Journals / Türk Göğüs Kalp Damar Cerrahisi Dergisi / 2018 / Cilt: 26 - Sayı: 3

Machine learning techniques in cardiac risk assessment

Makine öğrenmesi teknikleriyle kardiyak risk değerlendirmesi

Pages
394–401
Publication Date
2022-09-02
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Development of hospital-specific local assessment systems using hospital data, such as the application in this study, would be beneficial for both patients and doctors.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Background: The objective of this study was to predict the mortality risk of patients during or shortly after cardiac surgery by using machine learning techniques and their learning abilities from collected data.Methods: The dataset was obtained from Acıbadem Maslak Hospital.
📊
Key Findings & Data:First, Standard EuroSCORE scores of patients were calculated and risk groups were determined, because 30-day follow-up information of patients was not available in the dataset.
💡
Conclusion & Impact:Development of hospital-specific local assessment systems using hospital data, such as the application in this study, would be beneficial for both patients and doctors.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Domain Parameters

Abstract

Background: The objective of this study was to predict the mortality risk of patients during or shortly after cardiac surgery by using machine learning techniques and their learning abilities from collected data.Methods: The dataset was obtained from Acıbadem Maslak Hospital. Risk factors of the European System for Cardiac Operative Risk Evaluation (EuroSCORE) were used to predict mortality risk. First, Standard EuroSCORE scores of patients were calculated and risk groups were determined, because 30-day follow-up information of patients was not available in the dataset. Models were created with five different machine learning algorithms and two different datasets including age, serum creatinine, left ventricular dysfunction, and pulmonary hypertension were numeric in Dataset 1 and categorical in Dataset 2. Model performance evaluation was performed with 10-fold cross-validation.Results: Data analysis and performance evaluation were performed with R, RStudio and Shiny. C4.5 was selected as the best algorithm for risk prediction (accuracy= 0.989) in Dataset 1. This model indicated that pulmonary hypertension, recent myocardial infarct, surgery on thoracic aorta are the primary three risk factors that affect the mortality risk of patients during or shortly after cardiac surgery. Also, this model is used to develop a dynamic web application which is also accessible from mobile devices (https://elifkartal.shinyapps.io/euSCR/).Conclusion: The C4.5 decision tree model was identified as having the highest performance in Dataset 1 in predicting the mortality risk of patients. Using the numerical values of the risk factors can be useful in increasing the performance of machine learning models. Development of hospital-specific local assessment systems using hospital data, such as the application in this study, would be beneficial for both patients and doctors.

Özet

Amaç: Bu çalışmada amaç; makine öğrenmesi tekniklerini ve bu tekniklerin veriden öğrenme yeteneğini kullanarak kalp ameliyatı sırasında ya da kalp ameliyatı geçirdikten kısa bir süre sonra hastanın mortalite riskini öngörebilmektir. Çalış­ma­ planı:­ Veri seti Acıbadem Maslak Hastanesi'nden temin edildi. European System for Cardiac Operative Risk Evaluation (EuroSCORE) risk faktörleri, mortalite riskini tahmin etmek için kullanıldı. Hastaların 30 günlük takip bilgileri veri setinde mevcut olmadığından çalışmada ilk olarak Standart EuroSCORE puanları hesaplandı ve risk grupları belirlendi. Modeller; beş farklı makine öğrenmesi algoritması ile yaş, serum kreatinin, sol ventrikül disfonksiyonu ve pulmoner hipertansiyonun Dataset 1’de sayısal, Dataset 2’de kategorik olduğu iki farklı veri kümesiyle oluşturuldu. Model performans değerlendirmesi, 10-kat çapraz geçerleme ile yapıldı. Bulgular: Veri analizi ve performans değerlendirmesi R, RStudio ve Shiny ile gerçekleştirildi. C4.5 algoritmasıyla Dataset 1 üzerinde kurulan model risk tahmini için en iyi model olarak seçildi (doğruluk= 0.989). Bu model; pulmoner hipertansiyon, geçirilmiş miyokard enfarktüsü ve torasik aort cerrahisi niteliklerini bir hastanın mortalite riskini etkileyen ilk üç risk faktörü olduğuna işaret etmektedir. Ayrıca, bu model mobil cihazlardan da erişilebilen dinamik bir web uygulaması geliştirmek için kullanıldı (https://elifkartal.shinyapps.io/euSCR/). Sonuç:­ Bir hastanın mortalite riskini öngörebilmede C4.5 karar ağacı modeli, kullanılan veri seti olan Dataset 1’de en iyi performansa sahip bir model olduğu belirlendi. Risk faktörlerinin sayısal değerlerini kullanmak, makine öğrenmesi modellerinin performansını artırmada yararlı olabilir. Bu çalışmadaki uygulamada olduğu gibi hastaneye özgü veri kullanılarak yerel değerlendirme sistemlerinin geliştirilmesi hem hastalar hem de doktorlar için yararlı olabilecektir.

