Data Analysis And Tools And Application Assignment Sample

Introduction Of Data Analysis And Tools And Application

Data Analysis And Tools And Application On A Multi Camera And Multimodal Dataset For Posture And Gait Analysis

The posture and gait analysis play a major role in different fields, and those fields are sports science, healthcare, and ergonomics. Through the human movement patterns understanding and detecting normal posture deviations and gait can provide major information regarding the major injury risk, performance, and physical health of an individual. For the advanced development, and research development in this regard the dataset of multimodal, and multi-camera is required to analyze and capture human gait, and posture in a manner that is comprehensive.

Feeling overwhelmed with your assignments? Take a breather and let New Assignment Help ease your burden! Our professional team provides unmatched assignment writing services in the UK, guaranteeing academic success. Explore our free assignment samples for inspiration and guidance.

Group project motivation / problems

The difficulties and constraints associated with the analysis of the available gait and posture datasets have served as the primary driving force behind this group endeavor (Gu et al. 2020). On the single camera recordings many present datasets focus on majorly, or they use limited modalities' sensor through which richness, and accuracy are restricted of the data that is captured (Yang et al. 2022). In addition, that the diversity of lack is also noticed in the subjects through which generally is limited of the result from the analysis.

Group project aim

High quality, and comprehensive dataset creation for the gait and posture analysis is the major aim of this group project (Albert et al. 2020). Through the multiple synchronized cameras usage, and through the incorporation of the various modalities sensor the movement of the human from various perspective is aimed to be captured, through which more detailed, and accurate analysis is enabled of the gait, and posture (Moro et al. 2022). The design of the dataset will be done for addressing the existing dataset limitation, and valuable resources is provided for practitioners, and researchers in the field is provided through it.

Objectives for the analysis

Various major objectives are present that are focused on through the dataset analysis.

Analyses of the posture

• To analyzing, and extracting body segment orientation, and joint angles for the posture quality assessment.
• To normal patterns of posture identification, and making the normal alignment deviations.
• To get balance during various activities, and stability regarding posture.

Gait analysis

• To analyze, and capture gait parameters that include stride duration, cadence, and length.
• To classify and detect abnormal patterns of gait, named asymmetrical gait, shuffling, and limping.
• To assess the symmetry of gait, and coordination between various segments of body.

Analysis of the biochemical

• To determine the estimated moments, and joint forces for the evaluation of the various joints mechanical stress at the time of movement.
• To analyze the expenditure regarding energy, and efficiency of various patterns of gait.
• To investigate the relationship between major musculoskeletal disorder, gait, and posture.

Through these objectives achievements the subsequent analysis, and dataset will make the contribution to the development of more effective, and accurate gait, and posture assessment methods that leads to improved interventions of healthcare, ergonomic design, and optimization of sports performance.

Methodology

Data collection

The data set has been gathered with the help online website. The dataset has been retrieved with the help of the internet.

Data preparation

The data that is collected on that data the process of data preparation that is going to be used is the descriptive statistics, regression, and correlation.

The variables that have been collected in this regard are gender, Age (year), Body mass (kg), Body height (m), Hip Height (m), Shoe length, shoulder height, shoulder width, elbow span, wrist span, Hip Width, Knee height, and Ankle Height. There are 14 variables are present in the dataset. The dataset that is collected in that dataset there are no missing value is present, that is the reason pre-processing of data is not needed in this regard.

Data analysis

To get results the analysis, approach, method, techniques that are used in this research on the data set are descriptive statistics, correlation, and regression.

Figure 1: Descriptive analysis

(Source: In MS Excel self created)

The image that is presented in this image it is clearly visible that various statistical for every variable is presented, and those statistical observations are done here. At first the observation is done regarding the age. 25.36 years is the approximate mean age, with a 2.31 standard deviation (Yadav et al. 2022). ,From 22 to 30 the ages range sample is present in this regard. The positive skewed slightly is noticed in the data, and 0.50 is the skewness value that is indicated.

The second observation is done regarding the body mass. 69.73 kg is the mean body mass, with 11.45 of standard deviation. From 50.5 kg, to 89 kg the range of body mass is present. The positive skewed nature is noticed in the data, with a 0.11 value of skewness.

1.72 meters is the approximate mean body height with a 0.10 standard deviation. From 1.51 meters to 1.85 meters is the body height range in the sample (Khokhlova et al. 2019). The negative skewed nature is slightly noticed in the data, that by a -0.65 skewness value is indicated.

