بررسی و طبقه‌بندی مشخصه‌های کاربردی در تکنیک شبکه‌ی عصبی به منظور تشخیص بیماری‌های ایسکمیک قلبی: مروری سیستماتیک

نوع مقاله : مقاله مروری

نویسندگان

1 دکتری تخصصی انفورماتیک پزشکی، گروه علوم کامپیوتر، مؤسسه‌ی آموزش عالی غیر دولتی فاران، تهران، ایران

2 دانشجوی کارشناسی ارشد، گروه انفورماتیک پزشکی، دانشکده‌ی مدیریت و اطلاع‌رسانی، دانشگاه علوم پزشکی ایران، تهران، ایران

3 مربی ،گروه مدیریت و فن‌آوری اطلاعات سلامت، دانشکده مدیریت و اطلاع‌رسانی پزشکی، دانشگاه علوم پزشکی اصفهان، اصفهان، ایران

4 دانشجوی پزشکی، دانشکده‌ی پزشکی، دانشگاه علوم پزشکی جندی‌شاپور، اهواز، ایران

چکیده

مقدمه: امروزه، شیوع بیماری‌های ایسکمیک قلبی به عوارض جبران ناپذیری نظیر مرگ بیماران منجر می‌شود. تشخیص دیر هنگام این گونه بیماری‌ها و راه‌کار تهاجمی تشخیصی آن‌ها، سبب شده است محققین به منظور تشخیص به موقع بیماری، نسبت به تهیه‌ی سیستم تصمیم‌یار تشخیصی مبتنی بر تکنیک شبکه‌ی عصبی ضمن به کارگیری حداقل داده‌ها اقدام نمایند. در این راستا، انتخاب حداقل مشخصه‌های مفید برای طراحی ساختار شبکه‌ی عصبی از اهمیت ویژه‌ای برخوردار است و زمینه‌ی دست‌یابی به بیشترین دقت در اخذ نتایج تصمیم را فراهم می‌آورد.روش‌ها: ابتدا مقالات مرتبط با «تشخیص ایسکمیک قلبی با استفاده از شبکه‌های عصبی مصنوعی» از پایگاه داده‌های معتبر و با استفاده از کلید واژه‌های حساس استخراج گردیدند. سپس، تحلیل و طبقه‌بندی محتوا به روش‌های علمی انجام شد.یافته‌ها: طراحی ساختار شبکه‌ی عصبی با استفاده از مشخصه‌های گوناگون قابل اخذ از داده‌های دموگرافیک، تاریخچه‌ی پزشکی، علایم و نشانه‌های بیماری به ویژه آزمایش‌های پاراکلینیک انجام می‌پذیرد. در این بین، مشخصه‌های روش الکتروکاردیوگرام که در گروه آزمایش‌های پاراکلینیک قرار داشتند، به افزایش چشم‌گیر کارایی شبکه‌ی عصبی منجر شدند.نتیجه‌گیری: بهره‌برداری از این گونه سیستم‌های تصمیم‌یار تشخیصی در محیط‌های عملی، به ضریب اطمینان بالای آن‌ها و برخورداری از مقبولیت پزشکان وابسته است. از این رو، لحاظ کردن مواردی نظیر ارتقای میزان بلوغ طراحی ساختار شبکه‌ی عصبی که به انتخاب حداقل مشخصه‌های بهینه وابسته است و ایجاد زیر ساخت‌های لازم جهت ورود داده‌های واقعی، صحیح و در جریان بیماران به این سیستم‌ها راه‌گشا می‌باشند.

کلیدواژه‌ها


عنوان مقاله [English]

Survey and Classification of Functional Characteristics in Neural Network Technique for the Diagnosis of Ischemic Heart Disease: A Systematic Review

نویسندگان [English]

  • Shirin Ayani 1
  • Khadijeh Moulaei 2
  • Maryam Jahanbakhsh 3
  • Reza Moulaei 4
1 PhD in Medical Informatics, Department of Computer Sciences, Faran (Mehr Danesh) Non-governmental Institute of Virtual Higher Education, Tehran, Iran
2 MSc Student, Department of Medical Informatics, School of Health Managment and Information Siences, Iran University of Medical Sciences, Tehran, Iran
3 Instructor, Department of Management and Health Information Technology, School of Management and Medical Information, Isfahan University of Medical Sciences, Isfahan, Iran
4 Student of Medicine, School of Medicine, Jundishapur University of Medical Sciences, Ahvaz, Iran
چکیده [English]

