Review of Advanced Machine Learning Methods for Lung Disease Diagnoses: Pre-processing, Features Extraction, Classification

Diseases of the lung are common throughout the globe. (COPD) chronic obstructive pulmonary disease, Tuberculosis diseases, pneumonia diseases, asthma, fibrosis, and Lung cancer are the most prevalent types of lung illness. It is crucial to detect lung disease at the earliest opportunity. Many deep learning (DL) and machine learning (ML) patterns have been designed for this aim. Patients with lung diseases are at higher risk of needing ventilators, experiencing breathing problems, and facing oral health issues compared to the general population. Early detection of these diseases can lead to quicker recovery and help prevent the illness from spreading throughout the body. Early diagnosis is crucial for better outcomes. If problems are detected and accurately reflected, additional details will be provided to begin risk improvement and management decision-making. There are many risks in the lung illness detection field of study; one of the issues is the low-resolution medical images utilized to construct a general model to treat various medical images. As an illustration (Computed Tomography, X-ray, and Magnetic Resonance Imaging) are used to create a model that can solve multi-class problems on any image and dimension (resize images), classification according to multiple labels problem. This paper outlines the key advancements in the application of artificial intelligence in medical image processing, namely in evaluating patients’ people in the pulmonary medicine field.