Employing classification, clusetring and neural network algorithms to analysis the anthropometric data

Abstract

Background: Anthropometry is a branch of human science that considers the physical measurement of the human body, especially size and shape. One application of anthropometrical data in ergonomics is to design working space and the development of industrialized products. Therefore, anthropometrical study on tools, equipment and workstations, which are designed based on the physical dimensions of the workers, can increase productivity.These increases the possibility of work-related injuries, particularly musculoskeletal disorders. Therefore, most countries have made great efforts to build own anthropometrics databases for various groups of citizens. Due to the lack of comprehensive databases in Iran, often information from the Western countries, especially the United States are being used. The western nations anthropometric dimensions differ with that of the Iranian population, and western manufacturers design and develop tools and machines based on their own criteria. The mismatching of these designed tools and workstations with the dimensions of the Iranian user's body can cause complications such as fatigue or other physical damage. So, Iranian researchers have suggested that at least one comprehensive, up-to-date and general national database is needed. The results of several studies have shown a correlation between the dimensions of a body. These correlations can be used to create regression equations for estimating anthropometric dimensions. Identifying, categorizing, and determining the type of relationship between anthropometric dimensions play a significant role in treatment, fitness talent, and clothing production, etc. But, the relationship between these dimensions is affected by the environmental, economic and social factors’ change. The important affecting physical factors include age, sex, race, body structure, occupation, diet, and physical activity. Among these factors, the race has a very critical role in the variation of body size. So, the differences among diverse races are more than the variations between different nations. Ethnic diversity is a crucial factor that can affect anthropometric data and its application areas. For example, this variety in body dimensions between people with different sex and race can produce many problems in product design. Therefore, of a modern method is essential to build a comprehensive anthropometric database from different races in the country. This is to consider the dificulties of extracting appropriate information from massive data and transforming them into knowledge on one hand, and time-consuming and high expense process for collecting data on the other hand, especially in the Iranian population with many races These information banks are the basis of the ergonomic design of the work environment and production products and should be updated every five years, taking into account geographical, regional and racial conditions, as well as gender. It is sticky and hard. Therefore, studying and finding relationships and linear and non-linear equations in measurement has been of great interest to researchers. And the present study was also conducted in order to investigate the relationships and patterns governing racial, gender changes and anthropometric classification and clustering of workers of 6 Iranian ethnicities in the range of 20-60. Methodology: The type of this study is descriptive-analytical with a cross-sectional approach, which was implemented in order to improve and complete the methods of measuring and evaluating anthropometric indicators and capabilities. In the first stage of the research, after the stages of data preparation and conversion into the necessary formats, visualization, clustering, and classification stages were performed using Veca software to evaluate the quantitative and qualitative status of the data. In the second step, by using sensitivity analysis, important variables were identified, and these variables were used to train the neural network in MATLAB software, and anthropometric functions were estimated, and three algorithms of the Prestpron multilayer neural network, neurophase inference system, functions the base radius was compared in terms of accuracy and predictive power. Results: In the analysis stage of the main components, 21 components were identified as the main components, and in the data mining stage, 6 clusters were calculated with the Key-Means algorithm. Three clusters for women and three clusters for men, and in the classification and decision tree stages, important relationships and laws were also identified, and in the artificial neural networks stage, neurophasic networks were stronger than the other two models in terms of accuracy and accuracy. Discussion and conclusion: Researches related to anthropometry are very necessary and important due to the complexity of products, devices, equipment and systems for users. In anthropometric researches, identification of racial and gender differences is an important factor that can be used in the data anthropometric and its fields to be effective. Discovering the knowledge needed by designers and ergonomists requires tools that, in addition to categorizing data and reducing the complexity of calculations, are also simple and understandable, and neural networks and decision trees are part of these tools. The results of this study can be used in forensic medicine and sports in addition to manufacturing and clothing industries.