Background: Nowadays, the prevalence of ischemic heart diseases (IHDs) leads to destructive effects such as patient death. Late diagnosis of such diseases as well as their invasive diagnostic approaches made researchers provide a decision support system based on neural network techniques, while using minimum data set for timely diagnosis. In this regard, selecting minimum useful features is significant for designing neural network structure and it paves the way to attain maximum accuracy in obtaining the results.Methods: In this systematic review, valid databases using sensitive keywords were initially searched out to find articles related to "diagnosing the ischemic heart disease using artificial neural networks" and afterwards, scientific methods were used to analyze and classify the content.Findings: Researchers applied various extractable features from demographic data, medical history, signs and symptoms, and paraclinical examinations, to design the neural network structure. Among them, the features obtained from electrocardiographic test, embedded in paraclinical examinations, had led to a remarkable increase of efficiency in neural network.Conclusion: Utilizing such diagnostic decision support systems in practical environments depends on their high confidence coefficient and physicians’ acceptability. Therefore, it can be useful to improve maturity in the design of the neural network structure depending on the choice of the minimum optimal features, and to create required infrastructures to input patients’ real, accurate, and flowing data in these systems.

کلیدواژه‌ها [English]

  • Decision support systems
  • Ischemic Heart Disease
  • Neural Network
  • Electrocardiogram
  1. Benjamin EJ, Blaha MJ, Chiuve SE, Cushman M, Das SR, Deo R, et al. Heart disease and stroke statistics-2017 update: A report from the American Heart Association. Circulation 2017; 135(10): e146-e603.
  2. World Health Organization. About cardiovascular diseases [Online]. [cited 2017]; Available from: URL: http://www.who.int/cardiovascular_diseases/about_cvd/en/
  3. World Health Organization. The top 10 causes of death [Online]. [cited 2017 Jan]; Available from: URL: http://www.who.int/mediacentre/factsheets/fs310/en/
  4. Gaziano TA, Bitton A, Anand S, Abrahams-Gessel S, Murphy A. Growing epidemic of coronary heart disease in low- and middle-income countries. Curr Probl Cardiol 2010; 35(2): 72-115.
  5. Fauci A, Braunwald E, Kasper D, Hauser S, Longo D, Jameson J, et al. Harrison's principles of internal medicine. 17th ed. New York, NY: McGraw-Hill; 2008. p. 71-3.
  6. Kobashigawa J. Coronary computed tomography angiography: Is it time to replace the conventional coronary angiogram in heart transplant patients? J Am Coll Cardiol 2014; 63(19): 2005-6.
  7. Tasoulis DK, Vladutu L, Plagianakos VP, Bezerianos A, Vrahatis MN. Online neural network training for automatic ischemia episode detection. Berlin, Heidelberg, Germany: Springer Berlin Heidelberg; 2004 p. 1062-8.
  8. Kumar A, Maheshwari U. An expert system for identifying cardio vascular disease. International Journal of Innovative Research in Computer and Communication Engineering 2014; 2(3): 144-9.
  9. Dehnavi AR, Farahabadi I, Rabbani H, Farahabadi A, Mahjoob MP, Dehnavi NR. Detection and classification of cardiac ischemia using vectorcardiogram signal via neural network. J Res Med Sci 2011; 16(2): 136-42.
  10. Mehdi B, Khan T, Ali ZA. Artificial neural network based electrocardiography analyzer. Proceedings of the 3rd International Conference on Computer, Control and Communication; 2013 Sep 25-26; Karachi, Pakistan.
  11. Niranjana Murthy HS, Meenakshi M. Efficient algorithm for early detection of myocardial ischemia using pca based features. Indian J Sci Technol 2016; 9(39): 1-13.
  12. Thamarai Selvi S, Arumugam S, Ganesan L. BIONET: An artificial neural network model for diagnosis of diseases. Pattern Recognit Lett 2000; 21(8): 721-40.
