Cancerous Lung Nodules Detection and Classification Based on Computer Aided Diagnosis System
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2017, Vol 5, Issue 6
Abstract
Lung cancer is the leading cause of cancer death among both men and women; about 1 out of 4 cancer deaths are from lung cancer, referring to extraordinary growth of lung nodule in the lung. Computed tomography (CT) scan is employed to detect cancerous lung nodule, which is round lesion with variable size. The methodology used in proposed system uses image processing techniques followed by Region of Interest(ROI) extraction which is further processed by image transformation using Discrete Wavelet Transform (DWT), here db1, db2, db3 and db4 are computed. Further 11 features are computed on transformed image finally two best features are selected as input to Support Vector Machine (SVM) which is used to distinguish diseased image with healthy image.
Authors and Affiliations
Dr. Mallikarjun Swamy, Amit Kumar, Arpit Singh, Nihal Shree, Sangeet Sudhanshu
Keywords
Related Articles
- EP ID EP24517
- DOI -
- Views 329
- Downloads 10
Hybrid Isolated Wind–Hydro System by Using Induction Generators and Battery Storage
Now a days Most of natural energy sources such as wind, geothermal, solar, biomass, and hydro show predominance for energy production. Out of these natural energy sources, small wind and small hydro energy which are com...
PLC Based Automatic Nozzle Welding Control System
In industries, to produce different products we need various types of raw materials. Those can be in solid, liquid or gaseous state. To store such large quantity of material, we need big containers. Hence large containe...
ritical Studies on the Influence of Silica Fume and Steel Fiber Enhancing Properties of Concrete
This paper presents results of concrete containing Portland Pozzolona cement (PPC) containing fly ash with silica fume (SF) and steel fibers. The tests are carried out in laboratory on M30 grade of concrete at 28 and 90...
Secured Login Authentication on Key Attackers
An innovative solution is proposed to avoid the problem such as password security at application layer. For password security all lower case characters, upper case characters, special characters and digits in password c...
An Error Reduction Hybrid for Achieving Near to Exact Classification Using KSTAR and IBK Classifier
The data mining which is the source of the all the discovery of knowledge, classification always play a vital role and it really means, so without proper classification it is impossible to find the proper accuracy, so t...