International Research Journal of Engineering and Technology (IRJET) Volume: 10 Issue: 04 | Apr 2023
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
Lung Cancer Detection Using Deep Learning Algorithms Ritesh Jaiswal1, Pooja Shukla2, Mohini Varpe3, Pranay Krishna More4 1,2,3,4 Organization Name:- VPPCOE & VA
Department :- IT University Name:- University Of Mumbai -------------------------------------------------------------------------***-----------------------------------------------------------------------become apparent so it is very tough to identify in its Abstract: One of the leading causes of death worldwide,
beginning stage. Because of this, compared to all other cancer forms, lung cancer has a particularly high mortality rate. The two kind of lung disease which develop and spread in an unexpected way, are small cell lung malignancies (SCLC) and non-little cell lung tumors (NSCLC) [1]. The phase of lung disease alludes to the degree to which the growth has spread in the lung. The World Health Organization reported that more than 7.6 million people worldwide lost their lives to lung cancer each year. Moreover, the death rates of lung cancer are expected upon to keep rising, to wind up around 17 million worldwide in 2030[2]. Despite being the best imaging tool in the medical sector, clinicians find it challenging to interpret and detect cancer from CT scan data. In year of 2005, around 1,362,825 new cancer cases are expected and around 571,590 deaths are expected to happen due to cancer in the United States. It was evaluated that there will be 162,921 deaths from lung cancer, which occurs 30% of all cancer deaths. [3] The extent of the spread of cancer is the basis for the division of lung cancer into stages. It comprises of four stages namely stage I-The cancer is confined to the lung, stages II and III-the cancer is confined to the chest (with larger and more invasive tumor classified as stage III) and Stage IV-Cancer has spread from the chest to other parts of the body.
in both men and women, is lung cancer. According to WHO, the estimated number of lung cancer cases per year is two million. The overall 5-year survival rate for lung cancer patients increases from 16 to 56% if the disease is detected in time Computed Tomography (CT) scan can provide valuable information in the diagnosis of lung diseases This work's primary goals are to identify cancerous lung nodules from the provided input lung image and to categorise lung cancer according to its severity. This study employs cutting-edge Deep learning techniques to locate the malignant lung nodules. This study employs cuttingedge Deep learning techniques to locate the malignant lung nodules. Cancer patients' CT scanned lung images are obtained from various facilities. using image processing techniques like pre-processing, segmentation techniques such as watershed algorithm and feature extraction, area of interest is separated. Features such as texture, geometric, volumetric and intensity features are extracted. Finally, these features are classified using CNN.
Keywords- Lung Cancer, CT, Deep Learning, Watershed, CNN. (Key words)
I. INTRODUCTION Lung cancer disease is the second largest death threat to the world after heart attack, as this cancer is responsible for the largest number of deaths, compared to the number of deaths caused by any other cancer type. [1]. Lung cancer is characterised by unchecked cell proliferation that results in the development of lung nodules. It is reported that lung cancer is responsible for around 19% deaths globally mostly due to alcohol and tobacco consumption. The rate of survival is assured by only 15% survival chances, for a survival period of 5 years. [2]. The main reason for such a high fatality rate is because therapy is delayed due to discovery occurring at a later stage. Chances of survival can rise by 50–70% if lung cancer is discovered sooner. Non-small cell lung cancer and small cell lung cancer are the two major groups into which the lung cancer can be classified based on the cell characteristics. [7] non-small cell lung cancer is the most common type of lung cancer contributing to about 85-90% of total lung cancer cases, while the other 10-15% of the cases is diagnosed with small cell lung cancer. The leading cause of cancer-related deaths worldwide is lung cancer. The ultimate stage of lung cancer is when the symptoms
© 2023, IRJET
|
Impact Factor value: 8.226
There are many techniques to diagnose the lung cancer such as X-rays, Computed Tomography (CT), Magnetic Resonance Imaging (MRI scan), and Sputum Cytology. The problem with these techniques is that it can be time consuming and makes detection possible only at later stages. Despite being the best imaging tool in the medical sector, clinicians find it challenging to interpret and detect cancer from CT scan data. Hence, computer assisted diagnosis might be useful for clinicians to precisely identify the malignant cells. Computer aided techniques such as Deep learning and image processing have been implemented. In our proposed algorithm we have tried to solve these problems. Our developed algorithm can detect cancer affected cell and the corresponding stage such as initial, middle, or final stage. If no cancer affected cell is found in the input image, then it checks the probability of lung cancer.
|
ISO 9001:2008 Certified Journal
|
Page 346