  13. Aminsharifi A, Irani D, Pooyesh S, Parvin H, Dehghani S, Yousofi K, et al. Artificial neural network system to predict the postoperative outcome of percutaneous nephrolithotomy. J Endourol 2017; 31(5): 461-7.
  14. Khan IY, Zope PH, Suralkar SR. Importance of artificial neural network in medical diagnosis disease like acute nephritis disease and heart disease. International Journal of Engineering Science and Innovative Technology 2013; 2(2): 210-7.
  15. Peng H, Long F, Ding C. Feature selection based on mutual information: Criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 2005; 27(8): 1226-38.
  16. Nguyen DT, Kim KW, Hong HG, Koo JH, Kim MC, Park KR. Gender recognition from human-body images using visible-light and thermal camera videos based on a convolutional neural network for image feature extraction. Sensors (Basel) 2017; 17(3).
  17. Guyon I, Elisseeff A. An Introduction to variable and feature selection. J Mach Learn Res 2003; 3: 1157-82.
  18. Suzuki K. Artificial neural networks - methodological advances and biomedical applications. Rijeka, Croatia: InTech; 2011.
  19. Khandaker MH, Espinosa RE, Nishimura RA, Sinak LJ, Hayes SN, Melduni RM, et al. Pericardial disease: Diagnosis and management. Mayo Clin Proc 2010; 85(6): 572-93.
  20. Obaloluwa Olaniyi E, Kayode Oyedotun O. Heart diseases diagnosis using Neural Networks arbitration. I J Intelligent Systems and Applications 2015; 12: 75-8.
  21. Beer S, Rosler KM, Hess CW. Diagnostic value of paraclinical tests in multiple sclerosis: Relative sensitivities and specificities for reclassification according to the Poser committee criteria. J Neurol Neurosurg Psychiatry 1995; 59(2): 152-9.
  22. Quality guidelines for endodontic treatment: Consensus report of the European Society of Endodontology. Int Endod J 2006; 39(12): 921-30.
  23. Smith B, Ceusters W, Goldberg LJ, Ohrbach RK. Towards an ontology of pain and of pain-related phenomena. Proceedings of the Conference on Ontology and Analytical Metaphysics; 2011 Feb 24-25; Tokyo, Japan. Tokyo, Japan: Keio University Press; p. 1-6.
  24. Sign. Stedman’s Medical Dictionary. Wolters Kluwer Health. [Online]. [cited 2013 Dec 12]; Available from: URL: http://www.medilexicon.com/dictionary/81800
  25. Campbell EW, Lynn CK. The physical examination. In: Walker HK, Hall WD, Hurst JW, editors. Clinical methods: The history, physical, and laboratory examinations. 3rd ed. Boston, MA: Butterworths; 1990.
  26. Sato Y, Fujiwara H, Takatsu Y. Biochemical markers in heart failure. J Cardiol 2012; 59(1): 1-7.
  27. Martis RJ, Acharya UR, Ray AK, Chakraborty C. Application of higher order cumulants to ECG signals for the cardiac health diagnosis. Conf Proc IEEE Eng Med Biol Soc 2011; 2011: 1697-700.
  28. McMahon EM, Korinek J, Yoshifuku S, Sengupta PP, Manduca A, Belohlavek M. Classification of acute myocardial ischemia by artificial neural network using echocardiographic strain waveforms. Comput Biol Med 2008; 38(4): 416-24.
  29. De Gaetano A, Panunzi S, Rinaldi F, Risi A, Sciandrone M. A patient adaptable ECG beat classifier based on neural networks. Applied Mathematics and Computation 2009; 213(1): 243-9.
  30. Arif M, Malagore IA, Afsar FA. Automatic Detection and Localization of Myocardial Infarction Using Back Propagation Neural Networks. 2010 p. 1-4.
  31. Afsar FA, Arif M. Detection of ST segment deviation episodes in the ECG using KLT with an ensemble neural classifier. 2007 p. 11-6.
  32. Baxt WG, Shofer FS, Sites FD, Hollander JE. A neural network aid for the early diagnosis of cardiac ischemia in patients presenting to the emergency department with chest pain. Ann Emerg Med 2002; 40(6): 575-83.
  33. Kumar A, Singh M. Statistical Analysis of ST Segments for ischemia detection in electrocardiogram signals. J Med Imaging Health Inform 2016; 6(2): 431-40.
  34. Papaloukas C, Fotiadis DI, Likas A, Michalis LK. An ischemia detection method based on artificial neural networks. Artif Intell Med 2002; 24(2): 167-78.
