Nur Ikhwan Mohamad

Isotonic Muscle Fatigue Prediction for Sport Training Using Artificial Neural Network Modelling

Fatigue prediction is part of the muscle endurance analysis, which is normally based on expert experience and guided by muscle signal chart such as surface electromyography. The overall endurance training plan is normally pre-determined. Rapid updates on the training plan based on the athlete fitness state is hard to achieve in this case. Hence, this has resulted in inefficient muscle optimization during endurance training. Real time muscle monitoring and feedback can be realized through computational modelling. Many research has been done on isometrics muscle analysis. However, less attention was paid to the isotonic muscle fatigue prediction. This paper focuses on fatigue prediction using artificial neural network (ANN) model to support personalized sport training program. The ANN model aims to predict the fatigue state in isotonic muscle training. Selected feature extraction methods from time and frequency domains, i.e. the median frequency, mean frequency, mean absolute value, root mean squares, simple square integral, variance length, and waveform length were used as model predictors. The ANN model has achieved minimum mean squared error at 0.23 with overall regression value of 0.6571. The best validation performance has been attained at epoch 11. Although the result is not as good as the fatigue prediction for isometrics muscle analysis, it has shed light on the possibility of using computational modelling to predict muscle fatigue in isotonic training. Nevertheless, future work needs to be done on noise management in isotonic contractions to further improve the data quality for better prediction.

Relationship Between Muscle Architecture and Jumping Abilities Among Recreationally Active Men

The objective of this study is to determine the relationships between muscle architectures and jumping abilities. Thirty recreationally active men (mean age = 22.21 ± 1.59 years old) were recruited as study participants and underwent testing for muscle architecture and jumping abilities. Ultrasonography method was used to measure fascicle length, pennation angle and muscle thickness of biceps femoris (BF) and vastus lateralis (VL) for muscle architecture variables. Vertical jump and standing long jump were tested for jumping abilities. Pearson Correlation was used to determine the relationship between all muscle architectures and the jumping abilities tested. Results showed that the muscle thickness and pennation angle of BF and the muscle thickness of VL were shown to be positively correlated while the fascicle length of VL was shown to be negatively correlated with vertical jump performance. Besides that, muscle thickness of BF and the pennation angle of VL and BF were shown to be positively correlated with standing long jump performance. Results demonstrated the advantages of having thicker and more pennated muscle fascicles in enhancing jumping abilities.

Metabolic Cost of Continuous Body Weight Circuit Training with Aerobic-Based Exercise Interval for Muscle Strength and Endurance on Young Healthy Adults

Metabolic cost and cardiorespiratory responses produced by a body weight circuit training program were assessed for the purpose of this study. Fourteen recreationally active males aged 22.57 ± 1.70 years old with body weight of 64.53 ± 8.79 kg and height of 1.68 ± 0.04 m were recruited. A body weight circuit training program was prescribed to all participants. The body weight resistance circuit was performed in a continuous mode, with a combination of aerobic-based exercise (cycling) each body weight exercises. Portable desktop metabolic analyzer with a face mask (mouthpiece) was used to assess cardiorespiratory responses during the performance. Lower post-exercise blood pressure level was recorded, with significant differences observed between the pre- and post-exercise data (p = 0.018). Heart rate response significantly increased after performing the exercise protocol (p = 0.001). Peak oxygen consumption recorded during the performance was 24.36 ± 2.53 ml/kg/min, peak amount of oxygen extracted by the lung from the air was 16.85 ± 0.45%, maximum heart rate recorded was 156.29 ± 12.65 bpm, and respiratory frequency was 34.39 ± 5.18 l/min. Metabolic cost based on energy expenditure recorded was 468.71 ± 66.84 kcal (estimated per hour), and the Metabolic Equivalent of Task (MET) recorded was 6.86 ± 0.77. Albeit lesser total duration, continuous mode used to ensure the metabolic cost recorded was in accordance with other previous findings related to circuit-based training. Body weight circuit training with active recovery in between set is an excellent way of developing and maintaining overall physical fitness level.

Repeated Sprint Ability with Inclusion of Changing Direction among Veteran Soccer Players

This study was conducted to determine the repeated sprint ability (RSA) with the inclusion of changing direction among veteran soccer players. Twelve main players from a university veteran soccer team were recruited and were required to perform the RSA test in two conditions; i) without ball and ii) with ball. Descriptive statistics and Wilcoxon signed rank test was conducted to determine the mean score and the differences of sprint time, percentage of decrement score of sprinting (Sdec) and fatigue index (FI) between the two conditions. Both conditions demonstrated significantly drop in speed in the fifth sprint. Results showed sprint time, Sdec and FI were found to be significantly different between both conditions. Findings of this study demonstrated the important for specific training to sports for performance enhancement.

Kinematics Analysis of Dominant and Non-Dominant Lower Limb during Knee Strike among MuayThai Beginners

This study was conducted to determine and compare the kinematics of knee strike in MuayThai between dominant and non-dominant lower limb. Ten MuayThai beginners (mean age = 20 ± 1 years old) with less than one week experiences in MuayThai training were recruited and were asked to perform three trials of knee strikes for each leg (dominant and non-dominant). Joint angles and angular velocity of the movement were assessed for each trial. Repeated measure multivariate analyses of variances (MANOVA) were performed to compare the kinematics data between the dominant and non-dominant lower limb. Results showed no significant differences existed in all the joint kinematics examined between dominant and non-dominant lower limb. As the conclusion, MuayThai beginners demonstrated no differences of joint kinematics during knee strike between dominant and non-dominant lower limb.

A Comparison of Butterworth Noise Filteration Frequency for Isotonic Muscle Fatigue Analysis

In sports training, fatigue prediction using surface electromyography analysis is manually monitored by human coach. Decisions rely very much on experience. Hence, the endurance training plan for an athlete needs to be individually designed by an experienced coach. The pre-designed training plan suits the athlete fitness state in general, but not in real time. Real-time muscle monitoring and feedback help in understanding every fitness states throughout the training to optimise muscle performance. This can be realized with muscle fatigue prediction using computational modelling. Due to the higher amount of motion artefact, research in isotonic muscle fatigue prediction is very much lesser than the isometric prediction. Thus, this paper investigates the Butterworth high-pass noise filter on isotonic muscle fatigue data. Three cut-off thresholds, i.e. 5 Hz, 10 Hz, and 20 Hz, were compared using the Fuzzy c-Mean Radial Basis Function Network model. Several features of time and frequency domains, i.e. the median frequency, mean frequency, mean absolute value, root mean squares, simple square integral, variance length, and waveform length were used as model predictors. The cut-off threshold at 10 Hz is the best frequency with the lowest average mean squared error of 0.0282 and best validation performance at epoch 972.