An Innovative Hybrid Deep Learning Technique Based on Neural Networks for Early Detection of Lung Cancer

With a high death rate among affected individuals, lung cancer is a deadly illness. Patients can be saved by receiving an early diagnosis and correctly determining the stage of lung cancer. Lung cancer can be detected using a variety of image processing; biomarker based, and machine automation techniques, although early detection and accuracy are difficult for medical professionals to achieve. This work uses the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) to extract the Computed Tomography (CT) scan images. Conventional techniques use manual CT scans to determine if a patient has lung cancer. This study suggests a brand-new technique called Cancer Cell Detection utilizing Hybrid Neural Network (CCDC-HNN) for an early and precise diagnosis. Deep neural networks are used to extract the features from the CT scan images. To save the patient from this deadly illness, early detection of malignant cells depends heavily on feature extraction accuracy. An advanced 3Dconvolution neural network (3D-CNN) is also used in this study to increase diagnosis accuracy. Additionally, the proposed method makes it possible to distinguish between benign and malignant tumors. The outcomes validate the feasibility of the suggested hybrid deep learning (DL) approach for early lung cancer detection when assessed using conventional statistical methods.