Machine learning-based coronary artery disease diagnosis: A comprehensive review

Alizadehsani, Roohallah; Abdar, Moloud; Roshanzamir, Mohamad; Khosravi, Abbas; Kebria, Parham M.; Khozeimeh, Fahime; Nahavandi, Saeid; Sarrafzadegan, Nizal; Acharya, U. Rajendra

doi:10.1016/j.compbiomed.2019.103346

Cited by 185 publications

(100 citation statements)

References 113 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In addition, there is no way to verify claims in any of the publications for the accuracy of the algorithms, as the computer code has not been made publicly available 1 . A PubMed search using the same keywords identified ten original articles and one comprehensive review article [9]. Of the ten original articles, three used the Cleveland dataset, whereas the remaining seven used proprietary datasets.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Algorithms for Predicting Coronary Artery Disease: Efforts Toward an Open Source Solution

Akella¹,

Kaushik²

2020

Preprint

View full text Add to dashboard Cite

The development of Coronary Artery Disease (CAD), one of the most prevalent diseases in the world, is heavily influenced by several modifiable risk factors. Predictive models built using machine learning (ML) algorithms may assist healthcare practitioners in timely detection of CAD, and ultimately, may improve outcomes. In this study, we have applied six different ML algorithms to predict the presence of CAD amongst patients listed in an openly available dataset provided by the University of California Irvine (UCI) Machine Learning Repository, named "the Cleveland dataset." All six ML algorithms achieved accuracies greater than 80%, with the "Neural Network" algorithm achieving accuracy greater than 93%. The recall achieved with the "Neural Network" model is also highest of the six models (0.93). Additionally, five of the six algorithms resulted in very similar AUC-ROC curves. The AUC-ROC curve corresponding to the "Neural Network" algorithm is slightly steeper implying higher "true positive percentage" achieved with this model. We also extracted the variables of importance in the "Neural Network" model to help in the risk assessment. We have released the full computer code generated in this study in the public domain as a preliminary effort toward developing an open solution for predicting the presence of coronary artery disease in a given population and present a workflow model for implementing a possible solution.

show abstract

Section: Introductionmentioning

confidence: 99%

Machine Learning Algorithms for Predicting Coronary Artery Disease: Efforts Toward an Open Source Solution

Akella¹,

Kaushik²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Ten-fold cross-validation was used on the extracted data to evaluate the performance of the algorithms. We used the accuracy, sensitivity, specificity, and area under the curve (AUC) of the algorithms for their performance comparison [54][55][56]. The accuracy of these algorithms is shown in Table 2.…”

Section: Resultsmentioning

confidence: 99%

“…Table 4 shows the specificity of the different algorithms used in this research. Specificity is also called the true negative rate and estimates the proportion of actual negatives that are correctly recognized as such [54]. Again, the neural network has the best specificity rate, followed by rule induction, KNN (K = 22), KNN (K = 25), and LDA.…”

Section: Resultsmentioning

confidence: 99%

Physical Fatigue Detection Using Entropy Analysis of Heart Rate Signals

et al. 2020

Self Cite

View full text Add to dashboard Cite

Physical fatigue is one of the most important and highly prevalent occupational hazards in different industries. This research adopts a new analytical framework to detect workers’ physical fatigue using heart rate measurements. First, desired features are extracted from the heart signals using different entropies and statistical measures. Then, a feature selection method is used to rank features according to their role in classification. Finally, using some of the frequently used classification algorithms, physical fatigue is detected. The experimental results show that the proposed method has excellent performance in recognizing the physical fatigue. The achieved accuracy, sensitivity, and specificity rates for fatigue detection are 90.36%, 82.26%, and 96.2%, respectively. The proposed method provides an efficient tool for accurate and real-time monitoring of physical fatigue and aids to enhance workers’ safety and prevent accidents. It can be useful to develop warning systems against high levels of physical fatigue and design better resting times to improve workers’ safety. This research ultimately aids to improve social sustainability through minimizing work accidents and injuries arising from fatigue.

show abstract

“…However, as with all technological innovations, there are certain limitations which are yet to be addressed. Using ML in the field of diagnostic and treatment is always accompanied by the dataset limitation such that the decreased availability of features to be fed into the system can impact on ML performance, especially in disease diagnostics [21]. This is further reinforced by the vulnerability of the system given its dependence on the data used for training; hence, erroneous or biased data will result in flawed outputs.…”

Section: Resultsmentioning

confidence: 99%