Journal Article

Early Detection of Malignant Tumor in Lungs Using Feed-Forward Neural Network and K-Nearest Neighbor Classifier

Naveen Kumar, H N, Guru Prasad, M S, Mohan Naik, R, Parveen, Gujjar, Avinash, S

2023

Overview

Cancer is one among the prime causes for high mortality rate worldwide. This can be minimized by detecting the cancer disease at primitive state. In the proposed work, lung cancer is taken into consideration since lungs are one among the leading cancer sites. In the proposed work, X-ray image of lungs is used to detect the malignant tumor at its primitive state. The segmentation is carried out on chest X-ray images using Otsu’s thresholding algorithm. The counts of white and black pixels, histogram features of the segmented image are the elements of feature vector. The Feed-Forward Neural Network with Back Propagation (FFNN-BP) and K-Nearest Neighborhood Classifier (KNNC) are employed on feature vector for the classification of tumor as benign or malignant. The work is implemented on Kaggle dataset. The recognition accuracy of the proposed work is found to be superior compared to state-of-the-art approaches.

• Publisher: Springer Nature Singapore,Springer Nature B.V
• Subject: Advances in Computational Intelligence for Artificial Intelligence / Algorithms / Automation / Back propagation networks / Classification / Classifiers / Computer Imaging / Computer Science / Computer Systems Organization and Communication Networks / Data Structures and Information Theory / Datasets / Image databases / Information Systems and Communication Service / Internet of Things and Data Analytics / K-nearest neighbors algorithm / Lung cancer / Machine Learning / Magnetic resonance imaging / Medical imaging / Metastasis / Morphology / Neural networks / Original Research / Pattern Recognition and Graphics / Radiation / Software Engineering/Programming and Operating Systems / Support vector machines / Tumors / Vision