The 0.96 meters is the mean hip height, with a 0.05 standard deviation. In the sample from 0.86 meters to 1.07 meters the range of hip height in the sample is noticed. The positive skewed nature is noticed in the data, with a value of skewness is 0.39.

2.33 is the mean shoe length with a 7.68 standard deviation. The shoe length range in the sample is from 0.25 to 29. The positively skewed nature is noticed in the dataset heavily with a 3.74 value of skewness.

140 meters is the mean shoulder height, with a 0.09 standard deviation. From 1.18 meters, to 1.54 is the shoulder widths range. The slightly positive skewed nature is noticed in the data with a -0.84 value of skewness.

0.32 meters is the mean shoulder width approximately with a 0.04 standard deviation. In the sample the shoulder width range is from 0.25 meters to 0.39 meters. The slightly positive skewed nature is noticed in the data with the 0.27 value of skewness.

0.90 meters span is for the mean elbow with a 0.08 standard deviation. The elbow spans range in the sample is 0.77 to 1.06 meters. The positive skewed nature is noticed in the data, and 0.21 is the value of skewness in this regard.

1.36 meters is the mean wrist span value with a 0.09 standard deviation. From 1.18 meters to 1.51 meters is the range of wrist span in the sample. The slight negative skewed nature is noticed, and -0.01 is the skewness value here.

1.72 meters is the arm span mean, with a 0.11 standard deviation (Yadav et al. 2021). From 1.51 meters to 1.88 meters is the arm span range in the sample. Slight negative skewed nature is noticed in the data with a value of the skewness of -0.40.

0.28 is the mean hip width approximately with a 0.02 standard deviation. From 0.24 to 0.32 is the hip widths range in the sample. Slight negative skewed nature is noticed in the data, and the skewness value in this regard is -0.12.

0.49 is the mean knee height with a standard deviation of 0.04. From 0.43 to 0.53 is the knee height range in the sample. In the data slightly negative skewed nature is noticed, with a value of the skewness is -0.72.

0.15 meters is the ankle height mean with the 0.21 standard deviation. From 0.08 to 0.9 is the ankle height range in the sample. The data in this regard is skewed positively with a 3.73 value of the skewness.

Figure 2: Correlation analysis

(Source: In MS Excel self created)

The image that is presented in that image it is shown that in the provided data the presence of correlation matrix is noticed, and in this regard the relationship is shown between various variables. Every column and row represents the particular variable, and in the cells the values represents the direction, and strength of the relationship among the variables.

For the analysis of the data in the matrix the values are needed to be observed to understand the relationship among the variables.

The positive relationships are noticed when in the matrix values are present positive it indicates between the variables a positive relationship is present (Nambiar et al. 2019). An example in this regard between the body mass, hip height, body height, length of the shoe, height of the shoulder, span of the elbow, width of the shoulder, arm span, wrists span, knee height, hip width, and length of the ankle positive relationship is noticed.

The negative relationship is noticed when in the matrix values are present negative among the variables (Guo et al. 2023). An example regarding it is the negative relationship is present between the variables that are mentioned previously.

The strengths of the relationship is judged through the values magnitudes of the relationship between variables (Gavrilova et al. 2021). Closer to -1 or 1 the values that are present indicates a relationship that is stronger, and closer to 0 the values those are present indicates the relationship that is weaker.

In the matrix the diagonal value represents each variable relationship with itself that are correlated perfectly with it. [Referred to appendix 1]

The image that is presented it shows the regression analysis. The important information regarding it is presented below.

This image data contains observations that are numbered 14, and the independent variables are present in this regard are 12. Regarding the ANOVA, regression statistics, standard errors, p-values, t-stat the information is provided in the output. 0.9995 is the value of the “Multiple R”, and it indicates that between the independent variables, and dependent variables a positive, and strong correlation is present. 0.9991 is the value of the “R square”, and it meant that 99.9% of the variance is the approximate value in the dependent variable, and by the independent variable it can be explained (Rodrigues et al. 2019). The sum of square, “F-values”, “P-values”, “mean square” are shown in the table of Anova for the residual terms, and regression with the sum of square total. 88.2046 is the regression of the F value, and it is very high also, and 0.083 is the “P-value” that is associated, and it is the lower value also. It suggests that the statistically significant nature is noticed in the model of regression.

