Artificial Intelligence Based Lung Cancer Prediction Using Convolutional Neural Network Algorithm

Medical imaging equipment is essential for the prevention of lung cancer and for monitoring lung cancer patients while they are receiving therapy. For this generation, and for the foreseeable future, lung cancer will continue to be the major cause of cancer-related deaths. If lung cancer signs are identified early, treatment may be available. Sustainable prototyping models for lung cancer therapy can be created using recent advancements in artificial intelligence without having a negative effect on the environment. Reduce the amount of wasted resources, work, and time needed to accomplish manual tasks, saving time and money. For people, estimating movement using conventional techniques is challenging. Due to this, some machine learning algorithms provide efficient and quick predictions with low error on uncertain raw data. The proposed method use Artificial Intelligence (AI)-based Convolutional Neural Networks (CNNs) to build efficient models to identify high-risk groups for lung cancer, allowing early detection to avoid long-term complications from intervention. Initially, dataset images collected from standard repository. Then, started the image preprocessing stage to reduce the noises ad imbalanced data from the image dataset. Second stage is segmentation step is segmenting each image features based on the thresholding algorithm. Then, the features color equalization using for set several methods that make an effort to change or enhance the aesthetic appeal of an image. Thirdly, feature extraction from preprocessed stage photos. Extraction of features based on maximum threshold weight using Support Social Spider optimization (S3O) involves selecting the values of features. Before classification, the features must be evaluated using the SoftMax activation function to estimate the feature weights. Finally, Classification based on AI based CNN algorithm evaluating the risk factors uses the proposed model, and each feature obtained from the softback activation function receives its input from the model’s final fully linked layer. A combination of AI based CNN and softback layers yields 96.52% accuracy. The proposed model is a stable and consistent diagnostic model for lung cancer detection.