Obtaining the appropriate feature space in reporting for medical images with a Greedy algorithm and DistilBert language model

Obtaining the appropriate feature space in reporting for medical images with a Greedy algorithm and DistilBert language model
With the increasing progress of science and the integration of the two fields of computer vision and natural language processing, researchers have succeeded in preparing medical reports, which, of course, have shortcomings such as the brevity of production reports. But good help was given to doctors in serious matters and the process of diagnosis and treatment of diseases was made faster. Reporting for medical images is one of the emerging fields of artificial intelligence, which despite extensive research in this field, is still full of challenges and has become one of the active research fields. Our goal in this research is to take effective action to remove the limitations of medical reporting so that we can have a more suitable space in the process of reporting medical images with chest X-rays and more complete and longer medical reports. We have used the DistilBert language model and the greedy algorithm to obtain this feature space suitable for generating medical reports. Then, our proposed model is based on encoder-decoder architecture and attention mechanism of vision training, finally, we evaluated the model using natural language measurement criteria such as BLEU, ROUGE METEOR, and Elmo and succeeded in producing longer reports.