Generation of Histopathological Images Caption Using CNN and LSTM

The study of image caption generation has consistently sparked researchers’ interest in the artificial intelligence division. Histopathological image generation, Robotic Vision, Business, and many other fields would benefit greatly from the ability to program a machine to accurately describe an image or an environment like a typical human. In the history of artificial intelligence, this has proven to be a difficult task. In this paper, we focus on the various RNN techniques and analyze their impact on sentence generation to present different histopathological image caption-generating models based on deep neural networks. To determine which feature extraction and encoder model provides the best accuracy and produce the desired results, we also created captions for sample medical images and compared them to one another.