A Comprehensive Framework For Accurate Classification of Pulmonary Nodules

A precise computerized lung nodule diagnosis framework is very important for helping radiologists to diagnose lung nodules at an early stage. In this manuscript, a novel system for pulmonary nodule diagnosis, utilizing features extracted from single computed tomography (CT) scans, is proposed. This system combines robust descriptors for both texture and contour features to give a prediction of the nodule's growth rate, which is the standard clinical information for pulmonary nodules diagnosis. Spherical Sector Isosurfaces Histogram of Oriented Gradient is developed to describe the nodule's texture, taking spatial information into account. A Multi-views Peripheral Sum Curvature Scale Space is used to demonstrate the nodule's contour complexity. Finally, the two modeled features are augmented together utilizing a deep neural network to diagnose the nodules malignancy. For the validation purpose, the proposed system utilized 727 nodules from the Lung Image Database Consortium. The proposed system classification accuracy was 94.50%.

1. Introduction

Lung cancer is one of the most dangerous and aggressive diseases worldwide. The death rate of lung cancer is increased to be the most serious cancer among every single other sort of cancer. The reason behind that increase is that the survival rate of the lung cancer is very low after diagnosis, because the diagnosis’ dependency on the growth rate of the pulmonary nodule to determine whether it is malignant or benign [1]. The early diagnosis of the pulmonary nodule will increase the survival rate by giving chance to choose the right treatment. The recommended tool for diagnosis of the pulmonary nodule is the low-dose computed tomography (CT) [2]. The main difference between the CT and the original X-ray is that the X-ray uses a fixed X-ray tube instead of having motorized X-ray source that rotates around the patient like in the CT. This technique makes the CT scan contain a lot of data, which is saved in the scans, that needs the radiologist’s reading. Computed aided diagnosis (CAD) system is one of the main tools that is used to help the radiologist as a second reader of the CT scans. A lot of research has been made to integrate the artificial intelligence and deep learning algorithms to enhance the CAD system performance in the diagnosis of pulmonary nodules [3], [4].