Experimental Techniques

https://doi.org/10.1007/s40799-018-0255-0

Force Model for Ultrasonic Needle Insertion

L. Tan 1,2 & J.A. Jones 2 & A.C. Barnett 2 & H. Zhang 1 & J.Z. Moore 2 & Q. Zhang 1

Received: 9 November 2017 / Accepted: 9 May 2018

# The Society for Experimental Mechanics, Inc 2018

Abstract
A better understanding of needle-tissue interaction forces when needle is inserted into tissue during injection and minimally
invasive surgery can lead to more accurate needle placement, better needle path planning, and better surgical simulation. In this
study, a vibratory needle insertion force model is experimentally developed to explore the effect of ultrasonic amplitude on the
insertion force. The maximum insertion force is comprised of three forces: tearing force, spreading force, and friction force.
Ultrasonic vibration needle insertion experiments are performed on a tissue simulant to measure how vibratory parameters
influence friction force, dynamic fracture toughness, and crack length. Results show that with the vibration amplitude changing
from 0 to 517.2 μm, the maximum puncture force reduces up to 60.8%, the friction force reduces up to 80.7%, the dynamic
fracture toughness of the phantom decreases up to 76.8%, and the crack length increases up to 325.8%. The increase in crack
length created by the high amplitude vibration allows for the substantial decrease in frictional force.

Keywords Tissue cutting . Needle insertion . Ultrasonic cutting . Force model

Introduction require high insertion accuracy to achieve effective interven-

tion efficacy and avoid biopsy miss-sampling. Poor needle
Well-designed needles are widely used in medical procedures, placement accuracy can be the results of several reasons in-
such as biopsy, localized drug delivery, radioactive seed place- cluding: asymmetric forces caused by the inhomogeneity of
ment, injection, and blood sampling [1–4]. Some procedures organs [5], asymmetric needle tip geometry [6], and high nee-
dle insertion forces creating bending of the needle [7]. High
needle insertion forces have been shown to create higher
* J.Z. Moore levels of pain and diminish the accuracy of needle placement
jzm14@psu.edu [8–11].
Several methods have been utilized to reduce the insertion
L. Tan force, such as rotating insertion [12], vibration needle inser-
Lei_tan7@hotmail.com
tion [13, 14], micro-needle [15], and improving needle tip
J.A. Jones geometry [16]. Rotating needle insertion can only very slight-
jjone417@gmail.com ly reduce the insertion force by 10% during needle insertion
A.C. Barnett into chicken breast, but it causes more tissue damage [17, 18].
acb279@psu.edu Needle tip geometry has been modeled, and sharper tip geom-
H. Zhang etry (higher inclination and rake angles) contributes to smaller
hongcaizh@sdu.edu.cn insertion force [19, 20]. As widely used in traditional machin-
Q. Zhang ing, ultrasonic vibration is utilized in milling and turning.
zhangqh@sdu.edu.cn These applications have shown that creating a vibration be-
tween the tool and workpiece contributes to reducing cutting
1
School of Mechanical Engineering, Key Laboratory of High forces and improving surface finish [21]. Vibrational cutting
Efficiency and Clean Mechanical Manufacture of Ministry of has also been applied in medicine to improve the cutting of
Education, Shandong University, 17923 Jingshi Road,
Jinan 250061, Shandong Province, China tissue. Ultrasonic scalpel can shorten surgery time, significant-
2
ly reduce the blood loss, and cause less postoperative pain [22,
Department of Mechanical and Nuclear Engineering, The
Pennsylvania State University, 319 Leonhard Building, University 23]. Recent ultrasonic needle insertion research has also
Park, State College, PA 16802, USA shown that adding vibration to needles can greatly reduce
 Exp Tech

1.6
the insertion force [14, 24, 25]. However, currently this re- First Phase Second Phase Third Phase
1.4
search lacks a firm understanding specifically how vibratory Max Insertion Force
parameters influence the cutting force on tissue, the topic of 1.2

