Analyzing and detecting risk factors for the diagnosis of angina pectoris with machine learning
| dc.contributor.author | Özhan, Onural | |
| dc.contributor.author | Çiçek, İpek Balıkçı | |
| dc.contributor.author | Tunç, Zeynep | |
| dc.date.accessioned | 2024-08-04T19:42:43Z | |
| dc.date.available | 2024-08-04T19:42:43Z | |
| dc.date.issued | 2023 | |
| dc.department | İnönü Üniversitesi | en_US |
| dc.description.abstract | Aim: To classify angina pectoris (AP) in women by applying the Bagged CART approach, which is one of the machine learning (ML) methods, to the open-access AP dataset. Another aim is to reveal the risk factors associated with AP in women through modeling. Materials and Methods: In the current study, modeling was done with the Bagged CART technique utilizing an open-access data set containing the factors associated with AP. Model results were assessed with accuracy (ACC), sensitivity (Sen), balanced accu racy (BACC), positive predictive value (PPV), specificity (Spe), negative predictive value (NPV), and F1-score performance criteria. In addition, a 5-fold cross-validation approach was applied in the modeling phase. Finally, variable importance was derived with model ing. Results: ACC, BACC, Sen, Spe, PPV, NPV, and F1-score from Bagged CART modeling were 98.5%, 98.5%, 99.0%, 98.0%, 98.0%, 99.0%, and 98.5%, respectively. Depending on the variable importance values calculated for the input variables investigated in the current study, age, family history of myocardial infarction: yes, the average number of cigarettes smoked per day smoking status: current, family history of angina: yes, hypertensive condition: moderate, smoking status: ex, hypertensive condition: mild, family history of stroke: yes, whether the woman has diabetes: yes were obtained as the most important variables associated with AP. Conclusion: With the ML model used, the AP dataset was classified successfully, and the associated risk factors were revealed. ML models can be used as clinical decision support systems for early diagnosis and treatment. | en_US |
| dc.identifier.doi | 10.5455/annalsmedres.2023.02.043 | |
| dc.identifier.endpage | 485 | en_US |
| dc.identifier.issn | 2636-7688 | |
| dc.identifier.issue | 4 | en_US |
| dc.identifier.startpage | 481 | en_US |
| dc.identifier.trdizinid | 1165291 | en_US |
| dc.identifier.uri | https://doi.org/10.5455/annalsmedres.2023.02.043 | |
| dc.identifier.uri | https://search.trdizin.gov.tr/yayin/detay/1165291 | |
| dc.identifier.uri | https://hdl.handle.net/11616/88615 | |
| dc.identifier.volume | 30 | en_US |
| dc.indekslendigikaynak | TR-Dizin | en_US |
| dc.language.iso | en | en_US |
| dc.relation.ispartof | Annals of Medical Research | en_US |
| dc.relation.publicationcategory | Makale - Ulusal Hakemli Dergi - Kurum Öğretim Elemanı | en_US |
| dc.rights | info:eu-repo/semantics/openAccess | en_US |
| dc.title | Analyzing and detecting risk factors for the diagnosis of angina pectoris with machine learning | en_US |
| dc.type | Article | en_US |
