A Comprehensive Review On Content Based Pulmonary Nodules Retrieval

Pulmonary cancer is one of the deadliest cancers worldwide. Chain-smoking results in lung cancer. Researches assert that early detection of lung cancer considerably increases the chances of recovery rate. A lung nodule represents a range of abnormalities and irregularities in lung tissue. The low dose spiral or helical CT scan gives an effective and efficient way for early-stage lung cancer diagnosis. Currently, the computer tomography research- based analysis includes nodule detection or localization and classification of nodules. This is a crucial task in the initial. stages of pulmonary cancer diagnosis and treatment. The low-dose spiral CT scan gives three-dimensional X-rays of the lungs and a detailed analysis of the lung abnormalities. The deep-learning model for detecting and classifying lung nodules with clinical factors reduces the early-stage lung cancer's misdiagnosis and false-positive (FP) test results. Most pulmonary nodule deep learning detection techniques use the standard LIDC-IDRI dataset for testing and give better detection results than the existing methods.