Insertion Force (N)

this paper. 1
Initial Puncture Force
Needle insertion force models play an important role in
0.8
showing the relationship between insertion force and exper-
0.6
iment parameters, and contribute to reducing insertion forces
by allowing for optimal parameter selection. Researchers 0.4

have developed non-vibratory needle insertion force models 0.2

using experimental fitting of data, finite element analysis,
0
and analytical modeling techniques [26–31]. Some recent 0 2 4 6 8 10 12 14
researches have been developed to explain how needle vi- Insertion Depth (mm)

bration influences the frictional force on a needle [13, 32, Fig. 1 Needle insertion force in three puncture phase
33]. However, total insertion force consists of several forces
excluding friction force. What’s more, there is no research During the entire second phase, where the cutting is occur-
showing the effect of vibration on different forces and tissue ring, there are three forces acting to resist the needle motion:
properties. tearing force, spreading force and friction force. These three
This paper develops an experimentally based fracture me- forces are illustrated in Fig. 2, and will all have a dependence
chanics model to determine how ultrasonic vibratory parame- on the vibrational amplitude (A) during vibratory insertion.
ters influence the three major components of tissue cutting The tearing force Ft(A) is the force needed to create a crack
force: tearing force, spreading force, and friction force. This in the material. The spreading force Fs(A) is the insertion force
model explores the parameters of the workpiece to explain the needed to spread the crack open to allow the needle to pass
effect of the ultrasonic vibration amplitude on the physical through. The friction force Ff(A) is the frictional force that
characteristics. Synthetic materials is widely used in clinic occurs between the needle and the workpiece. These three
surgical simulations and cutting study because of providing forces sum to form the total insertion force F(A). The max
consistent results [34, 35]. Based on former experiment re- insertion force FM(A) at the end of second phase can be de-
sults, polyurethane (40A) is chosen as the skin simulation in scribed as in equation (1).
this paper [36]. Experiments are performed where the hypo-
dermic needle is inserted into the tissue simulant with a range
FM ðAÞ ¼ F t ðAÞ þ F s ðAÞ þ F f ðAÞ ð1Þ
of ultrasonic vibratory amplitudes. This paper describes nee-
dle cutting mechanics and the including fracture mechanics
model, describes the experimental procedure used to test
workpiece physical parameters, and fits models for the param-
eters of workpiece property. The final completed force model Fracture Mechanics Model
is obtained by incorporating all the fitted parameter models
and validated with experimental data. The change of the main Fracture mechanics modeling can be utilized to interpret the
forces with the increasing vibration amplitude is discussed. work done by forces [37]. A tissue fracture mechanics model
is formed dependent on ultrasonic vibration amplitude. In the

Needle Insertion Mechanics TISSUE

Force Model Spreading Force
As shown in Fig. 1, the needle insertion procedure starts from
the instant the needle touches the workpiece, and it can be
divided into three phases. In the first phase, the needle deflects NEEDLE
the material until puncture occurs. At the end of the first phase,
the initial puncture force occurs as at the moment that cutting Tearing Force
of the material begins and is marked by a sudden change in
force. In the second phase, the needle tip cuts the material until Friction Force
the entire needle tip punctures through the sample. In the third
phase, the needle is entirely through the workpiece and only
friction acts on the needle surface. Fig. 2 Three forces during needle insertion
Exp Tech

Table 1 Outline of experiment

procedures Testing Sample Needle Frequency (kHz) Peak-Peak Amplitude (μm)

Polyurethane(40A) Hypodermic16G 20 0, 29.7,57.2, 116.4, 148.5, 169.2,

230.2, 267.3, 288.5, 344.8,
386.1402.3, 459.7, 504.9, 517.2

second phase, for a differential insertion distance dl, the total F M ðAÞdl ¼ J Id ðAÞ*aðAÞdl þ μðAÞ*R2* f ðAÞdl
work of the cutting force is FM (A)dl. The work done by
tearing force is dWc(A), with the crack propagation. The dif- þ F f ðAÞdl ð7Þ
ferential work done by spreading force is stored in the phan-
tom as strain energy, dSE(A). The ultrasonic vibration pro- After canceling dl from both sides of the equation, the max-
duces energy in two ways: heat generation and cavitation imum force on a needle that occurs during the puncture can be
[38]. In a solid material, cavitation is difficult; therefore, cav- found as shown in equation (8). This assumes that the tissue
itation energy is not considered. Summing the energy trans- parameters are consistent throughout the insertion depth.
ferred creates the fracture mechanics equation as shown in
equation (2).
F M ðAÞ ¼ J Id ðAÞ*aðAÞ þ μðAÞ*R2* f ðAÞ þ F f ðAÞ ð8Þ

