Abstract:One of the leading causes of cancer death for both men and women is the lung cancer. The best way to improve the patient’s chances for survival is the early detection of potentially cancerous cells. But, the conventional systems fails to segment the cancerous cells of various types namely, well-circumscribed, juxta-pleural, juxta-vascular and pleural-tail at its early stage (i.e., less than 3 mm) that leads to less classification accuracy. It is also noted that none of the existing systems achieved accuracy more than 98%. In this paper, we propose an optimal diagnosis system not only for early detection of lung cancer nodules and also to improve the accuracy in Fog computing environment. The Fog environment is used for storage of the high volume CT scanned images to achieve high privacy, low latency and mobility support. In our approach, for the accurate segmentation of Region of Interest (ROI), the hybrid technique namely Fuzzy C-Means (FCM) and region growing segmentation algorithms are used. Then, the important features of the nodule of interest such as geometric, texture and statistical or intensity features are extracted. From the above extracted features, the optimal features used for the classification of lung cancer are identified using the Cuckoo search optimization algorithm. Finally, the SVM classifier is trained using these optimal features, which in turn helps us to classify the lung cancer as either of type benign or malignant. The accuracy of the proposed system is tested using Early Lung Cancer Action Program (ELCAP) public database CT lung images. The total sensitivity and specificity attained in our system for the above said database are 98.13 and 98.79% respectively. This results in a mean accuracy of 98.51% for training and testing in a sample of 103 nodules occurring in 50 exams. The rate of false positives per exam was 0.109. Also, a high receiver operating characteristic (ROC) of 0.9962 has been achieved.

Automatic Detection and Classification of Lung Nodules in CT Image Using Optimized Neuro Fuzzy Classifier with Cuckoo Search Algorithm

An intelligent lung cancer diagnosis system using cuckoo search optimization and support vector machine classifier

An effective neural network model for lung nodule detection in CT images with optimal fuzzy model

A Neural Network and Optimization Based Lung Cancer Detection System in CT Images

Hybrid Deep Learning Model and Fuzzy C Means Clustering Method for Pulmonary Nodule Detection in CT Images

An Automatic Lung Nodule Classification System Based on Hybrid Transfer Learning Approach

An Improved Convolution Neural Network and Modified Regularized K-Means-Based Automatic Lung Nodule Detection and Classification

Automated Detection and Classification for Early Stage Lung Cancer on CT Images Using Deep Learning

Early detection of lung cancer from CT images: nodule segmentation and classification using deep learning

Optimized convolutional neural network for automatic lung nodule detection with a new active contour segmentation

Lung Nodule Classification of CT Images Based on the Deep Learning Algorithms

Lung Nodule Classification on CT Images Using Deep Convolutional Neural Network Based on Geometric Feature Extraction

Automatic Computer Aided System for Lung Cancer in Chest CTs Using MD-RFCN Combined with Tri-Level Region Proposal Network

A Hybrid deep learning model for effective segmentation and classification of lung nodules from CT images

An automatic classification of pulmonary nodules for lung cancer diagnosis using novel LLXcepNN classifier

An Efficient Model for Lungs Nodule Classification Using Supervised Learning Technique

Knowledge Based Optimum Feature Selection for Lung Nodule Diagnosis on Thin Section Thoracic CT

Automated detection of pulmonary nodules in CT images with support vector machines

Lung cancer classification and identification framework with automatic nodule segmentation screening using machine learning

A New Strategy to Detect Lung Cancer on CT Images

An early prediction and classification of lung nodule diagnosis on CT images based on hybrid deep learning techniques