lung Nodule Detection

Automatic Lung Nodule Detection in CT Images

Noor Khehrah, Muhammad Shahid Farid, Saira Bilal, Muhammad Hassan Khan

Introduction:
Cancer is hard to analyze and is a heterogeneous disease which adds to the trouble of conclusion and forecast. The lung tumor is among the most detrimental kind of malignancy. It has high occurrence rate and high death rate as it is frequently analyzed at the later stages. Therefore, a noteworthy research exertion is being made to help the oncologists in early lung disease diagnosis. Computed tomography (CT) scans are broadly used to distinguish the disease; it envisions little nodules or tumors which cannot be seen with a plain film X-ray. Personal computer helped gadgets are being created to analyze the ailment at prior stages productively. In this paper, we present a fully automatic framework for nodule detection from lungs CT images. The working of the proposed system can be divided into two phases: lung segmentation and nodule detection. The preparatory phase of lung malignancy analysis using computer-aided design is lung segmentation from the chest CT scans. Ac curate lung segmentation is momentous in such frameworks as the execution of the later stages in such examination to a great extent relies upon the division accuracy. In the proposed system, a histogram of the grayscale CT image is built to automatically evaluate the limit to isolate the lung locale from the foundation. Then the associated segments are processed to expel any residual foundation. Morphological operators are utilized to enhance the segmentation accuracy. In the second phase, the internal structures are extracted from the parenchyma. A threshold-based technique is proposed to separate the nodules candidates from the other structures such as bronchioles and blood vessels. Different statistical features and shape-based features are extracted for these nodules candidates to formulate a nodule feature vector. These features are then classified using support vector machines to obtain the nodules. The proposed method is evaluated on a large dataset of lungs CT scans collected from the Lung Image Database Consortium (LIDC). This collection con tained 70 cases of lungs acquired using different CT scanners including Siemens, Toshiba, and General Electric. Each scan contains varying numbers of image slices. Our method achieved excellent results compared to the current state-of-the-art,achieving a sensitivity rate of 93.75% and false positive rate of 0.13 per image which shows its effectiveness. The experimental evaluation and the comparison with the existing CAD systems demonstrate that the proposed framework is highly accurate in lung nodule detection. The proposed system can provide a great assistance to the radiologists in lung nodule analysis. Moreover, the proposed system is economical to design as it requires regular computers for deployment which are usually already available in hospitals and clinics or can be easily procured.

Download Matlab Code

Download Dataset (LIDC)

Last updated: Feb 2, 2020