VISUAL INQUIRY ANSWER ON MEDICAL SPACE

Abstract

This proposal thinks about strategies to tackle Visual Inquiry Answer in Medical Space (VQA-Medical) assignments with a Deep Learning system. As a primer advance, we investigate Long Short Memory (LSTM) networks utilized in Natural Language Processing (NLP) to handle Question-Replying (content based). We at that point adjust the past model to acknowledge a picture as a contribution to expansion to the inquiry. For this reason, we investigate the Origin ResNet v2 systems to extricate visual highlights from the picture. These are converged with the word inserting of the inquiry to foresee the appropriate response. This work was a piece of the Visual Inquiry Noting Challenge CLEF2018 and data set which was released by Nature Recently. The created programming has pursued the best programming practices and Python code style, giving a predictable gauge in Keras for various designs.