Early detection of lung cancer through CT scan images using MobileNetV2

Early detection of lung cancer stands essential because it represents one of the leading causes of global cancer- connected fatalities. The research establishes a classification system using MobileNetV2 deep learning methodology to sort lung CT scan images among three diagnostic categories of normal and benign and malignant types, this version of MobileNetV2 proved suitable for processing CT lung images through the adaptation of the network using the IQ-OTH/NCCD dataset. Through a transfer learning approach, the model attains accurate results in classification tasks while decreasing both training op- orations and infrastructure requirements because it successfully extracts intricate features. The model demonstrates remarkable performance by achieving an average accuracy of 91% which shows its capability to assist medical diagnosis while helping with large-scale lung cancer screening even in environments that lack substantial computational power.