F M ðAÞdl ¼ d Wc ðAÞ þ d SE ðAÞ þ F f ðAÞdl ð2Þ Experiments are designed to obtain the dynamic fracture
toughness JId(A), friction force Ff(A), and crack length a(A).
Shear modulus μ(A) is considered as changing little under the
The energy, dWc(A), released by the crack generation in
effect of vibration, and the value of 417 kPa obtained by
model–I is given in equations (3) and (4) where JId(A) is the
Barnett et al. is utilized [36]. Contact factor f(A) is obtained
dynamic fracture toughness of phantom and da(A) is the incre-
by substituting tissue parameters and experimental data into
ment in crack area [39]. As ultrasonic cutting was introduced in
equation (8). All other factors other than vibratory amplitude;
the experiments, the vibrating needle tip presented stress waves
including frequency, insertion speed, and material properties,
during crack propagation. Therefore, the dynamic fracture
do not vary in the experiments.
toughness is utilized in this dynamic fracture procedure [40].
For a line Model–I crack of width a(A), da(A) = a(A)*dl where
a(A) is the crack length, dl is the crack depth [41].
Experimental Procedures

d Wc ðAÞ ¼ J Id ðAÞ*daðAÞ ð3Þ Twice puncture experiments were performed to obtained the
JId(A) and Ff(A). The experimental setup is discussed in Sec.
3.2 and the testing parameters are shown in Table 1. The
d Wc ðAÞ ¼ J Id ðAÞ*aðAÞ*dl ð4Þ
amplitude can be varied by altering the actuator input voltage.
In the range of the actuator input voltage, experiments were
The strain energy, dSE(A), per insertion length to open the taken at ten different vibration amplitudes (0, 57.2, 116.4,
crack, can be modeled as enlarging the radius of a circle from 169.2, 230.2, 288.5, 344.8, 402.3, 459.7, 517.2 μm) with 5
0 to R, where R is the needle radius [42]. Assuming plane
strain, the equation is given by equation (5). The crack hole
does not enfold the needle tightly, and it isn’t a regular circle.
Therefore, a non-dimensional contact factor f(A) is introduced
to balance the equation as shown in equation (6).
Micrometer

d SE ðAÞ ¼ 1=2 π μðAÞR2 dl ð5Þ Ultrasonic Horn Needle

d SE ðAÞ ¼ μR2 f ðAÞdl ð6Þ

Fiber-Optic Probe
By substituting equations (4) and (6) into (2), the equation
complete needle insertion force equation becomes equation (7). Fig. 3 Vibration amplitude measurement platform
 Exp Tech

Fig. 4 Vibratory Insertion

Ultrasonic
Experimental platform Plate Plates
Actuator

Needle Ultrasonic Horn

Needle

Force Sensor

Polyurethane (40A)
Linear Motor

trials for each amplitude. The validation experiments were in the second phase is described as equation (8). During the
performed at another 5 vibration amplitudes (29.7, 148.5, second puncture into the same hole, the force that occurs,
267.3, 386.1, 504.9 μm) with 3 trials for each amplitude. FM’(A), is equal to the sum of the spreading force and the
The crack length a(A) was measured using a stereo micro- friction force as shown in equation (9).
scope. The amplitude was measured using a fiber optic probe
as described in Sec. 3.1. F M ’ ðAÞdl ¼ d SE ðAÞ þ F f ðAÞdl ð9Þ

Vibration Amplitude Validation Experiments Subtracting equations (9) from (8), the equation is:

The needle amplitude was measured utilizing the experi-

mental setup shown in Fig. 3. The ultrasonic piezoelectric F M ðAÞ− F M ’ ðAÞ dl ¼ J Id ðAÞ*aðAÞdl ð10Þ
actuator used in experiments was at a resonance frequency
of 20 kHz. The amplitude measurements were taken on a
1.5
shock absorption table (Integrity 1, Newport, US). A
fiber-optic probe (MTI Instruments) was used to obtain 1.3
F
the vibratory amplitude through measuring the displace- 1.1
ment between needle tip and probe in real time. Reflective 0.9
F'
Force (N)

tape was attached to the tip of the needle to make the

0.7
reflected light stronger and easier to be recognized by
the probe. With this setup, the relationship between input 0.5

voltage and amplitude was determined. F-F'

0.3

0.1 a1 a2
Needle Insertion Experiments
-0.1 0 2 4 6 8 10 12 14
Insertion Depth (mm)
As shown in Fig. 4, an experimental setup including a force
sensor, linear motor, and ultrasonic piezoelectric vibratory ac- (a)
1
tuator is utilized to perform needle force experiments. The six- 0.9
axis force sensor (ATI Industries, Apex, NC) is used to record 0.8
the force on the needle during the needle insertion procedure. 0.7
The workpiece is a tissue simulant, 1.5 mm thick 0.6

