Surface EMG-based Profiling and Fatigue Analysis of the Biceps

Brachii Muscle of Cricket Bowlers

Muhammad Usama Rizwan, Nadeem Ahmad Khan, Rushda Basir Ahmad and Muneeb Ijaz

Department of Electrical Engineering, Lahore University of Management Sciences, Lahore, Pakistan

Keywords: EMG, Mean Power Frequency, Fatigue Analysis, Cricket Bowling, Player Profile, Player Development.

Abstract: Cricket bowling action is a complex repetitive task involving multiple muscles. In this paper we present a

protocol to analyse accumulated localized fatigue in muscles during cricket bowling action. Biceps Brachii

(BB) muscle in case of fast delivery for a novice player is analysed to illustrate the methodology.

Synchronized video recording with the surface EMG signal was captured from the medial position of the BB

muscle to enable segmentation of the EMG signal in six intervals corresponding to the six phases of the

bowling action. This enables study of the activation pattern of the muscle along with the fatigue trend during

bowling. Both integrated EMG and Mean Power frequency (MPF) are used as measures to analyse fatigue.

Though we have plotted the trends for a single muscle, a similar exercise should be repeated for all important

muscles involved. Analysing localized fatigue in individual muscles is important for injury prevention as well

as player performance development. It can help to see how individual muscle fatigue contributes in declining

performance during cricket bowling. Such an analysis can also be used to support minimum bowling overs

and suitable inter-over breaks for a specific bowler with regard to injury prevention and optimal performance.

1 INTRODUCTION

As sports are becoming more and more competitive

the maximum player performance has become

inevitable in order to outsmart others in this

competition. The repetitive muscles activity results in

the fatigue accumulation in the muscles. The fatigue

in the muscle results in the retarded muscles

efficiency and that is why the detection of fatigue in

athletes is a crucial task. The literature reviews in this

area suggest that a lot of attempts had been made to

model fatigue accumulation and fatigue graphs have

been plotted using non-invasive biopotential

techniques such as Electromyography (EMG). This

paper aims at EMG profiling of cricket bowlers of

Fast category and using it for performance

development of cricket bowlers in combination with

other captured sensory data.

The study includes the processing of EMG signals

as well as video data from a subject to correlate the

two and bring them in a form that enables us to see

how individual muscle fatigue contributes in

declining performance during cricket bowling. The

two sources of data have to be synchronized first in

order to develop correlation between them. EMG

signals were recorded using Shimmer device, which

is a commercially available device for acquiring bio-

potential data. The data is logged and stored in

Structured Query Language (SQL) based database

with configurable sampling rate. Whereas, videos

were recorded using high speed camera set up and

interfaced to VICON software by NEXUS.

This paper contributes with increase in the Quality

of Fast Bowling by giving a detailed interpretation of

the results concluded from recorded data. Detection

of partial range of performance can be done while

analysing the data, the level of muscle fatigue can be

interpreted using results. A degradation in the quality

of bowling in fast bowling, over the successive bowls

is an indicator of the rising level of fatigue in muscles.

In this way weak muscles can be identified that are

contributing to the decreasing performance and

deteriorating the quality of bowling. Such trends can

help us indicate severe injuries in players and reduce

such injuries by timely measures.

With the help of the protocol defined in this paper,

profiling will indeed help in Player development by

supporting a sports trainer who will be analyzing

current acquisition data with player profile. If trainer

identifies that muscle activation has reduced,

dedicated muscle exercises will be advised. If there is

a flaw in technique, that can be identified using video

192

Rizwan, M., Khan, N., Ahmad, R. and Ijaz, M.

Surface EMG-based Proﬁling and Fatigue Analysis of the Biceps Brachii Muscle of Cricket Bowlers.

DOI: 10.5220/0010250801920199

In Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2021) - Volume 4: BIOSIGNALS, pages 192-199

ISBN: 978-989-758-490-9

analysis, then player will be working on correcting

the techique of bowling action that will lead to player

development.

The compact model for a bowling data cannot be

materialized due to inter-subject variability and wide

range of bowling actions. The solution to this is that

instead of developing a single model for Fast bowlers,

individual profiles of players should be built, and

each player should be analyzed based on his own

profile. For setting some standard for comparison a

profile of an elite can be logged but caveat will still

be inter-subject variability.

