRA TOOL

INJURY RISK ASSESSMENT

COMPUTATION OF SUFEHM CRITERIA

VIA WEB SIMULATION

TEST HOUSE

6D acceleration curves

STR

ASB

OU

RG

U

NIV

ERSI

TY

60

STEP 1. IMPLEMENTATION OF LOADING CURVES

61

STEP 4. INJURY RISK ASSESSMENT

HEAD PROTECTIVE SYSTEMS

EVALUATION & OPTIMISATION

HELMET CONSUMER TESTS:

TOWARDS NEW HELMET

STANDARDS

64

TANGENTIAL TEST METHOD (CEN TC158)

64

Three tangential impacts at 6.5 m/s

65

TANGENTIAL TEST METHOD (CEN TC158)

65

66

HELMET CONSUMER TESTS IN FRANCE

66

Journal title:

60 Millions de consomateurs (F)August 2015

12 moto helmets evaluatedbased on SUFEHM criteria

67

HELMET CONSUMER TESTS IN GERMANY

67

Journal title:

Stiftung WarentestAugust 2015

18 bicy. helmets evaluatedbased on SUFEHM criteria

PEDESTRIAN AND PASSAGER

PROTECTION

VIRTUAL TESTING IN AUTOMOTIVE ENVIRONMENT

Safe-EV projectPedestrian Passive Safety

RESULTS OF PEDESTRIAN SIMULATIONS

OVERVIEW OF ASSESSMENTS

• Assessment of head injury risk

(using SUFEHM –IRA tool under VPS)

• Further possible injury risk indicators (based on max. pl. strain analysis)

– ribs

pelvis

• tibia/fibula and femur

2 enclumes

ECE R022 standardDevelopment of helmet FEM and coupling with SUFEHM

Injury riks evaluationwith SUFEHM

BICYCLE HELMET

v=5.42 m/s

Validation Head injury risks3 4

Helmet optimisation againstbiomechanical criteria

5

0 10 20 30 40 50 60 70 80

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

Pro

bab

ilit

y (

DA

I)

Brain Von Mises stress [kPa]

LESS-THAN-LETHAL WEAPONS

74

CAD

Meshing Mechanical properties definition

Validation againstexperimental data :

Experiments on rigid wall

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

0

1

2

3

4

5

6

str

ess [M

Pa

]

strain [-]

Foamdescription

Evaluation of head injury risks1

2 3

45

0 10 20 30 40 50 60 70 80

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

Pro

bab

ilit

y (

DA

I)

Brain Von Mises stress [kPa]

Linear and depressive skull failuresprediction

BACK EFFECT : MILITARY HELMET

VALIDATION

A LEGAL MEDICINE CASE

Head injury risks calculated with SUFEHM

77

CONCLUSION-HEAD

• Advanced brain FE models, Computation of axon strain

• Consolidated head trauma database with 125 cases.

• Very high Nagelkerke R² value (R²=0.876) for brain injury

• Best candidate parameter for brain injury is axon strain

• The model based head injury criteria are:

• Axon strain for brain AIS2+ (ea= 15%)

• Skull strain energy for fracture (0.5 J)

• Head injury prediction tool for end users

77

NECK MODELLING AND

WHIPLASH INJURY CRITERIA

Millimetric scanner sections

Surface reconstruction

Meshing criteria

1 2

4

Stereolithographic format

METHODS

Height : 1m72 Weight : 72 Kg Age : 33 Years (50th)

Criterion Values

Length min 2.25 mm

Length max 3 mm

Aspect ratio [1-2]

Warpage [0-5]

Angle quad(°) [70-110]

Angle Tria [50-80]

Jacobien [0.7-1]

% of tria 6

FE NECK MODEL : GEOMETRY

FEM OF THE HEAD-NECK SYSTEM

Intervertebral Discs

Atlas

Axis

C3

C4

C5

C6

C7

T1

Capsularligament

Ligament commun anterior

Interspinalligament

Flavumligament

Ligament commun posterior

Brain

CSF

Falx

Tentorium Skull

Scalp

Face

Number of elements25 661

NECK FINITE ELEMENT MODEL (UNISTRA)

81

NECK FINITE ELEMENT MODEL (UNISTRA)

82

HEAD-NECK UNISTRA COUPLING

83

Coupling to THUMS Model (LS-Dyna)…..TUC !

Moment at the OccipitalCondyles calculated with theTHUMS/Unistra-Head-Neck FEMsuperposed the corridor defineby Patrick and Chou 1971

Force at the Occipital Condylescalculated with theTHUMS/Unistra-Head-NeckFEM

Lateral Impact (N.B.D.L) Ewing et al. 1977

FOLKSAM DATABASE : 87 REAR IMPACTS ACCIDENTS

No neck injury Initial symptoms Symptoms > 1 month0

10

20

30

40

50

60

Nu

mbe

r of occup

an

t

Driver

Front Seat Passenger

=

122 accident cases :

• 77 no neck injury• 30 initial symtoms• 15 symptoms over than 1 month

No neck injury

Initial symptoms

Symptoms > 1 month

Delta V of pulses versus injury severity.

