An Adaptive Neuro-Fuzzy Inference System-Based Lung Cancer Detection System

Abstract

In the last forty years, there have been dramatic growths in the area of medical and healthcare systems. During this period, the true causes of several illnesses were uncovered, fresh diagnostic techniques were created, and new medications were created. Despite these advances, illnesses like cancer continue to plague us because people remain susceptible to the system. Cancer is the second-biggest reason of mortality worldwide, accounting for around one in every six deaths. Therefore, early illness detection considerably increases the likelihood of survival. Among all cancers, lung cancer has the highest risk. Using the capabilities of Artificial Intelligence (AI), tumour diagnosis may be automated to analysis larger capacity in lesser time and at a lesser cost. A Machine Learning based Lung Cancer Detection System (ML-LCDS) is suggested in this research. The automated identification and localization of tumour locations in lung imaging are increasingly crucial for saving patients' lives via prompt medical therapy. In this study, a lung tumour detection, categorization, and segmentation method based on machine learning are suggested. The tumour categorization stage first applies an adaptive median filtration to the original lung computerised tomography picture and then applies Discrete-Timing Complicated Wavelet Transformation (DT-CWT) to divide the whole picture into several sub-bands. In addition to the deconstructed sub-bands, Discrete Wavelet Transformation (DWT), and co-occurrence characteristics are calculated and identified using an ANFIS. The tumour segmentation stage detects tumour locations on this identified abnormal lung image using morphological features. The proposed system exhibits a precision of 93.4%, accuracy of 95.1%, specificity of 90.6%, sensitivity of 92.8%, False positive rate of 0.22%, false negative ratio of 0.18%, and classification accuracy of 98.2%. The outcomes of the simulation show that the proposed system for finding and predicting lung cancer is accurate and precise. The proposed method outperforms all methods and provides better lung cancer detection accuracy than others.