2 RELATED WORK

A lot of research has been carried out in the field of

sports kinematics and EMG. Our study expands on

the present work by integrating EMG techniques with

existing bio-mechanical techniques.

Surface Electromyography (EMG) is the science

and basic technique used for the quantiﬁcation of

muscle activity during movement which is popular

being a non-invasive technique. It is a hassle-free

procedure that can be used to determine the amount

of muscle activation throughout a given movement

and its activation timings and is an essential tool in

biomechanical and biomedical investigations.

Normally the EMG signal is between 5-450 Hz

window. The importance of muscles is based on the

EMG activity shown by them. Based on above

discussion, muscle showing the higher trend during

fast cricket bowling should be selected. Therefore,

Biceps Brachii (BB) muscle has been opted for the

analysis.

The studies by (Schmidt et al.,1999) and (Lloyd

et al.,2000) observe only the physical movement of

upper extremities. Since sports techniques are much

more elaborate than joint angles, EMG signals can be

used to provide more information and insight into

different actions, as well as their relation to muscle

fatigue and injuries. EMG signals refer to the

electrical signals generated when a muscle contracts,

and are measured using EMG sensors.

During the delivery of the ball, the most upper

limb muscles; BB, pectoralis major, deltoid,

trapezius, latissimus dorsi, infraspinatus, trapezius,

serratus anterior, and supraspinatus muscles are

mostly active. (Ahamed et al.,2014)

The EMG frequency decreases with time,

indicating an increasing degree of fatigue. This

provides us with a useful method of injury prevention.

The relationship between EMG and fatigue is, as

(Florimond,2008) has written, that “muscle fatigue is

accompanied by a decrease in motor unit firing rate

and conduction velocity.” This result has been

observed in many studies, including those by (Raez et

al.,2006). Thus, the median and mean frequencies

decrease as fatigue increases therefore, muscle

fatigue index has been defined using the mean and

median frequency of the EMG spectrum. The muscle

fatigue index decreases as contraction time increases.

While both median and mean frequencies may be

used as the fatigue index, as per (Florimond,2008),

median frequency is less sensitive to noise and more

sensitive to biomechanical factors so using mean

power frequency can be used for analysis.

Furthermore, as another indicator of fatigue, the

EMG amplitude is also found to increase with fatigue.

This is due to additional fibres being used to generate

the same level of force. EMG amplitude is used to

indicate fatigue when movement is required, in

exercises (Florimond,2008). Similar results have

been drafted by (Shorter et. al. 2010) for

interpretation of injury which is done through muscle

activity.

(Ahamed et al.,2014) have defined phases of a

bowling action as Run-up, Pre-Delivery Stride, Mid

Bound, Back-Foot Contract, Front-Foot Contract,

Release of the Ball and Follow through and compared

different activation trends during these phases.

3 METHODOLOGY AND

PROTOCOL

This study focuses only on BB muscles but same or

similar procedure can be used to analyse fatigue in

other muscles such as latissimus dorsi and soleus

muscles also. We plan to design a protocol for

professional players and then increase number of

subjects under study.

3.1 Subject Preparation

The subject is given an overview of the experiment if

he is interested a consent form is to be signed by the

subject. Then the Subject Data Performa is filled in

order to have an idea of BMI and exercise history of

the subject. The subjects are specifically inquired

about any history of neuromuscular injury. The initial

biceps envelop recordings are made and trial EMG of

subject is recorded using bicep curls experiment to

see that recordability of the muscle of the specific

subject.

A Bicep envelop is formed using measurement

tape and the electrode positions are marked. A typical

Bicep envelop has been shown in Figure 1. The bold

Surface EMG-based Proﬁling and Fatigue Analysis of the Biceps Brachii Muscle of Cricket Bowlers

193

positions in Figure.1 are medial and lateral points, on

has been marked. For reference an electrode position

has been marked near bony structure of elbow. The

skin must be shaved and cleansed with alcohol so that

the EMG signal is noise free. After that the skin is

abraded with skin cleaning gel and alcohol swabs are

rubbed at skin contact point to remove dead skin cell

and to have better data capturing.

Figure 1: Biceps Envelop for Electrode placement.

For the scope of this paper the data acquired from

Medial Channel has been reported. The electrode

placement on subject has been shown in Figure 2.

Figure 2: Electrode Placement on a Subject.