Average age of 46 year old

METHODOLOGY

(FOLKSAM)Linear and Angular Accelerations (T1 & headrest)

Ω(t)

UdS MEF

Injury risk curves/R² score

0 50 100 150 200

0

10

20

30

40

50

60

Acc

ele

ratio

n [m

/s²]

Time [ms]

0 50 100 150 200

-40

-20

0

20

40

60

80

100

Acce

lera

tio

n [m

/s²]

Time [ms]

Ax

Ay

Az

0 50 100 150 200-150

-100

-50

0

50

100

150

Ang

ula

r acce

lera

tio

n [R

ad/s

²]

Time [ms]

Arx

Ary

Arz

V(t)Ω(t)

V(t)NIC, Nkm, MY, Fx,Fz

0

50

100

150

200

250

300

Head a

cc M

ax

2971

630

004

2968

730

005

2960

129

971

3014

930

236

2992

430

237

3015

729

795

2984

929

966

3016

929

911

2967

730

115

2997

530

196

2949

129

880

2977

330

068

3025

129

908

3027

930

318

3031

630

063

2977

829

976

2978

129

951

3032

129

706

2965

230

232

3017

030

024

2996

529

732

3032

630

360

2961

430

029

3024

430

078

2996

830

075

3026

330

278

3015

329

807

3025

330

334

3013

0

3001

629

945

3000

730

001

3016

030

097

2978

029

974

2966

430

125

2973

729

733

3005

230

111

2969

329

967

3003

230

259

3004

9

2997

229

533

2973

930

082

2987

6

2972

729

577

2952

129

602

3001

3

Symptom 0 Symptom 1 Symptom 2 Symptom 3

0

10

20

30

40

50

60

HIC

3000

429

716

2968

730

005

2997

129

601

3014

930

236

2992

430

237

2996

630

157

2979

529

849

3016

929

911

2967

730

196

2997

529

773

3011

529

491

2988

030

068

3025

129

908

3027

930

318

2977

829

976

3031

630

063

2995

129

781

2970

630

321

2965

230

232

3017

030

024

2973

229

965

3036

030

326

2961

430

029

3007

829

968

3024

430

263

3007

530

278

3015

329

807

3025

330

334

3013

0

3001

629

945

3000

730

001

3016

029

780

2997

430

097

2966

430

125

2973

730

052

2973

330

111

2969

329

967

3003

230

259

3004

9

2997

229

739

2953

330

082

2987

6

2957

729

727

2952

129

602

3001

3

Symptom 0 Symptom 1 Symptom 2 Symptom 3

0

5

10

15

20

25

NIC

2949

130

251

2960

129

614

3026

329

652

2967

729

924

2997

530

004

3000

530

024

3006

330

115

3014

930

157

3027

830

236

3031

830

237

3032

130

279

3031

630

360

2968

729

706

2980

729

716

2973

229

880

2996

629

773

2977

829

781

2979

530

170

2984

929

911

2995

130

029

2996

529

971

3032

629

976

3023

230

244

3007

530

334

3025

330

078

3006

830

130

2990

830

153

3016

930

196

2996

8

2994

530

097

2966

429

693

3000

129

733

2973

729

780

2996

729

974

3001

630

049

3003

230

052

3011

130

125

3016

030

259

3000

7

2953

329

972

2973

929

876

3008

2

2952

129

577

2972

729

602

3001

3

Symptom 0 Symptom 1 Symptom 2 Symptom 3

HIC, HEAD Max Acceleration, NIC

6

RESULTS

R² Stade 1 R² Stade 2 R² Stade 3

NIC 0.017 0.073 0.109

Nkm 0.086 0.324 0.266

Fx upper 0.108 0.363 0.361

Fz upper 0.071 0.076 0.047

My upper 0.127 0.433 0.495

Abs (C1-C7) 0.223 0.545 0.842

6

R² WAD1 R² WAD2 R² WAD3

Abs (C1-C7) 0.223 0.545 0.842

Meyer et al.2012 , ‘Development and Validation of a Coupled Head-neck FEM – Application to Whiplash Injury Criteria Investigation’, International Journal of Crashworthiness, 2012, 1–24

NECK INJURY CRITERIA BASED ON FE NECK MODEL AND REAL

WHIPLASH ACCIDENT RECONSTRUCTION

CONCLUSION-NECK

• Full validated neck model (time & frequency)

• Model based neck injury criteria (F,L,R,V)

• SUFEHM_onNeck coupled to THUMS-v3

• Transferred in automotive industry

• Injury Risk Assessment tool

89

Model basedHead & Neck injury criteria

Deck C, Meyer F, Bourdet N, Willinger R.Strasbourg University

Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie (Icube)

Equipe Matériaux multi-échelles et Biomécanique (MMB)

Rémy WILLINGERremy.willinger@unistra.fr

Menschmodelle- DynamoreStuttgart June 2016

91

HEAD INJURY RISK ON « BONNET POINT »

Real pedestrian test

0 10 20 30 40 50

0

250

500

750

1000

1250

1500

Resultant

Lin

ear

Head A

ccele

ration

[m/s

²]

Time [ms]

0 10 20 30 40 50

-8000

-6000

-4000

-2000

0

2000

4000

6000

Y R

ota

tional

Head A

ccele

ration

[rad/s

²]

Time [ms]

6 head accelerations

0 10 20 30 40 50 60 700

1000

2000

3000

4000

5000

6000

7000

8000

9000

Norm

ale

Forc

e [N

]

Displacement [mm]

Force-displacement curves computed

Pedestrian test simulation

0 10 20 30 40 500

250

500

750

1000

1250

1500Finit Element ModelLumped Model

Resultant Lin

ear

Head A

cce

lera

tion [m

/s²]

Time [ms]

« Bonnet point » validation

« Bonnet point » validated model

Step 1: Validation procedure

Predictif head FE model

Injury risk based on biomechanical criteria

Step 2: FE test procedure