Towards a Deep Learnig for Lung Cancer Prediction Approach

Lung cancer stands as one of the primary contributors to mortality, primarily because its symptoms are often detected very late. This delayed detection is partly attributed to the fact that lung cancer shares common symptoms with the flu, making it challenging to identify the disease in its early stages. Moreover, several contributory factors, such as hereditary predisposition and unhealthy habits of the patient, can further exacerbate the risk of developing lung cancer.Numerous research studies have been dedicated to lung cancer, with a primary focus on employing Convolutional Neural Networks (CNN) for image-based diagnosis. However, a significant challenge persists: conventional imaging techniques often identify the disease at an advanced stage, rendering it challenging to save patients. In our paper, we introduce an innovative approach named “DeepLCP” that combines the capabilities of Natural Language Processing (NLP) and CNN. Our proposed methodology seeks to tackle this issue by harnessing the strengths of both NLP and CNN. Preliminary experimental findings reveal promising results, characterized by high accuracy and minimal data loss during the validation process.