Özet:

Abstract Chest X-ray images are extremely difficult to interpret due to the fact that they are produced using a projection imaging modality. This is largely owing to the fact that anatomical structure and disease are closely intertwined. A large number of chest X-rays helps radiologists develop their knowledge and diagnostic abilities after they have mastered the principles of chest X-ray analysis.Droplets fill the lungs and make breathing difficult as a result of pericardial effusion caused by pneumonia. Pneumonia can be treated more effectively and with a higher chance of survival if caught early. Chest X-ray imaging is the most routinely used diagnostic technique for pneumonia. Examining chest X-rays, on the other hand, is a tough task with a high degree of subjectivity. In this study, we employed chest X-ray pictures to develop a computer-aided specialised diagnostic system capable of identifying pneumonia.Researchers have seen how machine learning algorithms can be used to tell if a chest X-ray shows signs of pneumonia. The most important thing about this study's conclusion is that it sorts out the different types of pneumonia. The combined Scale Invariant Fourier Transform (SIFT)and Local Binary Pattern (LBP)features are extracted from each training image and fed into machine learning models such as the Random Forest (RF), Artificial Neural Network (ANN) and Decision Tree (DT) model. After that, the classification model was created and tested on a set of test images. With an accuracy of 91.29%, RF was able to correctly classify all of the patients who had been diagnosed with pneumonia. Based on these results, we can say that the proposed method described in this research paper may help doctors figure out what's wrong with people with typical pneumonia.

Anahtar Kelimeler: