Speech driven natural language understanding for hands-busy recording of clinical information

Abstract

The hands-busy nature of many clinical examinations means that keyboard and mouse driven interface paradigms are unable to capture medical information at source. The impact of compromised data integrity when using such systems and their inability to serve the needs of clinicians in an endoscopic context is examined. A speech driven application has been developed to serve the clinical process of endoscopy and record data at source. The system exploits the power of a natural narrative to capture and generate consistent visual and textual clinical information.