Abstract: Generating radiology reports is time-consuming and requires extensive expertise in practice. Radiology report generation remains challenging in regards to understanding and linking complicated medical visual contents with accurate natural language descriptions. The system focus on generating detailed findings for a report. Such a model can provide interpretable justifications as part of a computer-aided reporting system to assist clinicians in making decisions (i.e. a multi-task scheme consisting of chest x-ray image classification and report generation.) Dataset: public datasets (e.g. Open-I).
Contact: David Ahmedt-Aristizaba