Identifying Deepfakes in CT Scans of Lung Cancer using an Ensemble Architecture

The growing utilization of deep learning in medical imaging has sparked concerns about the authenticity of diagnostic images, particularly in lung cancer CT scans. To address this issue, this work presents a novel deep fake detection system specifically designed for lung cancer CT images. An extensive dataset was assembled, containing both genuine scans and artificially generated deep fake images. This paper employs an ensemble architecture that integrates two pre-trained deep learning models, DenseNet121 and VGG16, to train a convolutional neural network (CNN) capable of accurately distinguishing between authentic and fake CT scans. Advanced techniques such as data augmentation and transfer learning were utilized to enhance the model’s performance and generalization. The system’s effectiveness was evaluated using comprehensive metrics, including accuracy, precision, recall, and F1-score, confirming its capability to identify deep fake images in lung cancer CT scans.