Polyurethane (40A), which is fixed between two aluminum 0.5

0.4
plates. A linear motor (Dunkermotoren, Bonndorf, Germany)
0.3 Slope=Dynamic Fracture Toughness
provides insertion motion for needle at a stable speed of 1 mm/
0.2
s. The ultrasonic piezoelectric vibratory actuator (Kyoto
0.1
Kagaku, Kyoto, Japan) connects to the linear motor and pro- 0
vides the ultrasonic vibration. A horn is utilized to increase the 0 0.0005 0.001 0.0015 0.002 0.0025
amplitude of the needle vibration.
The phantom is punctured twice at the same position to (b)
calculate the fracture toughness as suggested by McCarthy Fig. 5 a Force against insertion depth for first and second insertion, b
et al. [43]. During the first puncture, the mechanical formula dynamic fracture toughness in the fracture work
Exp Tech

0.5mm 0.5mm

(a) (b)

0.5mm 0.5mm
(c) (d)
Fig. 6 Cracks formed with vibratory amplitude of a 116.4 μm, b 169.2 μm, c 344.8 μm, d 517.2 μm

In the steady cutting second phase, integrating both sides of fracture work, the integral of the difference in force FM(A)-
equation (10) creates equation (11) where a1 is the beginning FM’(A) with respect to insertion depth, is plotted against the
of second phase, a2 is the end of the steady cutting area. crack area in Fig. 5(b). The crack area can be obtained with the
a2
 0
 a2
crack length results using integration method. As shown in
∫a1 F M ðAÞ−F M ðAÞ dl ¼ J Id ðAÞaðAÞ∫a1 dl ð11Þ Fig. 5(b), the slope of the curve is the dynamic fracture tough-
ness of tissue simulation at the corresponding vibration
As shown in Fig. 5(a), a1 starts at the first force peak where amplitude.
the needle tip punctures into the phantom. The end of the As crack length is used during calculating the dynamic
steady cutting is at a2 where a second force peak occurs. fracture toughness, the polyurethane cracks are photographed
Utilizing the method proofed by Azar and Hayward [41], the by the electronic eyepiece from an optical microscope

Max puncture force

350
2
Friction force
300
1.5 Difference between two forces
250
Amplitude (µm)

Force (N)

200 1

150
0.5
Amplitude = 2.87 × Voltage
100

50 0
0 100 200 300 400 500 600
0 Vibration Amplitude (µm)
0 20 40 60 80 100 120
Fig. 8 Maximum puncture force and friction force are changed with
Voltage (V)
vibratory amplitude. The mean and standard deviation of each kind
Fig. 7 Vibratory amplitude compared to input voltage forces are provided (n = 5)
 Exp Tech

Fig. 9 Dynamic fracture 500

toughness is changed with 450

vibratory amplitude. Error bars
represent standard deviation from

Dynamic Fracture Toughness (kJ/m2)

400
the mean (n = 5)
350

300

250 J =326.9e(-((A-56.956)/144.2)^2)-107.4
d(A)

200

150

100

50

0
0 100 200 300 400 500 600
Vibration Amplitude (µm)

(Acscope, USA) and then measured using the microscope The amplitude measurements in these experiments are per-
software on the computer. Examples of crack length measure- formed on the shock absorption table in the air and under a no-
ments can be seen in Fig. 6. As noted depending on the am- load condition. Therefore, the actual amplitude during the
plitude, cracks created are either lines (lower vibratory ampli- cutting experiments will be lower due to the added cutting
tudes) or semi circles (higher vibratory amplitudes). force that will act to dampen the needle amplitude.

Influence of Vibration Amplitude on the Forces

Results and Discussion
Max puncture force FM(A), the friction forces Ff(A), and their
Vibration Measurement Results difference FM(A)-Ff(A) are plotted as squares, diamonds, tri-
angles separately in Fig. 8. Each point in the figure represents
As shown in Fig. 7, the peak-peak vibratory amplitude in- the mean experimental force, and the error bars indicate the
creases with the input voltage linearly. The trend line in Fig. standard deviations. From the 0 μm vibration amplitude to the
7 is fitted using the Curve Fitting Toolbox in MATLAB to 517.2 μm amplitude, the max puncture force and the friction
obtain the best-fit curve, and other curving fittings in this force are reduced by 60.8 and 80.7%, respectively. As can be
paper are the same. Through linear regression, the best fit seen from the Fig. 8, the max puncture force FM(A) decreases
equation of the data is shown in equation (12), and this has with the increasing vibration amplitude, which is mainly
an adjusted R2 value of 0.9998. caused by the decreasing of friction force Ff(A). Judging from
equation (1), the difference between the max puncture force
A ¼ 2:87* Voltage ð12Þ FM(A) and the friction force Ff(A) also means the sum of the

