Knowledge Translator for Clinical Decision Support in Treating Hearing Disorders

Tinnitus, commonly known as a "ringing in the ears," is a common hearing disorder. There are currently no approved drugs to cure tinnitus, but there were methods developed to manage it. One of them is tinnitus retraining therapy (TRT), which in multiple clinical trials proved a clinical success rate of 80%. However, the therapy is not widely offered and there is still no Standard of Care for tinnitus. TRT is administered in specialized clinics, and its successful practice requires specific and not widely available training as well as years of practice to develop expertise in it. Within this research, we are developing eTRT, a knowledge-based clinical decision support system designed to electronically assist practitioners to provide TRT. Within this paper, we describe a knowledge translator which is a component of eTRT that extracts machine-learned rules and automatically builds them into a knowledge base of TRT-based diagnostics and treatment actions. We describe the experiments measuring the volume and time of developing the TRT knowledge base. The developed translator is generic and can be applied for other projects that would benefit from automatic development of rule knowledge base to further reason from the rules and provide explainable output.