| Autor: |
M, SureshKumar, Perumal, Varalakshmi, Yuvaraj, Gowtham, Rajasekar, Sakthi Jaya Sundar |
| Předmět: |
|
| Zdroj: |
Concurrent Engineering: Research & Applications; Dec2022, Vol. 30 Issue 4, p325-334, 10p |
| Abstrakt: |
The survival percentage of lung patients can be improved if pneumonia is detected early. Images of the chest X-ray (CXR) are the most common way of identifying and diagnosing pneumonia. A competent radiologist faces a difficult problem in detecting pneumonia from CXR images. Many people are at danger of contracting pneumonia, especially in developing countries where billions of people live in energy poverty and rely on polluting energy sources. Though there are effective tools in existence to prevent, diagnose and treat pneumonia, pneumonia-related deaths are prevalent in most of the countries. But only a small amount of health budgets is allocated to eradicate pneumonia. If the diagnosis of the disease is made in more reliable and cost effective way, tackling the disease won't be a herculean task. Machine learning algorithms paved a great way to easily identify, diagnose and predict the disease with minimal amount of time. This paper represents the identification of pneumonia from chest X-Ray by implementing traditional machine learning algorithms with ensemble using optimal number of image features with the help of correlation co-efficient. Also deep learning approach has been implemented. The proposed method traditional machine learning approach and deep learning approach achieved accuracy rates of 93.57% and 93.59% and time required for pneumonia detection is 157,452 s (approx.) and 240,253 s (approx.) respectively. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
