Enhanced Artificial Neural Network Algorithm for Predicting Lung Cancer

This study explores the application of deep learning models for the detection of lung cancer subtypes utilizing histopathological images. Leveraging a diverse dataset containing images of adenocarcinoma, squamous cell carcinoma, normal tissue, and large cell carcinoma, our research aims to develop accurate classification models to aid in early diagnosis and treatment planning. The dataset is meticulously preprocessed, ensuring uniform image sizes and balanced class distributions. Two deep learning architectures, VGG16 and MobileNetV2, are employed and trained on the dataset. The VGG16 model is fine-tuned with additional layers for classification, while the MobileNetV2 model is utilized in its pre-trained form. Rigorous training and evaluation procedures are conducted, including stratified train-test splits and performance metric assessments. Experimental results demonstrate the effectiveness of both models in accurately classifying lung cancer subtypes, with the VGG16 model achieving particularly high accuracy. Comparativeanalysis between the two models highlights their respective strengths and performance differences. Overall, this research contributes to the advancement of computer-aided diagnosis systems for lung cancer detection, offering potential solutions for improving patient outcomes and healthcare decision-making.