AUTOMATIC DETECTION AND DIAGNOSIS OF LUNG CANCER USING ADVANCED NEURAL NETWORK CLASSIFIERS FOR MEDICAL IMAGE PROCESSING APPLICATIONS

Lung cancer is currently one of the most common causes of cancer-related death. Detection, and to provide at an early stage of its development of an accurate diagnosis of potentially cancerous lung nodules increase therapeutic effect, thereby reducing lung cancer mortality. A key obstacle to early detection is the absence of obvious symptoms until the cancer has spread. Diagnosis and the use of non-invasive imaging such as Computed Tomography (CT) screening is a potential solution. However, to achieve accurate automatic analysis of these high-resolution images of the potential need for this approach. Recently, image processing technology has been widely used in several medical fields of detection and treatment levels. For the detection of lung cancer in the image processing there are four different stages are analyzed (i) pre-processing (ii) segmentation (iii) feature extraction and (iv) classification. In the first stage of the process, the Adaptive median filtering and Gaussian filtering techniques are implemented to reduce the noise in the CT images. During the second stage of process, the lung cancer region are separated from the pre-processed image using Versatile Linear Iterative Clustering Algorithm (VLICA) and Suboptimal Clustering Technique (SCT). This process eliminates unnecessary areas of interest. As a solution to this, propose system developed for concentrates is used in segmental nodules, thereby helping radiologists to analyze the disease and detecting various stages of lung cancer. In third stage multi model the different features extraction values are obtained for further classification. The proposed Prognostic Assimilate Learning Classifier (PALC) and Contiguous Anisotropic Learning Classifier (CALC) is used to classify the lung cancer region in the digital images. The implemented PALC technique's performance is examined with different classification parameters like Recall, Precision, and F-measure.