3.2 Protocol of the Experiment

Each bowler must perform 3 overs as a trial. As there

are six deliveries in an over hence 18 bowls per trial

will be recorded. Players should not be exhausted or

should not have muscular tension prior to the trial.

Subjects should remain hydrated throughout the trail.

A rest period of up to five minutes is given between

the each over so to mimic the actual inter-over rest of

a cricket match. The Run ups for each delivery should

be constant.

3.3 Phases of Cricket Bowling

Following are the phase segmentation as described in

by (Ahamed et al.,2014) we segmented our deliveries

in six segments.

a) Run-Up (RU)

The ball is held in the palm with the arm hanging

straight down and the shoulder at 0

degree abduction

neutral rotation.

b) Pre –delivery Stride (PS)

Biceps muscle gets slightly contracted through the

external torque provided by low-load weihted ball.

This torque also enables elbow movement and the 90

abduction of shoulder muscles with max external

rotation

c) Mid Bound (MB)

Shoulder provided forward elevation and the arms are

lifted at greater than 90

angle backwards at ear level.

d) Back and Front-foot Contact (BFC)

Combined phase of Foot contractions. The end of the

stride will be identified by the contact of the back foot

with the crease.

e) Release of Ball (RB)

The point where the ball will be released and the arms

will be in a vertical condition. The elbow extension

strength at 100

– 120

of elbow joint continues until

the ball is released.

f) Follow-Through (FT)

The action after the ball release until the final interval

of arm motion dissipating the deceleration forces, is a

follow through.

Figure 3: Video Segmentation Results in Phases.

4 EXPERIMENTAL METHOD

The subject identification is important for the

experiment. For this experiment Novice subjects are

required who have at least 90 kph pace and level of

maturity in their line, length and bowling action

during a cricket bowling. The reason for selecting a

Novice Subject is a better fatigue development in the

subjects of this category due to less muscle

endurance.

Data Acquisition was conducted in ICC

Biomechanics Lab LUMS. EMG data lead us to

detailed muscle activity analysis while Video Data

was helpful in segmenting the deliveries of trial into

respective phases by picking the corresponding

segment in EMG data as given by the phase-wise

segmentation of Video of deliveries.

4.1 Set-up

With the help of Shimmer device multiple

Biopotential signals can be measured simultaneously

BIOSIGNALS 2021 - 14th International Conference on Bio-inspired Systems and Signal Processing

194

in real-time. ConsensysPRO software is helpful in

streaming for live visualization. The raw EMG data

can be logged in the personal computer storage and

also to an SD cards. Consensys Software interfaces

EMG data via Bluetooth and it is stored and compiled

in the form of definite sessions by labelling

descriptions and creating new Subjects for each trial.

Each over has six balls, so 18 sessions are stored per

trial. Consensys enables to visualize the live

acquisitions of data. The sampling rate of the software

must be matched with the hardware Shimmer device

in order to have synchronized data. After data logging

in Consensys the data is exported in the form of Excel

file which can be readable in MATLAB. For this

experiment the interfacing and analysis was

performed on MATLAB. It was the main software

used for data cleaning, pre-processing and data

analysis. From these corresponding points, data was

labelled using frame by frame video and the

corresponding per second reading of the data. This

project combines EMG with existing bio-mechanics

protocols. This addition of EMG can help the bio-

mechanics field expand its horizons in terms of player

injury profiles and fatigue analysis

4.2 Data Synchronization

The data from the phases was identified by looking at

the video after the correlation of EMG with the video

data. Shimmer device is used to acquire the EMG data

and is attached via electrodes on the Subject’s biceps

muscle. For this experiment we have selected

sampling rate to be 1024 samples per seconds.

Two high speed cameras, one was placed at the

back side and one was placed side-ways, recorded the

video data. The Video data is segmented into frames

using a MATLAB program. The Frame rate of high-

speed cameras is 125 fps whereas Sampling rate of

Shimmer is equal to 1024 samples per sec. So, each

frame of video data almost equals to 9 samples of

EMG data. First the delivery of a trial is segmented in

the subsequent phases as depicted in Figure 3 and the

starting and ending samples of Video are noted.

Based on the number of frames of video recording

and their relative fraction in the trial, the

corresponding EMG samples are then calculated

using the same fractional ratio in EMG data. Using

this technique and segmentation of video data, the

EMG signal is then segmented in the corresponding

windows of samples using frames of Video Data.

Each window of samples corresponds to the

segmented phase-wise activity of a cricket delivery.

