----------------------------- ----------------------------
Volume 8, Issue 2 (Summer 2022)                   JMIS 2022, 8(2): 126-139 | Back to browse issues page


XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Hosseinpoor M. Predicting Gestational Diabetes Using an Intelligent Algorithm Based on Artificial Neural Network. JMIS 2022; 8 (2) :126-139
URL: http://jmis.hums.ac.ir/article-1-335-en.html
Department of Computer Engineering, Estahban Branch, Islamic Azad University, Estahban, Iran.
Full-Text [PDF 4876 kb]   (199 Downloads)     |   Abstract (HTML)  (556 Views)
Full-Text:   (174 Views)
Introduction
Diagnosing diseases is a very challenging and complicated process for experts in medical sciences; therefore, it is necessary to use appropriate data mining methods to make a correct diagnosis in medical issues. The use of algorithms and data mining techniques can provide patterns and data regarding the factors of various diseases; based on these data, doctors and medical professionals can act for the prevention of diseases. Due to the existence of a huge amount of data about patients with diabetes, it is not be possible to extract predictors of diabetes from this huge amount of data. Data mining science can discover the predictors of diseases and help doctors and medical staff in the prediction and diagnosis of diseases.
Methods
The is an applied survey study conducted in 2020 using the dataset of Mirsharif et al.’s study. The study population includes 105 female patient data registered from 2011 to 2014 in a specialized women’s medical clinic in Tehran, of which 80 were for healthy women and 25 were for women with gestational diabetes. MATLAB software was used to analyze and check the results. The used algorithm was a hybrid artificial neural network- genetic algorithm. First, unsupervised clustering operation was performed to create two clusters of diabetic and non-diabetic women using genetic algorithm on the input dataset. After performing 100 times of execution, the best extract containing the optimal clusters was obtained. After this step, the output of the genetic algorithm was considered as the input of the artificial neural network to perform the training operation. The neural network was a feedforward neural network with 20 hidden layers. In this study, by default, 75% of the datasets were used as learning samples for system learning. Moreover, to check the accuracy of the system, 25% of the datasets were used for test and validation.
Results
The results and comparisons showed the high efficiency of the proposed method in predicting gestational diabetes. In the test dataset, 25 samples were used to evaluate the method, among which 15 were for non-diabetic women and 10 were for diabetic women. In the test of the proposed method, according to the produced confusion matrix, all 12 predictions of non-diabetic women were correctly classified. Out of 13 predictions of diabetic women, 69.2% were correctly classified and 30.8% were incorrectly classified. Out of 16 samples of non-diabetic women, 75% were correctly predicted as healthy women, while 25% were diabetic. All 9 samples of diabetic women were correctly classified as diabetic. Overall, 84% of predictions were correct and 16% were incorrect (Figure 1).

The performance of the proposed method was compared with the method proposed by Mirsharif et al. on the same dataset based on two criteria. As can be seen in Table 1, the efficiency and performance of the proposed method was better than the Mirsharif et al.’s method in both MSE and ACC criteria.


This indicates that the proposed method in unsupervised mode had the desired efficiency in facing different datasets to accurately diagnose gestational diabetes.
Discussion
In this study, an intelligent and unsupervised approach based on neural-artificial network techniques and genetic algorithm was used to predict gestational diabetes. The proposed method seeks to improve the process of differentiating women with gestational diabetes from healthy women. The neural network in the proposed method discovers hidden patterns in diabetic women compared to healthy women by examining the desired dataset. Since the function of the neural network depends on how its structure is defined, a genetic algorithm was used along with the proposed neural network whose task was to learn the structure of the neural network without supervision. The results of the implementation of the proposed system showed that the performance and accuracy of the algorithm in the desired dataset, compared to the previous similar system, were acceptable and improved significantly by 93.2%. Therefore, this intelligent and unsupervised method can be used to predict gestational diabetes.

Ethical Considerations
Compliance with ethical guidelines

This study has ethical approval from Estahban Branch, Islamic Azad University (Code: 1144819916004881397187659556 ).

Funding
This research did not receive any grant from funding agencies in the public, commercial, or non-profit sectors. 

Conflicts of interest
The authors declared no conflict of interest.

Acknowledgements
The author would like to thank Estahban Branch, Islamic Azad University for their cooperation.
 

