International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 06 | June 2022
p-ISSN: 2395-0072
www.irjet.net
LIFE EXPECTANCY PREDICTION FOR POST THORACIC SURGERY T.Mamatha, K.Sudeepa, K.Asha Nandini, S.Vaishnavi Department of Computer Sciences, Sreenidhi Institute of Science and Technology ------------------------------------------------------------------------***-----------------------------------------------------------------------Abstract In order to improve quality initiatives, healthcare administration, and consumer education, it is critical to track health outcomes. The data obtained from patients who had large lung resections for primary lung cancer is referred to as thoracic surgery. Attribute ranking and selection are critical components of successful health outcome prediction when using machine learning algorithms. Researchers used several procedures, such as early-stage examinations, to determine the type of cancer before symptoms appeared. The most relevant attributes are identified using attribute ranking and selection, and the duplicated and unnecessary attributes are removed from the dataset.
Let's learn about the scientific and biological things happen inside a human body. This way it becomes far more easier to work the technical advancements. Thoracic surgery is done when lungs stops working properly. In an eloborated way, lungs stops exchanging of gases which is obviously a death deal. Alveoli are the minute organs in lungs which are critical for exchange of gases. When alveoli fades or dies the septal cells also becomes dead which inturn form a dead tissue what we generally call a Tumor. What makes alveoli die? Many things especially tobacco. Tobacco contains Nicotino carcinoma which is deadly component. Tumor that is responsible for lung cancer can be detected in CT scans which is common way for detecting any kind of abnormality in humans. That is why one of our 17 attributes is smoking criteria.
The goal of our study is to look at patient mortality over the course of a year after surgery. More precisely, we're looking into the patients' underlying health issues, which could be a powerful predictor of surgical-related mortality.
Keywords:
Attribute ranking; Prediction; Thoracic Surgery.
Machine
learning;
1 Introduction The introduction of computer applications into the medical industry has had a direct impact on doctors' productivity and accuracy in recent years. One of these applications is the study of health outcomes.
2 Related Work Even though people are aware of how deadly a cancer can be somehow they are always reckless and careless about taking care. Lung cancer became very obvious that today we are doing a project related to it as a development in technology. Lung cancer cannot be cured but certainly can be prevented and avoided. The most prevalent cause of death after any sort of thoracic surgery is postoperative respiratory problems.
In most nations, cancer is now one of the leading causes of mortality. Thoracic surgery is the most common operation performed on lung cancer patients. Massive datasets of cancer have been collected and made available to medical professionals as a result of the advancement of new tools in the field of medicine.
The predictive models that are provided are based on various supervised machine learning techniques including logistic regression and Random forest as an aim to model cancer risk or patient outcomes.
Many machine learning techniques such as KNN, Logistic regression, random forest etc…are used to predict life expectancy for post thoracic surgery.
Their results indicated that simple logistic regression technique is better or other machine learning techniques with 81% prediction accuracy.
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 2049