Fig. 10 Crack length is plotted 6

against vibratory amplitude Error
bars indicate standard deviation 5
(n = 5)
Crack Legth (mm)

3
a(A)=1.31tan-1(0.0339A-7.70)+3.06
2

0
0 100 200 300 400 500
Vibration Amplitude (µm)
Exp Tech

Fig. 11 Contact factors are 1.5

changed with vibratory
1.3
amplitude. Error bars represent
standard deviation from the mean 1.1
(n = 5) f(A)=1.19e(-((A-111.6)/151.7)^2)-0.140
0.9

Contact Factor
0.7
0.5
0.3
0.1
-0.1 0 100 200 300 400 500 600
-0.3
-0.5 Vibration Amplitude (µm)

tearing force and the spreading force, Ft(A) + Fs(A). It in- amplitude varies from 57.2 to 517.2 μm. This could be con-
creases with increasing amplitude in the front half section sidered as the combined results of hardening by vibration and
and decreasing in the back half section. the energy transferred from ultrasonic vibration.
For polyurethane material, the dynamic fracture tough-
Parameters Changing with Amplitude ness is higher than the static fracture toughness [44].
Vibratory cutting can be considered as thousands of impacts
Dynamic fracture toughness per second on the polyurethane workpiece. Therefore, the
workpiece is hardened under vibration. During the ultrasonic
The results of dynamic fracture toughness are shown in Fig. 9. vibration cutting phase, the section of workpiece where
The dynamic fracture toughness reduces up to 76.8% with holding the cutting edge moves forwards and backwards
vibratory amplitude increasing from 0 to 517.2 μm. A thousands times along with needle when the motor moves
Gaussian equation (13), is fit to the data with an adjusted R2 forward per second. With the section of workpiece move-
value of 0.994. ment, part of ultrasonic vibration energy is transferred into
the crack tip. The energy required to grow a crack becomes
smaller. Therefore, the dynamic fracture toughness reduces
J Id ðAÞ ¼ 326:9eð−ððA−56:956Þ=144:2Þ Þ −107:4
2
ð13Þ with ultrasonic vibration.
When the amplitude varies from 0 to 57.2 μm, the energy
transferred to rubber is small because of small amplitude, but
As shown in Fig. 9, the dynamic fracture toughness in- the rubber is hardened under vibration. According to this com-
creases with amplitude when the vibration amplitude changes prehensive function, the dynamic fracture toughness increases
from 0 to 57.2 μm, and reduces with amplitude when the at first. During the amplitude ranging from 57.2 to 517.2 μm,

Fig. 12 Friction forces are plotted

1.4
against vibratory amplitudes.
Error bars represent one standard
1.2
deviation around the mean (n = 5)
Friction Force (N)

1
Ff(A)= 1.39e-0.003A
0.8

0.6

0.4

0.2

0
0 100 200 300 400 500 600
Amplitude (µm)
 Exp Tech

Fig. 13 Max puncture forces are 2.2

varied with vibratory amplitude. 2
Each point represent the mean 1.8
Vibration insertion force model

Max Puncture Force (N)

validation experimental force, for
1.6
3 trials, while the error bars
indicate the standard deviation 1.4
1.2
1
Validation experimental results
0.8
0.6
0.4
0.2
0
0 100 200 300 400 500 600
Vibration Amplitude (µm)

the thermal energy transferred into rubber grows fast with the According to equation (6), this leads to the contact factor
increasing amplitude, which leads to the reduction of dynamic increasing with the increase of amplitude in the first period.
fracture toughness under higher amplitude vibration. As shown in Fig. 10, the crack length grows rapidly when the
amplitude is greater than 116.4 μm. As a result, the hole
Crack length around the needle formed by cutting becomes much larger
with increasing amplitude. Then, less the work by spreading
The crack length of rubber after needle puncturing is mea- force is needed to enlarge the hole. Therefore, the contact
sured, and the mean results along with the standard deviation factor decreases with increasing amplitude when the ampli-
are plotted in Fig. 10. The crack length increases up to 325.8% tude is greater than 116.4 μm.
with vibratory amplitude. An arc tan fit is applied to the data The contact factor becomes negative after the amplitude
with an adjusted R2 value of 0.996. The equation of the fit is: larger than 344.8 μm. This is caused by dynamic fracture
toughness over estimation. After the first needle insertion,
aðAÞ ¼ 1:31tan−1 ð0:0339A−7:70Þ þ 3:06 ð14Þ Large amounts of energy remains in the workpiece around
needle because lots of thermal energy is generated by dry
friction force between the needle and workpiece under the
As shown in Fig. 10, the crack length increases with the
ultrasonic vibration. This energy causes the material to soften,
increasing vibration amplitude. As discussed in Sec. 4.3.1,
which means less work done by spreading force is needed to
less energy is required to grow the crack during the ultrasonic
spread the hole open on the second insertion. Therefore, the
vibration insertion because part of ultrasonic vibration energy
strain energy during the second insertion, d’SE(A), is much
is transferred into the crack tip. Larger vibration amplitude
smaller than the strain energy in first insertion.
results in more energy being transferred into the crack tip.
Therefore, the workpiece becomes easier to cut with the in-  0

