Abstract
In recent years, deep generative networks have reinforced the need for caution while consuming different formats of digital information. One method of deepfake generation involves the insertion and removal of tumors from medical scans. Significant drains on hospital resources or even loss of life are the consequences of failure to detect medical deepfakes. This research attempts to evaluate machine learning algorithms and pre-trained deep neural networks' (DNN) ability to distinguish tampered data and authentic data. Moreover, this research aims to classify cancer scans based on DNN. The experimental results show that the proposed method based on using DNN can enhance performance detection. Furthermore, the proposed system increased the detection accuracy rate and reduced the number of false alarms.