  35. Correa R, Arini PD, Valentinuzzi ME, Laciar E. Novel set of vectorcardiographic parameters for the identification of ischemic patients. Med Eng Phys 2013; 35(1): 16-22.
  36. Soltani-Aski M, Zakeri A, Rastegar H. Ischemic beats detection using bispectrum analysis of the ischemic episodes in heart rate variability. Indian Journal of Fundamental and Applied Life Sciences 2015; 5(S1): 5107-14.
  37. Rajeswari K, Vaithiyanathan V, Amirtharaj P. A novel risk level classification of ischemic heart disease using artificial neural network technique - an Indian case study. Int J Mach Learn Comput 2011; 1(3): 231-5.
  38. Gudipati P, Rajan PK. Ischemic episode detection in an ECG waveform using discrete cosine transform and artificial neural network. 2008 p. 218-21.
  39. Nakajima K, Matsuo S, Wakabayashi H, Yokoyama K, Bunko H, Okuda K, et al. Diagnostic performance of artificial neural network for detecting ischemia in myocardial perfusion imaging. Circ J 2015; 79(7): 1549-56.
  40. Rawson TM, Moore LSP, Hernandez B, Charani E, Castro-Sanchez E, Herrero P, et al. A systematic review of clinical decision support systems for antimicrobial management: are we failing to investigate these interventions appropriately? Clin Microbiol Infect 2017; 23(8): 524-32.
  41. Berner ES. Clinical decision support systems. New York, NY: Springer; 2007. p. 23-67.
  42. Al-Shayea QK. Artificial neural networks in medical diagnosis. International Journal of Computer Science Issues 2011; 8(2):150-4.
  43. Anagnostou T, Remzi M, Lykourinas M, Djavan B. Artificial neural networks for decision-making in urologic oncology. Eur Urol 2010; 43(6): 596-603.
  44. Njie GJ, Proia KK, Thota AB, Finnie RKC, Hopkins DP, Banks SM, et al. Clinical decision support systems and prevention: A community guide cardiovascular disease systematic review. Am J Prev Med 2015; 49(5): 784-95.
  45. AlMahamdy M, Riley HB. Performance study of different denoising methods for ECG signals. Procedia Comput Sci 2014; 37: 325-32.
  46. Battiti R. Using mutual information for selecting features in supervised neural net learning. IEEE Trans Neural Netw 1994; 5(4): 537-50.
  47. Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R. Dropout: A simple way to prevent neural networks from overfitting. J Mach Learn Res 2014; 15(1): 1929-58.
  48. Alves AP, Martins JO, Placido da Silva H, Lourenco A, Fred A, Ferreira H. Experimental study and evaluation of paper-based inkjet electrodes for ECG signal acquisition. Proceedings of the 1st International Conference on Physiological Computing Systems; 2014 Jan 7-9; Lisbon, Portugal.
  49. Dragan D, Ivetic D. A Comprehensive quality evaluation system for PACS. UbiCC Journal 2009; 4(3): 642-50.
  50. Hsieh JC, Lo HC. The clinical application of a PACS-dependent 12-lead ECG and image information system in E-medicine and telemedicine. J Digit Imaging 2010; 23(4): 501-13.
  51. Crawford J, Doherty L. Practical Aspects of ECG Recording. Cumbria, UK: M&K; 2012.
  52. Naya S, Soni MK, Bansal D. Filtering techniques for ECG signal processing. Int J Res Eng Appl Sci 2012; 2(2): 671-9.
  53. Hug CW, Clifford GD. An analysis of the errors in recorded heart rate and blood pressure in the ICU using a complex set of signal quality metrics. Proceedings of 2007 Computers in Cardiology; 2007 Sep 30- Oct 3; Durham, NC, USA. p. 641-4.
  54. Lin YJ, Chang HH. Automatic noise removal in MR images using bilateral filtering associated with artificial neural networks. Int J Pharm Med Biol Sci 2015; 4(1): 39-43.
  55. Limaye H, Deshmukh VV. ECG noise sources and various noise removal techniques: A survey. International Journal of Application or Innovation in Engineering and Management 2016; 5(2): 86-92.
  56. Ayer T, Chhatwal J, Alagoz O, Kahn CE, Jr., Woods RW, Burnside ES. Informatics in radiology: comparison of logistic regression and artificial neural network models in breast cancer risk estimation. Radiographics 2010; 30(1): 13-22.