crease of vibration amplitude. As a result, the crack length F M ðAÞ− F M ðAÞ dl ¼ J Id ðAÞ*aðAÞdl
increases with the increasing vibration amplitude.
þ ðd SE ðAÞ−d 0 SE ðAÞÞ ð16Þ
Contact factor
As can be seen from equation (16), JId (A)*a(A)dl in equa-
The contact factor over varying amplitudes is shown in Fig. tion (11) is over estimated. According to equation (8), it can be
11. A Gaussian fit of the relationship between the f and A is found that the contact factor becomes negative.
found in equation (15) with an adjusted R2 value of 0.9351.
Friction force
f ðAÞ ¼ 1:19eð−ððA−111:6Þ=151:7Þ Þ −0:140
2
ð15Þ
The friction force’s relationship to amplitude is shown in Fig.
12. The friction forces reduce up to 80.7% with amplitude. The
The contact factor increases initially and then decreases
Ff(A) is found to have an exponential relationship to amplitude
with the increase of vibration amplitude. As discussed in Sec
with an adjusted R2 value of 0.9877 as shown in equation (17).
4.3.1, the ultrasonic vibration insertion would harden the
workpiece, thereby more work done by spreading force is
needed to enlarge the hole to let the needle pass through. F f ðAÞ ¼ 1:39e−0:003A ð17Þ
Exp Tech