There were some synchronization problems in few

deliveries, which were eliminated using the visual cue

analysis for phase segmentation in EMG data.

4.3 Data Processing

The processing starts from importing Raw EMG data

in MATLAB. A Moving average filter with window

size of 100 samples is applied, which has Low Finite

Impulse Response (FIR) hence is used to smoothen

the signal. After data smoothening, a Band pass filter

of pass band 5-450 Hz is applied using a MATLAB

program. The resultant is a denoised and smoothen

signal. After this the phase segmentation through

Video Data analysis is incorporated. The phases wise

time domain plots are then used to visualize the

activity in each phase. The mean activity of each

phase is then plotted per ball in order to validate the

data as given by the literature review.

The data was changed into its frequency domain

transform and from there, mean frequencies were

calculated. Mean Power Frequency (MPF) of each

phase is also computed an analysed. The Integrated

EMG (IEMG) and MPF activity of each bowl is

computed and cumulative plots per over and per trials

are also plotted.

The Integrated EMG (IEMG) is the cumulative

sum of the absolute value of EMG signal.

𝐼𝐸𝑀𝐺   |𝐸



𝑛







The E(n) is the discrete activity function of EMG

signal obtained from the sensor, normally in mV, and

summing this activity function over a definite interval

having N samples gives IEMG of this interval. In

continuous case summation will be mapped to

integration, but as far as the scope of this project is

concerned the data is discrete. As the muscle activity

increases more and more muscle fibers are being

activated and |E(n)| increases, therefore the value of

IEMG increases as subject’s activity is increased.

There is also a normalized version of IEMG which is

called as Sample Normal IEMG and defined as IEMG

per sample of a given data of N samples

𝑆𝑎𝑚𝑝𝑙𝑒 𝑁𝑜𝑟𝑚𝑎𝑙 𝐼𝐸𝑀𝐺 

𝐼𝐸𝑀𝐺

𝑁

The Mean activity within a phase gives the

average unipolar IEMG of that phase per unit sample.

By definition, it is same as Sample Normal IEMG.

Mean activity of EMG data having N number of

samples is defined as

𝑀𝑒𝑎𝑛 𝐴𝑐𝑡𝑖𝑣𝑖𝑡𝑦 

∑

|𝐸



𝑛







𝑁

Surface EMG-based Proﬁling and Fatigue Analysis of the Biceps Brachii Muscle of Cricket Bowlers

195

Mean Power Frequency is the average frequency

of a signal, which is calculated as the sum of product

of the EMG power spectrum and the frequency

divided by the total sum of the power spectrum.

𝑀𝑃𝐹 

∑

𝑓𝑖 𝑃𝑖





∑

𝑃𝑖





where, fmax is the maximum frequency of the data

and fmin is the lowest one. And P(i) denotes the

power spectrum of the signal. There is a noticeable

decreasing trend in the MPF of an exercise. As the

fatigue point approaches the MPF shifts towards the

lower frequencies.

5 RESULTS

Once we had our data and it was processed, we

analysed the data to produce three distinct studies and

results; Phase-wise time segmentation, IEMG

Analysis and MPF analysis A detailed description of

these studies is given below:

The phase-wise segmentation of the trial was

performed using video analysis data as explained in

section 4.2 in detail. A typical phase segmentation is

shown in Figure 4 where the red lines shows partition

of respective phases. Ball wise time domain

segmentation of some deliveries of the trial are given

Figure 5.

Figure 4: Labelling of six phases of a typical delivery.

By definition, the mean activity is the same as

sample normal IEMG. There is a gradual increase in

the activity as the deliveries progress in the trial.

Figure. 6 provides a detailed Medial mean phase

activity of the trial. It can be observed that there are

significant activity trends in MB, BFC and RB

phases.

After delivery wise phase segmentation analysis,

the averages of all phases have been computed over

the trial and plotted in Figure. 7. The results

elucidates that the RB phase is the most active phase

of cricket bowling followed by BFC and MB as

concluded by (Ahamed et al.,2014).

Figure 5: Segmentation in six phases of different deliveries

in the same sequence as identified in Figure 4.

Figure 6: Phase-Wise Mean Activity of individual Delivery

3 (a), Delivery 8 (a), Delivery 16 (a), Delivery 18 (a) of

Trial.

