نوع مقاله: مقاله پژوهشی

نویسندگان

چکیده

مقدمه: سکته مغزی یکی از معضلات بزرگ بهداشتی، شایع‌ترین و نیز پر عارضه‌ترین بیماری مغز و اعصاب است و همه ساله حدود 5/5 میلیون نفر را در سراسر جهان مبتلا می‌کند. مطالعه حاضر با هدف تعیین عوامل موثر در ابتلا به سکته مغزی و تعیین عملکرد الگوریتم داده کاوی شبکه عصبی در پیش‌بینی این بیماری انجام شد.مواد و روش: این مطالعه از انواع مطالعات توصیفی – تحلیلی و کاربردی بود که به ‌صورت مقطعی گذشته-نگر انجام شد. در این مطالعه از روش کریسپ به عنوان یکی از قدرتمندترین روش‌های انجام مطالعات داده کاوی استفاده شد. همچنین برای  تحلیل داده ها از نرم افزار SPSS Modeler 14.2 و الگوریتم شبکه عصبی بهره گرفته شد.یافته‌ها: با توجه به یافته‌های پژوهش، دقت کلی مدل شبکه عصبی برای پیش‌بینی سکته مغزی 7/89 درصد بود که نشان‌دهنده قدرت بالای این مدل در پیش بینی ابتلا به سکته مغزی بود. همچنین مدل مذکور ریسک فاکتور دیابت را به عنوان مهم‌ترین عامل در ابتلای به سکته مغزی پیش‌بینی کرد. علاوه براین بر اساس مدل شبکه عصبی مواردی مانند سن، تروما و تصلب شرایین نیز به عنوان عوامل ابتلا در نظر گرفته شدند.بحث و نتیجه‌گیری: بر اساس مطالعه حاضر بیماران دیابتی بیشتر در معرض ابتلا به سکته مغزی قرار داشتند و هرچه سن بیمار افزایش پیدا کند ریسک ابتلا به این بیماری افزایش پیدا می‌کند. لذا پیشنهاد می شود اقدامات موثری در جهت بیماریابی از طریق احداث کلینیک‌های تشخیص و درمان دیابت در سطح شهر و نیز انجام غربالگری های دوره‌ای انجام شود.

کلیدواژه‌ها

عنوان مقاله [English]

Presentation a Model for Prediction of Cerebrovascular Accident using Data Mining Algorithm

نویسندگان [English]

  • Yousef Mehdipour
  • Saeid Ebrahimi
  • Afsaneh Karimi
  • Jahanpour Alipour
  • Mohammad Khammarnia
  • Fatemeh Siasar

چکیده [English]

Background and Objectives: Cerebrovascular Accident (CVA) is a major health problem and the most common neurological disease affecting 5.5 million individuals around the world annually. The present study aimed to assess the effective factors in CVA and determine the performance of data mining algorithm in predicting the disease.
Methods: This retrospective, cross-sectional, descriptive-analytical study used Crisp as one of the most powerful data mining techniques. The data were analyzed using SPSS Modeler 14.2 and neural network algorithm.
Results: According to the findings, the overall accuracy of the neural network model was 89.7%, which reflects the strength of this model in predicting the risk of CVA. Indeed, this model predicted the risk factors of diabetes as the most important factor in the risk of CVA. Age, trauma, and atherosclerosis were also determined as other risk factors.
Conclusion: According to this study, diabetic patients were more prone to CVA. Additionally, the risk of this disease increased with age.Therefore, effective measures are recommended to be taken for diagnosis and treatment of diabetes surveillance by establishing clinics and performing periodic screening.

کلیدواژه‌ها [English]

  • Data mining
  • Neural networks
  • Cerebrovascular accident
  • Risk factors
  • Prediction of the probability risk