Vibration in traditional machining has been shown to References

reduce friction in metal-metal condition [45]. Recent re-
searches have shown that vibration contributes to reduc- 1. Abolhassani N, Patel R, Moallem M (2007) Needle insertion into
ing the friction in metal-tissue condition [32, 46]. The soft tissue: a survey. Med Eng Phys 29:413–431
vibration changes the friction coefficient between needle 2. Elgezua I, Kobayashi Y, Fujie MG (2013) Survey on current state-
of-the-art in needle insertion robots: open challenges for application
and tissue and thereby reduces the amount of friction that in real surgery. Procedia CIrP 5:94–99
occurs on the needle. 3. Hirsch L, Gibney M, Berube J, Manocchio J (2012) Impact of a
modified needle tip geometry on penetration force as well as ac-
Force Model ceptability, preference, and perceived pain in subjects with diabetes.
J Diabetes Sci Technol 6:328–335
4. Yang C, Xie Y, Liu S, Sun D (2018) Force modeling, identification,
Overall, the max puncture force decreases with the ampli- and feedback control of robot-assisted needle insertion: a survey of
tude both in the validation experiments and in the model, the literature. Sensors 18:561
as shown in Fig. 13. The model results are obtained by 5. Datla NV, Konh B, Honarvar M, Podder TK, Dicker AP, Yu Y et al
substituting the value of parameters Eqs. (13), (14), (15) (2014) A model to predict deflection of bevel-tipped active needle
advancing in soft tissue. Med Eng Phys 36:285–293
and (17) into (8). The standard deviation for the model is
6. Sadjadi H, Hashtrudi-Zaad K, Fichtinger G (2014) Needle deflec-
extracted from experimental data where the largest stan- tion estimation: prostate brachytherapy phantom experiments. Int J
dard deviation is 0.2 N. Comput Assist Radiol Surg 9:921–929
In order to validate the force model, five different am- 7. Misra S, Reed KB, Douglas AS, Ramesh K, Okamura AM. (2008)
plitudes (20.9, 104.4, 187.9, 271.4, 354.8 μm) experi- Needle-tissue interaction forces for bevel-tip steerable needles.
2008 2nd IEEE RAS & EMBS International Conference on
ments are conducted with 3 trials for each amplitude. As Biomedical Robotics and Biomechatronics, pp. 224–231
can be seen from Fig. 13, the plotted experimental results 8. Egekvist H, Bjerring P, Arendt-Nielsen L (1999) Pain and mechan-
follow the model well and the model slightly over pre- ical injury of human skin following needle insertions. Eur J Pain 3:
dicts. This error is a result of overestimation of dynamic 41–49
fracture toughness, which is discussed in Sec. 4.3.1. 9. Halabian M, Beigzadeh B, Karimi A, Shirazi HA, Shaali MH
(2016) A combination of experimental and finite element analyses
However, the model is valid because all the experimental of needle–tissue interaction to compute the stresses and deforma-
results are within the model with the 0.2 N maximum tions during injection at different angles. J Clin Monit Comput 30:
error. 965–975
10. Barnett AC, Lee Y-S, Moore JZ (2016) Fracture mechanics model
of needle cutting tissue. J Manuf Sci Eng 138:011005
11. DiMaio SP, Salcudean SE (2003) Needle insertion modeling and
Conclusion simulation. IEEE Trans Robot Autom 19:864–875
12. Meltsner MA, Ferrier NJ, Thomadsen BR (2007) Observations on
Force model is built by utilizing fracture mechanics model, at rotating needle insertions using a brachytherapy robot. Phys Med
the same time, the differences between force model predic- Biol 52:6027–6037
13. Begg ND, Slocum AH (2014) Audible frequency vibration of
tions and the experimental results are within the standard de-
puncture-access medical devices. Med Eng Phys 36:371–377
viation for the model. The effect of vibration amplitude on 14. Huang YC, Tsai MC, Lin CH (2012) A piezoelectric vibration-
insertion force and phantom parameters are explored. The based syringe for reducing insertion force. IOP Conf Ser: Mater
max puncture forces and friction forces reduce up to 60.8 Sci Eng 42:012020
and 80.7% separately when vibration amplitude increases 15. Yang M, Zahn JD (2004) Microneedle insertion force reduction
using vibratory actuation. Biomed Microdevices 6:177–182
from 0 to 517.2 μm. Dynamic fracture toughness decreases
16. Moore JZ, McLaughlin PW, Shih AJ (2012) Novel needle cutting
up to 76.8% and crack length increases up to 325.8% with the edge geometry for end-cut biopsy. Med Phys 39:99–108
vibration amplitude increasing from 0 to 517.2 μm. The max 17. van Gerwen DJ, Dankelman J, van den Dobbelsteen JJ (2012)
puncture forces decrease with amplitude mainly because of Needle–tissue interaction forces–a survey of experimental data.
the reduction of friction forces. This also means tissue be- Med Eng Phys 34:665–680
comes easier to cut, but getting more damage. The amplitude 18. Wedlick TR, Okamura AM. (2012) Characterization of robotic nee-
dle insertion and rotation in artificial and ex vivo tissues. 2012 4th
makes little difference on the sum of tearing force and spread- IEEE RAS & EMBS International Conference on Biomedical
ing force. From the results, it is proofed that ultrasonic vibra- Robotics and Biomechatronics (BioRob), pp. 62–68
tion is effective to reduce insertion force. 19. Barnett AC, Lee Y-S, Moore JZ. (2016) Needle geometry effect on
vibration tissue cutting. Proc Inst Mech Eng B J Eng Manuf:
Acknowledgments This material was based upon the work supported by 0954405416654188
the National Science Foundation (NSF) under Grant No. CMMI- 20. Moore JZ, Shih AJ, McLaughlin P, McGill C, Zhang Q, Zheng H
1404916, the National Natural Science Foundation of China (Grant No. (2009) Blade oblique cutting of tissue for investigation of biopsy
51475274) and the Specialized Research Fund for the Doctoral Program needle insertion. Trans NAMRI/SME 37:49–56
of Higher Education (Grant No. 20130131110070). 21. Neppiras E (1964) Ultrasonic machining and forming. Ultrasonics
2:167–173
 Exp Tech

