A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm
Hyperinsulinemia is a condition with extremely high levels of insulin in the blood. Various factors can lead to hyperinsulinemia in children and adolescents. Puberty is a period of significant change in children and adolescents. They do not have to have explicit symptoms for prediabetes, and certain...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Healthcare |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-9032/10/5/921 |
| _version_ | 1797499474798444544 |
|---|---|
| author | Igor Lukić Nevena Ranković Nikola Savić Dragica Ranković Željko Popov Ana Vujić Nevena Folić |
| author_facet | Igor Lukić Nevena Ranković Nikola Savić Dragica Ranković Željko Popov Ana Vujić Nevena Folić |
| author_sort | Igor Lukić |
| collection | DOAJ |
| description | Hyperinsulinemia is a condition with extremely high levels of insulin in the blood. Various factors can lead to hyperinsulinemia in children and adolescents. Puberty is a period of significant change in children and adolescents. They do not have to have explicit symptoms for prediabetes, and certain health indicators may indicate a risk of developing this problem. The scientific study is designed as a cross-sectional study. In total, 674 children and adolescents of school age from 12 to 17 years old participated in the research. They received a recommendation from a pediatrician to do an OGTT (Oral Glucose Tolerance test) with insulinemia at a regular systematic examination. In addition to factor analysis, the study of the influence of individual factors was tested using RBF (Radial Basis Function) and SVM (Support Vector Machine) algorithm. The obtained results indicated statistically significant differences in the values of the monitored variables between the experimental and control groups. The obtained results showed that the number of adolescents at risk is increasing, and, in the presented research, it was 17.4%. Factor analysis and verification of the SVM algorithm changed the percentage of each risk factor. In addition, unlike previous research, three groups of children and adolescents at low, medium, and high risk were identified. The degree of risk can be of great diagnostic value for adopting corrective measures to prevent this problem and developing potential complications, primarily type 2 diabetes mellitus, cardiovascular disease, and other mass non-communicable diseases. The SVM algorithm is expected to determine the most accurate and reliable influence of risk factors. Using factor analysis and verification using the SVM algorithm, they significantly indicate an accurate, precise, and timely identification of children and adolescents at risk of hyperinsulinemia, which is of great importance for improving their health potential, and the health of society as a whole. |
| first_indexed | 2024-03-10T03:47:53Z |
| format | Article |
| id | doaj.art-f8ecee753c6344f0b6b5fe0f2056f726 |
| institution | Directory Open Access Journal |
| issn | 2227-9032 |
| language | English |
| last_indexed | 2024-03-10T03:47:53Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Healthcare |
| spelling | doaj.art-f8ecee753c6344f0b6b5fe0f2056f7262023-11-23T11:15:08ZengMDPI AGHealthcare2227-90322022-05-0110592110.3390/healthcare10050921A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning AlgorithmIgor Lukić0Nevena Ranković1Nikola Savić2Dragica Ranković3Željko Popov4Ana Vujić5Nevena Folić6Faculty of Medical Sciences, University of Kragujevac, 34000 Kragujevac, SerbiaDepartment of Computer Science, School of Computing, Union University, 11000 Belgrade, SerbiaFaculty of Health and Business Studies, Singidunum University, 14000 Valjevo, SerbiaDepartment of Mathematics and Statistics, Faculty of Applied Sciences in Nis, Union University “Nikola Tesla”, 18000 Nis, SerbiaSchool Center with Dormitory “Dositej Obradović”, 24000 Subotica, SerbiaFaculty of Medical Sciences, University of Kragujevac, 34000 Kragujevac, SerbiaFaculty of Medical Sciences, University of Kragujevac, 34000 Kragujevac, SerbiaHyperinsulinemia is a condition with extremely high levels of insulin in the blood. Various factors can lead to hyperinsulinemia in children and adolescents. Puberty is a period of significant change in children and adolescents. They do not have to have explicit symptoms for prediabetes, and certain health indicators may indicate a risk of developing this problem. The scientific study is designed as a cross-sectional study. In total, 674 children and adolescents of school age from 12 to 17 years old participated in the research. They received a recommendation from a pediatrician to do an OGTT (Oral Glucose Tolerance test) with insulinemia at a regular systematic examination. In addition to factor analysis, the study of the influence of individual factors was tested using RBF (Radial Basis Function) and SVM (Support Vector Machine) algorithm. The obtained results indicated statistically significant differences in the values of the monitored variables between the experimental and control groups. The obtained results showed that the number of adolescents at risk is increasing, and, in the presented research, it was 17.4%. Factor analysis and verification of the SVM algorithm changed the percentage of each risk factor. In addition, unlike previous research, three groups of children and adolescents at low, medium, and high risk were identified. The degree of risk can be of great diagnostic value for adopting corrective measures to prevent this problem and developing potential complications, primarily type 2 diabetes mellitus, cardiovascular disease, and other mass non-communicable diseases. The SVM algorithm is expected to determine the most accurate and reliable influence of risk factors. Using factor analysis and verification using the SVM algorithm, they significantly indicate an accurate, precise, and timely identification of children and adolescents at risk of hyperinsulinemia, which is of great importance for improving their health potential, and the health of society as a whole.https://www.mdpi.com/2227-9032/10/5/921hyperinsulinemiainsulinglucosechildren and adolescentsrisk factorsSVM |
| spellingShingle | Igor Lukić Nevena Ranković Nikola Savić Dragica Ranković Željko Popov Ana Vujić Nevena Folić A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm Healthcare hyperinsulinemia insulin glucose children and adolescents risk factors SVM |
| title | A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm |
| title_full | A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm |
| title_fullStr | A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm |
| title_full_unstemmed | A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm |
| title_short | A Novel Approach of Determining the Risks for the Development of Hyperinsulinemia in the Children and Adolescent Population Using Radial Basis Function and Support Vector Machine Learning Algorithm |
| title_sort | novel approach of determining the risks for the development of hyperinsulinemia in the children and adolescent population using radial basis function and support vector machine learning algorithm |
| topic | hyperinsulinemia insulin glucose children and adolescents risk factors SVM |
| url | https://www.mdpi.com/2227-9032/10/5/921 |
| work_keys_str_mv | AT igorlukic anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT nevenarankovic anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT nikolasavic anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT dragicarankovic anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT zeljkopopov anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT anavujic anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT nevenafolic anovelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT igorlukic novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT nevenarankovic novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT nikolasavic novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT dragicarankovic novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT zeljkopopov novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT anavujic novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm AT nevenafolic novelapproachofdeterminingtherisksforthedevelopmentofhyperinsulinemiainthechildrenandadolescentpopulationusingradialbasisfunctionandsupportvectormachinelearningalgorithm |