Employing the Capabilities of LSTM and Bi-LSTM for Lung Cancer Detection and Classification

Lung cancer (LC) is one of the leading causes of death globally, impacting approximately five million individuals each year.Early diagnosis of a malignant lung nodule may be beneficial in terms of increasing a patient's chances of survival while also helping pathologists.As a result, computing technology has been used to discover and anticipate numerous human diseases.Since its inception, deep learning (DL) has shown to be an extremely valuable approach for medical research inquiries.Our major goal is to detect tumor-forming lung nodules in lung imaging and assess the severity of LC.The given research aims to demonstrate a deep learning technique that combines convolutional neural networks (CNN), long short-term memory (LSTM), and bi-directional long short-term memory (Bi-LSTM) to automatically detect the presence of LC using CT scans.In this approach, CNN is employed for deep feature extraction, while LSTM and Bi-LSTM are used for classification with features from the IQ-OTH/NCCD dataset.The system produced the expected results on the dataset using the CNN-Bi LSTM model with overall accuracy, recall, precision, and F-score of 99.2,0.95,0.97 and 0.99 respectively, outperforming current innovative methods that use the dataset.