22. Deo S, Shukla N, Asthana S, Niranjan B, Srinivas G (2002) A 33. Tan L, Qin X, Zhang Q, Zhang H, Dong H, Guo T et al (2017)
comparative study of modified radical mastectomy using harmonic Effect of vibration frequency on biopsy needle insertion force. Med
scalpel and electrocautery. Singap Med J 43:226–228 Eng Phys 43:71–76
23. Kandil T, El Nakeeb A, El Hefnawy E (2010) Comparative study 34. Dąbrowska A, Rotaru GM, Derler S, Spano F, Camenzind M,
between clipless laparoscopic cholecystectomy by harmonic scalpel Annaheim S et al (2016) Materials used to simulate physical prop-
versus conventional method: a prospective randomized study. J erties of human skin. Skin Res Technol 22:3–14
Gastrointest Surg 14:323–328 35. Peattie R, Golden E, Nomoto R, Margossian C, Pancheri F, Edgar E
24. Liao X, Sadiq M, Corner G, Cochran S, Huang Z. (2013) Reduced et al (2015) A technique for comparing wall pressure distributions
penetration force through ultrasound activation of a standard nee- in steady flow through rigid versus flexible patient-based abdomi-
dle: an experimental and computational study. 2013 IEEE Int nal aortic aneurysm phantoms. Exp Tech
Ultrason Symp (IUS), pp. 1436–1439 36. Barnett AC, Tan L, Barrett J, Moore JZ. (2016) Needle Cutting of
25. Liao X, Sadiq M, Kuang Y, Corner G, Cochran S, Huang Z. (2014) Skin Simulants. ASME 2016 11th International Manufacturing
Performance optimization of ultrasonic needle actuating device for Science and Engineering Conference, pp. V002T003A008-
insertion operation into tissue mimics. 2014 IEEE international V002T003A008
Ultrasonics symposium, pp. 823–826 37. Mahvash M, Dupont PE (2010) Mechanics of dynamic needle inser-
26. Kobayashi Y, Sato T, Fujie MG. (2009) Modeling of friction tion into a biological material. IEEE Trans Biomed Eng 57:934–943
force based on relative velocity between liver tissue and 38. Cheeke JDN. (2012) Fundamentals and applications of ultrasonic
needle for needle insertion simulation. 2009 Annual waves, CRC press
International. Conference of the IEEE Engineering in 39. Shergold OA, Fleck NA. (2004) Mechanisms of deep penetration of
Medicine and Biology Society, pp. 5274–5278 soft solids, with application to the injection and wounding of skin.
Proceedings of the Royal Society of London A: Mathematical,
27. Asadian A, Kermani MR, Patel RV. (2010) A compact dynamic
Physical and Engineering Sciences, pp. 3037–3058
force model for needle-tissue interaction. 2010 Annual
40. Knauss W, Ravi-Chandar K. (1985) Some basic problems in stress
International Conference of the IEEE Engineering in Medicine
wave dominated fracture. Dynamic fracture. Springer, pp. 1–17
and Biology, pp. 2292–2295
41. Azar T, Hayward V. (2008) Estimation of the fracture toughness of
28. Jiang S, Li P, Yu Y, Liu J, Yang Z (2014) Experimental study of soft tissue from needle insertion. Inte Symp Biomed Simul, pp.
needle–tissue interaction forces: effect of needle geometries, inser- 166–175
tion methods and tissue characteristics. J Biomech 47:3344–3353 42. Shergold OA, Fleck NA (2005) Experimental investigation into the
29. Maghsoudi A, Jahed M (2012) Needle dynamics modelling and deep penetration of soft solids by sharp and blunt punches, with
control in prostate brachytherapy. IET Control Theory Appl 6: application to the piercing of skin. J Biomech Eng 127:838
1671–1681 43. McCarthy CT, Hussey M, Gilchrist MD (2007) On the sharpness of
30. Moore JZ, Malukhin K, Shih AJ, Ehmann KF (2011) straight edge blades in cutting soft solids: part I–indentation exper-
Hollow needle tissue insertion force model. CIRP Ann iments. Eng Fract Mech 74:2205–2224
Manuf Technol 60:157–160 44. Marsavina L, Linul E, Voiconi T, Sadowski T (2013) A comparison
31. Okamura AM, Simone C, O'Leary MD (2004) Force model- between dynamic and static fracture toughness of polyurethane
ing for needle insertion into soft tissue. IEEE Trans Biomed foams. Polym Test 32:673–680
Eng 51:1707–1716 45. Godfrey D (1967) Vibration reduces metal to metal contact and
32. Khalaji I, Hadavand M, Asadian A, Patel RV, Naish MD. (2013) causes an apparent reduction in friction. A S L E Trans 10:183–192
Analysis of needle-tissue friction during vibration-assisted needle 46. Sfakiotakis M, Pateromichelakis N, Tsakiris DP (2014) Vibration-
insertion. 2013 IEEE/RSJ International Conference on Intelligent induced frictional reduction in miniature Intracorporeal robots.
Robots and Systems, pp. 4099–4104 IEEE Trans Robot 30:1210–1221