🔖

Cite This Article

KARTAL, E., BALABAN, M. E. (2022). Machine learning techniques in cardiac risk assessment. Türk Göğüs Kalp Damar Cerrahisi Dergisi, Cilt: 26 - Sayı: 3, 394–401. https://ajindex.com/articles/423876

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (9)

  1. RStudio. shinyapps.io. Available at: http://www.shinyapps.io/. [Accessed: November 10, 2017]. 43. RStudio. Shiny. Available at: https://shiny.rstudio.com/. [Accessed: November 10, 2017]. 42. Chang W. Shinythemes: Themes for Shiny. 2016, Available at: https://CRAN.R- project.org/package=shinythemes. Accessed: November 10, 2017]. 41. Chang W, Cheng J, Allaire JJ, Xie Y, McPherson J. Shiny: Web Application Framework for R. 2017. Available at: https:// CRAN.R-project.org/package=shiny. [Accessed: November 10, 2017]. 40. Witten IH, Frank E. Data Mining: Practical machine learning tools and techniques. 2nd ed. San Francisco: Morgan Kaufmann
  2. Hornik K, Buchta C, Zeileis A. Open-source machine learning: R meets weka. Comput Stat 2009
  3. Wei B, Yang F, Wang X, Ge Y. knnGarden: Multi-distance based k-Nearest Neighbors, 2012, Available at: https:// CRAN.R-project.org/package=knnGarden. [Accessed: November 10, 2017]. 37. Meyer D, Dimitriadou E, Hornik K, Weingessel A, Leisch F. e1071: Misc Functions of the Department of Statistics (e1071), TU Wien, 2017, Available at: https://CRAN.R-project.org/package=e1071. [Accessed: November 10, 2017]. 36. RStudio, Take control of your R code. Available at: https://www.rstudio.com/products/rstudio/. [Accessed: November 10, 2017]. 35. R-project, The R Project for Statistical Computing. Available at: http://www.r-project.org/. [Accessed: November 10, 2017]. 34. Sokolova M., Lapalme G. A Systematic analysis of performance measures for classification tasks. Inf Proc Manage 2009
  4. Karabulut E, Alpar R. Lojistik regresyon, In: Alpar R, editör. Uygulamalı Çok Değişkenli İstatistiksel Yöntemler. Ankara: Detay Yayıncılık
  5. s. 591-660. 32. Dekhtyar A. CSC 466: Knowledge Discovery from Data - Distance/Similarity Measures. Available at: http://users.csc. calpoly.edu/~dekhtyar/560-Fall2009/lectures/lec09.466.pdf. [Accessed: November 10, 2017]. 31. Gan G. Data Clustering in C++: An Object-Oriented Approach. Florida: CRC Press
  6. Balaban ME, Kartal E. Veri Madenciliği ve Makine Öğrenmesi Temel Algoritmaları ve R Dili ile Uygulamaları. 1. Baskı. İstanbul: Çağlayan Kitabevi
  7. Özkan Y. Veri madenciliği yöntemleri. İstanbul: Papatya Yayıncılık Eğitim
  8. Han J, Kamber M. Data mining: concepts and techniques (the Morgan Kaufmann Series in data management systems). 2nd ed. California: Elsevier
  9. EuroSCORE for patients. Available at: http://www.euroscore.org/patienteuroscore2.