BREAST CANCER DIAGNOSIS BASED ON THERMOGRAPHY IMAGES USING PRE-TRAINED NETWORKS

Aim: Breast cancer is the leading cause of death among women around world. Because its low cost and fact that it does not emit hazardous radiation, infrared thermography has emerged as a viable approach for diagnosing condition in young women. This study aims to create computer-aided diagnostic system can process thermographic breast images classify with pre-trained networks order use method.
 Materials Methods: In this study, an open-access data set consisting was used purposes. The consists 179 healthy 101 from patients. were converted .txt format .jpeg format. acquired http://visual.ic.uff.br/dmi/. various reduce training time. Different metrics employed assess performance models.
 Results: obtained during modeling phase display both breasts image without distinguishing right left breasts, is, fragmenting images. According results different network models after preprocessing stages, best classification achieved ResNet50V2 model accuracy value 0.996.
 Conclusion: diagnosis created by developing interface addition experimental findings. web software based on proposed provided promising predictions developed help medical other healthcare professionals easily spot cancer.