  1. Ahmadi Ahangar A, Sanaat A, Saghebi R. “Assessment of Risk Factors in Stroke Patients”. Journal of Babol University of Medical Sciences. 2005; 7(3): 55-60. [In Persian [
  2. Iranmanesh F, Vaziri Nejad R, Gadari F, Rajabpoor N. “Study of Relationship between Prevalence of Post-Stroke Depression and Stroke Risk Factors”. Journal of Fasa University of Medical Sciences. 2012; 2(6): 266-70. [In Persian]
  3. Joozi M. “The Effect of Massage Therapy on Stroke Patients”. Journal of Medical Sciences, Islamic Azad University. 2009; 19(4): 256-61. [In Persian]
  4. Mazaheri T, Fallahpoor M, Karimlou M, Hoseini A. “Evaluating the Reliability of Farsi version of Stroke Impact Scale (SIS) among Patients with Stroke”. Research in Rehabilitation Sciences. 2011; 7(4): 549-56. [In Persian]
  5. Dorrotaj D. “Investigated 90 Cases of Stroke in Young People Admitted to Imam Khomeini Hospital (1993-1998) [dissertation]”. [Iran]: Tehran University of Medical Sciences and Health Services; 2001. P: 92.
  6. Baghbanian M, Ramezani M, Abedini M, Yazdani Charati J. “Difference between Risk Factors of Common Ischemic Brain Strokes on Anterior and Posterior Circulation”. Journal of Mazandaran University of Medical Sciences. 2013; 23(101): 49-54. [In Persian]
  7. Saadatnia M, Feiz M, Ziaei E, Hamzeh M, Ghorbani E, Hassanzadeh Keshteli A. “Lipid Profile in Patients with Ischemic and Hemorrhagic Stroke”. Journal of Isfahan Medical School. 2011; 29(129): 172-9. [In Persian]
  8. Amini L. “Manage and Control the Stroke with Data Mining in Al-Zahra Hospital in Isfahan and Ghaem Hospital in Mashhad [dissertation]”. [Iran]: PNU Tehran; 2012: 90.
  9. Sarafi Nejad A, Saeid A, Mohammad Rose I, Rohanimanesh A. “Modeling a Data Mining Decision Tree and Propose a New Model for the Diagnosis of Skin Cancer by Immunohistochemical Staining Methods”. Journal of Health and Biomedical Informatics. 2014; 1(1): 54-62. [In Persian]
  10. Azizi A, Zarei J, Nabovati E, Vakili Arki H, Abbasi E, Razavi A. “Determining of the Factors Affecting Mortality in Burn Patients Using a Decision Tree Data Mining Algorithm”. Journal of Health Information Management. 2013; 16(54): 34-45. [In Persian]
  11. Moghaddassi H, Hoseini A, Asadi F, Jahanbakhsh M. “Application of Data Mining in Health”. Journal of Health Information Management. 2012; 9(2): 297-304. [In Persian]
  12. Ahmadi Ahangar A, Khafri S, Mostafazadeh M, Akbarian M, Motavalli M. “Risk Factors for Strokes in Patients 15-45 Years Old Admitted at Ayatollah Roohani Hospital in Babol, Iran (2009-2010)”. Journal of Babol University of Medical Sciences. 2013; 15(6): 95-103. [In Persian]
  13. Ghaffari H, Amini L, Farahi A. “Stroke Diagnosis Using Data Mining Techniques”. The Fifth Iran Data Mining Conference. 2011. [In Persian]
  14. Hoseini A, Bazrafshan H. “The Role of Risk Factors for Stroke”. Journal of Gorgan University of Medical Sciences. 2001; 3(8): 76-80. [In Persian]
  15. Yeh D.Y, Cheng Ch.H, Chen Y.W. “A Predictive Model for Cerebro Vascular Disease Using Data Mining”. Expert Systems with Applications. 2011: 8970-7.
  16. Jianxin C, Yanwei X, Guangcheng X, Jianqiang Y, Dongbin Z. [A Comparison of Four Data Mining Models: Bayes, Neural Network, SVM and decision Trees in Identifying Syndromes in Coronary Heart Disease]. Springer- Verlag Berlin Heidelberg 2007:1274–1279
  17. Safdari R, Ghazi saeedi M, Gharooni M, Nasiri M, Arji G. Comparing performance of decision tree and neural network in predicting myocardial infarction. JPSR 2014; 3(2):26-37.
  18. Raeesi A, Ebrahimi S, Irfan Nia L, Arji G, Askani M. An investigation of data mining techniques of the performance of a decision tree algorithm for predicting causes of traumatic brain injuries in Khatamolanbya Hospital in Zahdan city, 2012 to 2013. J Health Man & Info. 2014;1(2):28-30.
  19. Sepehri MM, Rahnama P, Shadpour P, Teimourpour B. A data mining based model for selecting type of treatment for kidney stone patients. Tehran University Medical Journal 2009; 67(6): 421-7. [In Persian].