Delivery 3 Delivery 7

Delivery 15 Delivery 18

BIOSIGNALS 2021 - 14th International Conference on Bio-inspired Systems and Signal Processing

196

We computed the Average of Phase-wise Mean

activity of 18 Deliveries in Figure. 7, which helps in

player profile analysis and it gives the insight of

activity in each respective phase of cricket bowling.

The reason of not averaging different subjects in this

average is inter-subject variability. Due to Inter-

subject variability we cannot model all fast bowlers

using same model rather we have to analyze

individual player profiles (Rushda et al., 2020).

Figure 7: Average of Phase-wise Mean activity (in mV) of

18 Deliveries.

In Figure.8, complete trial IEMG of Medial channel

data is plotted. The general trends overwise shows an

increasing trend since, more action potentials are

being activated. The gap between the overs stabilize

the activity and due to this 5 to 7 minutes, rest IEMG

decreases before the onset of new over. The red line

marked in the graphs of Figure.8 shows the over

partition.

Figure 8: Delivery-wise IEMG trends for complete

Bowling Action.

The MPF trends are consistent for the segments of

the delivery in which there is relatively high activity

is observed. The Best window is the window of 1500

samples in a delivery having maximum activity as

shown in Figure 9. It is normally MB, BFC and RB

phases of the delivery. The reason for taking best

window is that MPF trends are significantly

decreasing in the areas of high activity i.e. best

window, in a trial.

Figure 9: Time Domain Best window plot of Delivery 4.

Best window MPFs of complete trial has be

plotted in the Figure. 10 while in Figure.11 phase-

wise MPF trends over the complete trial are plotted.

Figure 10: MPF trends of Best Window of complete trial.

Referring to the Figure.10 and Figure.11 the plots

are depicting that during all the phases of a cricket

bowling MPF values have different starting values.

This is due to the fact, that in different phases

different amount of power is required and therefore,

the MPF values have different values during each

phase. Another useful insight due from Figure.10 and

11 is that in all the phase of trial Medial side, the MPF

has a decreasing trend. With the onset of new over,

i.e.; at 7th and 13th delivery due to inter over resting

time the muscle activity is relaxed hence MPF results

are showing slight increases at those points, but the

Surface EMG-based Proﬁling and Fatigue Analysis of the Biceps Brachii Muscle of Cricket Bowlers

197

Figure 11: Separate MPF trends of six phases per trail.

overall trend is decreasing as supported by literature

review.

6 DISCUSSION AND

INTERPRETATION OF

RESULTS

MPF has been used as fatigue index, like stated in

(Hwaang et. al., 2016) that as fatigue sets in the

muscle the value MPF decreases to the 60% of its

initial value. This 60% point can be taken as failure

point or onset of fatigue. From Figure 10 and 11 it is

clear that in case of Novice subject the fatigue sets in

during the first over in BB muscle. BB muscle is

active during fast bowling. BB muscle is relatively

more active during MB and RB phases of cricket

Bowling trial. Signiﬁcant differences between the

phases of fast bowling were found. The entire results

support our hypothesis and validate the trends as

explained through literature reviews.

The Figure 8 depicts that the IEMG trends are

increasing with the increase of number of deliveries

of the trial. During the overs especially if best

windows of each deliveries are considered MPF

activity is decreasing. Same trends were visualized

when we analyse phase-wise MPFs trends or best

activity phases.

Mean phase activity trends follow the same

pattern as describes by literature review. This provide

a basic understanding of the BB muscle activation

pattern during a typical cricket bowling. By the better

understanding of activity trends and identification of

important phases, using video analysis, a Sports

trainer can focus and analyse the phases RB, BFC and

MB because being the phases of higher activity

makes them more prone to injury.

BIOSIGNALS 2021 - 14th International Conference on Bio-inspired Systems and Signal Processing

198

7 CONCLUSIONS

The main objective of this paper was developing a

protocol to set up a support system for data

acquisition for a trainer or cricket coach. After the

detailed literature review and interviews of Cricket

Coaches protocol of the acquisition is being set up.

After that phase-wise analysis of deliveries was

performed which validate the results from literature

reviews and elucidate that RB and MB phases are the

relative active phases of a normal delivery and during

phases as muscles are mostly active it makes the

bowler more prone to injuries and it makes sense

because during Release of Ball bowlers jerk to pull

out maximum speed of the ball and resultantly they

injure themselves. So, the coach must study

kinematics specifically in high activity phases in

order to expound the patterns of muscles activity and

relate it with injury. These patterns can also be used

during the rehabilitation and fast recovery of an

injured player.

