Explore chapters and articles related to this topic
Lung Nodule Classification Basedon the Integration of a Higher-Order Markov-Gibbs Random Field Appearance Model and Geometric Features
Published in Ayman El-Baz, Jasjit S. Suri, Lung Imaging and CADx, 2019
Ahmed Shaffie, Ahmed Soliman, Ali Mahmoud, Mohammed Ghazal, Hassan Hajjdiab, Robert Keynton, Guruprasad Giridharan, Adel Elmaghraby, Jasjit S. Suri, Ayman El-Baz
Lung cancer is the second most common cancer among men and women the world over. It is second only to prostate cancer in men and breast cancer in women. Moreover, it is considered the leading cause of cancer-related deaths among both genders in the United States, as the number of people who die each year of lung cancer is more than the number of people who die of breast and prostate cancers combined [1]. The number of patients suffering from lung cancer has recently increased significantly all over the world, increasing the motivation for developing accurate and fast diagnostic tools to detect lung cancer earlier in order to increase the patient's survival rate. Lung nodules are the first indication to start diagnosing lung cancer. Lung nodules can be benign (normal subjects) or malignant (cancerous subjects). Figure 9.1 shows some samples of benign and malignant lung nodules.
Lung Lesion Classification Scheme Using Optimization Techniques and Hybrid (KNN-SVM) Classifier
Published in IETE Journal of Research, 2022
K. Vijila Rani, S. Joseph Jawhar
Cancer is a class of diseases characterized by out-of-control cell growth. There are over 100 different types of cancer, including breast cancer, skin cancer, lung cancer, colon cancer, prostate cancer, and lymphoma. Each cancer is classified by the type of cells where the symptom is initially displayed. Cancer harms the body through damaging the cells and that divides uncontrollably, to form lumps or mass of tissue called tumours. Lung cancer is the most dangerous and has minimum survival rate. Early detection of lung cancer is a question mark to the researchers. Lung nodules are important abnormality of tissue that present in the lung. A pulmonary nodule is also lung abnormality that looks like round, white glooms on the Computerized Tomography (CT) scan [1, 2]. Various imaging modalities such as CT scan, Positron Emission Tomography (PET) scan, Bronchoscopy, or tissue biopsy are used to detect whether the nodules are tumorous or not. The clear image for lung nodule, size, shape, and location for nodules is obtained by using the CT scan. Early detection of lung cancer is very important, because there are 220,000 new cases and 160,000 deaths are reported per year in India and the survival rate is up to 5 years according to the study report of Malabar Cancer Centre, Kerala, India.
Watershed Segmentation with CAFIS and RCNN Classification for Pulmonary Nodule Detection
Published in IETE Journal of Research, 2021
S. Albert Jerome, K. Vijila Rani, K. S. Mithra, M. Eugine Prince
The demises at an anticipated rate are caused by cancer. A large volume of logbook scrutinized based on lung cancer finds top ranking for homo sapiens death. Hence, prior diagnosis by oncologists can effectively deescalate the effect of it. Cancer is mainly caused by the cell group called Obstreperous burgeon. These tissues are invaded to form malignant tumor. Thus for research, CT images are considered. For effective and precise scanning, CT scan images are considered as more sensitive. The objective of this work is to construct a system with CT images as inputs and corresponding outputs are obtained. The algorithm proposed is evaluated based on sensitivity, accuracy and specificity, and accuracy. Many research works have been brought forward for the efficient recognition of pulmonary nodules. In comparison of X-ray to CT images, the successive method has increased the diagnosis of pulmonary nodules. These abnormal cells grow and form new cells called cancer cells [1,2]. Lung cancer is a heterogenous disease that causes cancer demises. The abnormal cells collectively lead to cancerous tissues [3–5] forming nodule. Not all tumors are classified as cancerous [6,7]. Benign nodules are non-cancerous. Malignant nodules grow as non-sequence cells that obliterate the healthy tissues [8,9].
Lung nodules detection using grey wolf optimization by weighted filters and classification using CNN
Published in Journal of the Chinese Institute of Engineers, 2022
Anas Bilal, Guangmin Sun, Yu Li, Sarah Mazhar, Jahanzaib Latif
According to a previous study (Manser et al. 2013), lung cancer is one of the most common type of cancers to be identified and diagnosed at the earliest point in the field of medicine. Lung cancer can be predicted by a physician with the help of different signs of the disease, such as having a bloody cough, difficulty breathing, weight loss, chest irritation, exhaustion, cognitive impairment, knee pain, bone fracture, neurological complications, bleeding, facial stiffness, headache, and changes in sputum color. Various technological innovations have impacted the patient and been continuously used for diagnosis as screening procedures (Wang et al. 2015). These include genetic testing, bronchial scoping, reflex testing, fluid biopsy, and blood tests. From diagnostic strategies, the National Institute for Health and Care provides general guidelines for reliably predicting lung cancer and its stages. Screening methodologies have successfully studied lung tissue and the variations throughout the cells used to predict lung cancer, but calculating performance is challenging to assess. Among these screening techniques, computer-aided tomography (Hulbert et al. 2017) is an efficient screening procedure that accurately investigates the variations and changes in the body that are detected by X-rays diffusing the body. The X-rays imaging and modeling efficiently examines the internal anatomy structure and the tissues and affected sections, which are easy to process as compared to the positron emission tomography (PET) and magnetic resonance imaging screening methods. Lee et al. (2015) used CT images to automatically predict and diagnose lung cancer using a range of standard steps, along with image noise removal, field segmentation and feature extraction, feature collection, and tumor classification. Lung cancer can be divided into initial and late stages, where it spreads into other parts of the body in the late stage (Islam, Mahamud, and Rab 2019). CT scans are used to evaluate the nodules for cancer detection. Contrast-enhanced CT scans can provide high-resolution, sharp boundaries that are more useful for detecting nodules in the lung region (Chen et al. 2019). CT images contain multiple slices in a single scan. Detecting and diagnosing nodules from many slices is a difficult task for radiologists, which ultimately leads to an urgent demand for computer-aided diagnostic (CAD) CT images (Moradi and Jamzad 2019).