A Method for Predicting Cardiovascular Disorder using Machine Learning Techniques

Drastic increase of cardiovascular disease has led to a lot of adult's death. As per "News 18" very year, 12 million young people in India die from heart disease. In most of these cases people experience cardiac arrest, some of which are normal. First, families of patients are vulnera...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2024 International Conference on Recent Advances in Science and Engineering Technology (ICRASET) s. 1 - 7
Hlavní autoři: Hegde, Ramakrishna, Pavithra, D R, Shivashankara, S, Prasanna Kumar, G, Soumyasri, S M, Nagashree, S
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 21.11.2024
Témata:
Accuracy
Cardiac arrest
Cardiovascular Disease
Classification algorithms
Feature extraction
Heart
K Nearest Neighbour algorithm
Linear regression
Machine Learning
Nearest neighbor methods
Prediction algorithms
Predictive models
Random Forest Classifier
Random forests
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Drastic increase of cardiovascular disease has led to a lot of adult's death. As per "News 18" very year, 12 million young people in India die from heart disease. In most of these cases people experience cardiac arrest, some of which are normal. First, families of patients are vulnerable because it takes only a few minutes for a person to die of a heart attack, and it is difficult to get medical help in time. Second, most victims are under the age of 45. According to Dr. Maninder Sandhu, cardiologist and cardiologist at Artemis Hospital Gurgaon, 30 percent of people who suffer from heart disease are under the age of 45. This paper proposes a method that maybe used by laboratory technicians, doctors and users to understand whether there are any chances of having cardiovascular disease or not. The proposed system takes live data of patients and their lifestyle attributes these data are processed by the trained model to produce the output. In this model we haveused random forest classifier, linear regression, \mathbf{k}-neighbors classifier & x gradient boost classifier. This model has attained maximum accuracy and least type one error in "X- Gradient Boost". We have concentrated more on type one error. The developed system shows whether the patient has any chances of cardiovascular disease or not.
DOI:10.1109/ICRASET63057.2024.10895119