The compact model for a bowling data cannot be

materialized due to inter-subject variability and wide

range of bowling actions. The solution to this is that

instead of developing a single model for Fast bowlers,

individual profiles of players should be built, and

each player should be analysed based on his own

profile. For setting some standard for comparison a

profile of an elite can be logged but caveat will still

be inter-subject variability therefore, individual

profiles of players should be analysed.

Chances of incurring injury during bowling is

enhanced when a fatigued muscle exerts itself during

a bowling action. The above procedure can be used to

study phase-wise muscle activation pattern during a

bowling action and study the lowering of fatigue

index with repeated bowling actions. Based on

choosing a predefined threshold and the experimental

data acquired from a player, a safe number of allowed

overs and inter-over gaps can be selected for a player.

Fatigue related degradation in repeated

performance can be identified by collecting localized

fatigue data from multiple muscles during repeated

performance. Video monitoring and processing allow

computation of body kinematics in different phases of

bowling actions. Correlating this with fatiguing

pattern of individual muscles involved in the

kinematics can indicate which individual muscles

should be specifically focused for further training.

Improvement in bowling performance can be

planned better. Fast and slow fatiguing trend of a

muscle also indicate if the muscle can be invoked for

more power for better kinematics. Any lack in the

desired kinematics of a cricket bowling action can be

associated with either the lack in muscle capacity or

just the need of further training in a bowling

technique. In case of former the involved muscle

capacity shall be further improved before proceeding

with the latter.

REFERENCES

Ahamed, N.U, et al. “Surface electromyographic analysis

of the biceps brachii muscle of cricket bowlers during

bowling.” Australasian Physical & Engineering

Sciences in Medicine, vol. 37, no. 1, 2014, pp. 83-95

At, Au, and Kirsch RF. "EMG-based Prediction of Shoulder

and Elbow Kinematics in Able-bodied and Spinal Cord

Injured Individuals." IEEE Trans Rehabil Eng 8.4

(2000): 471- 80. Print.

Burden, Adrian. “Surface electromyography.”

Biomechanical Evaluation of Movement in Sport and

Exercise, edited by Carl J. Payton and Roger M.

Bartlett, Routledge, 2008, pp. 77-102.

Florimond, V. “Basics of Surface Electromyography

Applied to Psychophysiology.” eBook, Thought

Technology Ltd, 2008.

H.J Hwaang, W. H Chung, J.H Song, J.K Lim and H.S Kim,

2016. Prediction of biceps muscle fatigue and force

using electromyography signal analysis for repeated

isokinetic dumbbell curl exercise. Journal of

Mechanical Science and Technology, 30(11), Pp.

5329~5336.

Lloyd, David G., Jacqueline Anderson A., and Bruce Elliot

C. "An Upper Limb Kinematic Model for the

Examination of Cricket Bowling: A Case Study of

Mutiah Muralitharan." Journal of Sports Science 18

(2000): 975-82. Print.

Raez, M.B.I, et al. “Techniques of EMG Signal Analysis:

Detection, Processing, Classification and

Applications.” Biological Procedures Online 8 (2006):

11-35. PMC. Web. 4 Nov. 2016.

Ralf Schmidt, Catherine Disselhorst-Klug, Jiri Silny, and

Günter Rau. "A Marker-based Measurement Procedure

for Unconstrained Wrist and Elbow Motions." Journal

of Biomechanics 32.6 (June 1999): 615-21. Print

Rushda Basir Ahmad, Nadeem Ahmad Khan and

Muhammad Usama Rizwan.“A study on variation in

EMG trends under different muscular energy condition

for repeated Isokinetic Dumbbell Curl Exercise”

BioSignals 2020, December 2020.

Shorter K.,Smith N., Lauder M. and Khoury P., A

preliminary electromyographic investigation into

shoulder muscle activity in cricket seam bowling. In

Jensen R., Ebben W., Petushek C R. and Roemer, K

(Ed.), Proceedings of the XXVIII International Conf.

on Biomechanics in Sport, Northern Michigan

University, Michigan, USA, pp. 608-611, 2010.

Surface EMG-based Proﬁling and Fatigue Analysis of the Biceps Brachii Muscle of Cricket Bowlers

199