PERFORMANCE OF LUNG CANCER PREDICTION AT EARLIER STAGE USING NEURAL NETWORKS.
Lung malignancy is one of the most pervasive types of disease, representing one of five passing from malignant growth. When all is said in done, it is brought about by bizarre cell development in the lungs. Acknowledgment and finding of lung disease at the most punctual reference point stage can be extremely valuable in improving patient endurance rate. Yet, disease determination is one of the significant difficulties for radiologists. Lung malignant growth is analysed utilizing biopsy, automated tomography (CT) check, tomography-electronic positron outflow (PET-CT) examine. Multi-arrange grouping was utilized for the location of malignancy. This framework can anticipate the likelihood of lung malignant growth. In each phase of arrangement picture upgrade and division have been done independently. Dataset for preparing is gotten from Lung Image Database Consortium (LIDC) or cancer imaging archive. This paper displays a doable method to group the lung tumour as three phases as starting or mid-arrange or basic. The target of the present examination is to propose a basic yet quick CNN model joined with repetitive neural system (RNN).  The proposed CNN-RNN model nearly beat other AI calculations viable. The precision remained consistently higher than the other methods