Automated Generation of Medical Reports Using Deep Learning Techniques

Abstract

 Medical Imaging, being in the vanguard of technological transformation, aids in renovating the field of radiology. It is widely used in the diagnosis and treatment of various diseases. Such is done using the radiographs, commonly known as X-Rays. Automation of the diagnostics using radiographs for medical report generation will be beneficial to the field of radiology. It will accelerate the diagnosis as well as aid in the triaging of the report. This paper proposes a system that focuses on chest disease and hence will make use of chest X-Rays for diagnosis. The solution proposed is a method to automate medical report generation, by using Deep Learning Techniques. A system that uses Convolutional Neural Networks (CNNs) along with Long Short-Term Memory Networks (LSTMs) makes this possible.