همایش ملی بیوانفورماتیک ایران

صفحه اصلی / 4th international edition and 13th Iranian Conference on Bioinformatics

Modeling and Predicting the Use of Medications Antiplatelets and ARBs Using Logistic Regression

نویسندگان :

Ahmad Aliyari Boroujeni¹ Pouya Joze Soleimani² Shima Soltani³ Farzaneh Karamitanha⁴

1- دانشگاه زنجان 2- دانشگاه زنجان 3- دانشگاه زنجان 4- دانشگاه علوم پزشکی زنجان

کلمات کلیدی :

Cardiovascular Diseases،Antiplatelets،Angiotensin Receptor Blockers،Classification،Logistic Regression

چکیده :

Cardiovascular diseases account for 40% of mortality in Iran, and heart failure is one of the primary causes. Approximately 1% of individuals in their 50s and 10% in their 80s experience heart failure. One of the methods to manage or treat these diseases is through specific medications. Among these drugs are Antiplatelets, a group of cardiovascular medications that disrupt the process of platelet aggregation, which leads to blood clot formation. By inhibiting this process, Antiplatelets prevent the formation of blood clots, which are the main cause of heart attacks and strokes. Additionally, this medication is used for the prevention and treatment of vascular diseases, such as coronary artery disease. Another class of medications is Angiotensin II Receptor Blockers (ARBs), used to treat high blood pressure and certain other heart and kidney conditions. ARBs work by blocking the angiotensin II receptors in the body. By doing so, they help relax blood vessels and lower blood pressure. In patients with heart failure, ARBs can also improve heart function. Heart failure is usually a lifelong condition requiring ongoing treatment and management, which can significantly impact daily life. In this study, the Minnesota Living with Heart Failure Questionnaire (MLHFQ) (Bilbao et al., 2016) was utilized to assess the impact of heart failure on patients' lives. The questionnaire consists of 21 questions, each rated on a scale from 0 (no problem) to 5 (severe problem), covering three main domains: physical, emotional, and overall quality of life. It was distributed to 100 patients aged 46 to 96. Information related to marital status, income level, hospitalization history, and whether the patients had taken Antiplatelet or ARB medications was collected and evaluated. The overall process aimed to predict the necessity of taking the selected medications, both with and without considering personal information, hospitalization data, and other medications the patient may have been using. To predict the usage of these medications, Logistic Regression (Hosmer et al., 2013) was applied as a binary classification algorithm for the target feature. After performing Principal Component Analysis (PCA) (Jolliffe, 2002) for dimensionality reduction followed by Logistic Regression, the accuracy for Antiplatelet medication prediction was 90%, and for ARBs, it was 85%. Additionally, the results were evaluated in terms of classification metrics such as precision, recall and f1-score, which confirmed the efficacy of this study in predicting the use of these medications for heart patients.