References
  1. Mierzyński R, Poniedziałek-Czajkowska E, Dłuski D, Patro-Małysza J, Kimber-Trojnar Ż, Majsterek M, et al. Nesfatin-1 and vaspin as potential novel biomarkers for the prediction and early diagnosis of gestational Diabetes Mellitus. Int J Mol Sci. 2019; 20(1):159. [PMID]
  2. Pei T, Wang W, Zhang H, Ma T, Du Y, Zhou C. Density-based clustering for data containing two types of pointsInt J Geogr Inf Sci. 2015; 29(2):175-93. [DOI:10.1080/13658816.2014.955027]
  3. Jayalakshmi T, Santhakumaran A. A novel classification method for diagnosis of Diabetes Mellitus using artificial neural networks. Paper presented at: International Conference on Data Storage and Data Engineering. 9-10 February 2010; Bangalore, India. [DOI:10.1109/DSDE.2010.58]
  4. Kumar D, Palaniswami M, Rajasegarar S, Leckie C, Bezdek JC, Havens TC. clusiVAT: A mixed visual/numerical clustering algorithm for big data. Paper presented at: IEEE International Conference on Big Data. 6-9 October 2013;  Silicon Valley, USA. [DOI:10.1109/BigData.2013.6691561]
  5. Zhao L, Ren Y. A scalable genetic algorithm for discovering comprehensible anomaly detection rules using big data in computer cluster. Paper presented at: 3rd International Conference on Systems and Informatics (ICSAI). 19-21 November 2016; Shanghai, China. ‌[DOI:10.1109/ICSAI.2016.7811048]
  6. Johnson T, Kumar Singh S. Enhanced K Strange points clustering algorithm. Paper presented at: International Conference on Emerging Information Technology and Engineering Solutions. 20-21 February 2015; Washington, DC, United States. [DOI:10.1109/EITES.2015.14]
  7. Schneider AK, Leemaqz SY, Dalton J, Verburg PE, Mol BW, Dekker GA, et al. The interaction between metabolic syndrome and physical activity, and risk for gestational Diabetes Mellitus. Acta Diabetol. 2021; 58(7):939-47. [PMID]
  8. Berikov V. Weighted ensemble of algorithms for complex data clustering. Pattern Recognit Lett. 2014; 38:99-106. [DOI:10.1016/j.patrec.2013.11.012]
  9. Khan SR, Mohan H, Liu Y, Batchuluun B, Gohil H, Al Rijjal D, et al. Correction to: The discovery of novel predictive biomarkers and early-stage pathophysiology for the transition from gestational diabetes to type 2 diabetes. Diabetologia. 2019; 62(4):730-1. [PMID]
  10. Breuing J, Pieper D, Neuhaus AL, Heß S, Lütkemeier L, Haas F, et al. Barriers and facilitating factors in the prevention of diabetes type II and gestational diabetes in vulnerable groups: protocol for a scoping review. Syst Rev. 2018; 7(1):245. [PMID]
  11. Parsons J, Sparrow K, Ismail K, Hunt K, Rogers H, Forbes A. Experiences of gestational diabetes and gestational diabetes care: A focus group and interview study. BMC Pregnancy Childbirth. 2018; 18(1):25. [PMID]
  12. Hosseinpoor M, Parvin H, Nejatian S, Rezaei V. [Detection and extraction of potential promoter/enhancer interactions in genome of cancer patients using an evolutionary multi-objective algorithm (Persian)]. J Health Biomed Inform. 2018; 5(2):304-13. [Link]
  13. Mohammadi F, Nazari S. [Exodite segmentation in diabetic patients in retinal images using fuzzy clustering (Persian)]. Papaer presented at: 3rd International Congress on Computer, Electrical and Telecommunication. 22 September 2016; Torbat Heydariyeh, Iran. [Link] 
  14. Zahedi Fard MR, Malekzadeh ـJ, Habibi S. [Medical data mining: Pattern discovery for diabetics using significant variables in diabetes (Persian)]. Paper presented at: 12th National Conference on Intelligent Systems, Bam, Iranian Intelligent Systems Association. 3-5 February 2014; Bam, Iran. [Link]
  15. Mirsharif M, Rouhani S. [Data mining approach based on neural network and decision tree methods for the early diagnosis of risk of gestational Diabetes Mellitus (Persian)]. J Health Biomed Inform. 2017; 4(1):59-68. [Link]
  16. Khoshnamvand Z, Asadi F, Khoshnamvand S, Khoshnamvand M. Diagnosis of diabetes using water wave optimization algorithm and comparison with machine learning algorithms. Paper presented at: 5th International Conference on Knowledge Based Research in Computer Engineering and Information Technology. 21 July 2017; Tehran, Iran. [Link]
  17. Sun H, Saeedi P, Karuranga S, Pinkepank M, Ogurtsova K, Duncan BB, et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022; 183:109119. [PMID]
  18. Pouya S, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9(th) edition. Diabetes Res Clin Pract. 2019; 157:107843. [DOI:10.1016/j.diabres.2019.107843]
  19. Stotz SA, McNealy K, Begay RL, DeSanto K, Manson SM, Moore KR. Multi-level diabetes prevention and treatment interventions for native people in the USA and Canada: A scoping review. Curr Diab Rep. 2021; 21(11):46. [PMID]
  20. Teufel NI, Ritenbaugh CK. Development of a primary prevention program: insight gained in the Zuni Diabetes Prevention Program. Clin Pediatr (Phila). 1998; 37(2):131-41.[PMID]
  21. Chambers RA, Rosenstock S, Neault N, Kenney A, Richards J, Begay K, et al. A home-visiting diabetes prevention and management program for American Indian Youth: The together on Diabetes trial. Diabetes Educ. 2015; 41(6):729-47. [PMID]
  22. Hosseinpoor MJ, Parvin H, Nejatian S, Rezaie V. Gene regulatory elements extraction in breast cancer by Hi-C data using a meta-heuristic method. Russ J Genet. 2019; 55(9):1152-64. [DOI:10.1134 /S1022795419090072]
  23. Kulhawy-Wibe S, King-Shier KM, Barnabe C, Manns BJ, Hemmelgarn BR, Campbell DJT. Exploring structural barriers to diabetes self-management in Alberta First Nations communities. Diabetol Metab Syndr. 2018; 10:87. [PMID]
  24. HAPO Study Cooperative Research Group. Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study: Associations with neonatal anthropometrics. Diabetes. 2009; 58(2):453-9. [PMID]
  25. Marais C, Hall DR, van Wyk L, Conradie M. Randomized cross-over trial comparing the diagnosis of gestational diabetes by oral glucose tolerance test and a designed breakfast glucose profile. Int J Gynaecol Obstet. 2018; 141(1):85-90.[PMID]

 
Type of Study: Research | Subject: Special
Received: 2021/08/16 | Accepted: 2022/04/11 | Published: 2022/07/1

Add your comments about this article : Your username or Email:
CAPTCHA

Send email to the article author


Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

© 2023 CC BY-NC 4.0 | Journal of Modern Medical Information Sciences

Designed & Developed by: Yektaweb