The regression coefficients, and intercept estimates are provided through the table of coefficient with the “P-values”, “t-stat”, and “standard errors”. 36.39 is the intercept volume, and from -143.39 to 137.25 regarding every independent variable coefficient is present. The relatively small nature is noticed regarding the “standard errors”, and it indicates that the precise nature is noticed relatively regarding the estimates (Chatzitofis et al. 2020). For the test of the null hypothesis the “P-values”, and “t-stats” can be used, and in this regard equal to zero the value of the coefficient is present. If below the 0.05 “P-value” the null hypothesis can be rejected at the significance level of 5%.

Project management

Project plan

Project collaboration

In the form of a group I have worked with a greater team spirit. The discussion board, and MS team, are used for the shared document. There are various differences are present in the group and according to that the work is conducted among the group, through which all the members of the group will be able to collaborate with each other for the completion of the project.

Learning reflection

The learning outcomes those are achieved through this research are enhancement of understanding of the gait, and posture analysis, accurate algorithm development, improvement of the monitoring tools, and diagnostic, and gathering population specific differences insight. The issues that are faced in this regard is regarding the collection of data, data integration, and synchronization, and data annotation, and labeling of data. In the future the things that will go differently are regarding the data sharing, and collaboration, measurement protocol standardization, deep learning techniques incorporation, real world validation, and application.

Reference list

Journal

• Albert, J.A., Owolabi, V., Gebel, A., Brahms, C.M., Granacher, U. and Arnrich, B., 2020. Evaluation of the pose tracking performance of the azure kinect and kinect v2 for gait analysis in comparison with a gold standard: A pilot study.Sensors,20(18), p.5104.
• Chatzitofis, A., Saroglou, L., Boutis, P., Drakoulis, P., Zioulis, N., Subramanyam, S., Kevelham, B., Charbonnier, C., Cesar, P., Zarpalas, D. and Kollias, S., 2020. Human4d: A human-centric multimodal dataset for motions and immersive media.IEEE Access,8, pp.176241-176262.
• Gavrilova, M.L., Ahmed, F., Bari, A.H., Liu, R., Liu, T., Maret, Y., Sieu, B.K. and Sudhakar, T., 2021. Multi-modal motion-capture-based biometric systems for emergency response and patient rehabilitation. InResearch Anthology on Rehabilitation Practices and Therapy(pp. 653-678). IGI global.
• Gu, X., Guo, Y., Deligianni, F., Lo, B. and Yang, G.Z., 2020. Cross-subject and cross-modal transfer for generalized abnormal gait pattern recognition.IEEE Transactions on Neural Networks and Learning Systems,32(2), pp.546-560.
• Guo, Z., Hou, Y., Wang, P., Gao, Z., Xu, M. and Li, W., 2023. FT-HID: a large-scale RGB-D dataset for first-and third-person human interaction analysis.Neural Computing and Applications,35(2), pp.2007-2024.
• Khokhlova, M., Migniot, C., Morozov, A., Sushkova, O. and Dipanda, A., 2019. Normal and pathological gait classification LSTM model.Artificial intelligence in medicine,94, pp.54-66.
• Moro, M., Marchesi, G., Hesse, F., Odone, F. and Casadio, M., 2022. Markerless vs. marker-based gait analysis: A proof of concept study.Sensors,22(5), p.2011.
• Nambiar, A., Bernardino, A. and Nascimento, J.C., 2019. Gait-based person re-identification: A survey.ACM Computing Surveys (CSUR),52(2), pp.1-34.
• Rodrigues, T.B., Catháin, C.Ó., Devine, D., Moran, K., O'Connor, N.E. and Murray, N., 2019, June. An evaluation of a 3D multimodal marker-less motion analysis system. InProceedings of the 10th ACM multimedia systems Conference(pp. 213-221).
• Yadav, S.K., Luthra, A., Tiwari, K., Pandey, H.M. and Akbar, S.A., 2022. ARFDNet: An efficient activity recognition & fall detection system using latent feature pooling.Knowledge-Based Systems,239, p.107948.
• Yadav, S.K., Tiwari, K., Pandey, H.M. and Akbar, S.A., 2021. A review of multimodal human activity recognition with special emphasis on classification, applications, challenges and future directions.Knowledge-Based Systems,223, p.106970.
• Yang, C., Yang, Z., Li, W. and See, J., 2022. FatigueView: A Multi-Camera Video Dataset for Vision-Based Drowsiness Detection.IEEE Transactions on Intelligent Transportation Systems.