The overall objective of the project is to enable users to interact with the EMR via voice. With this in mind, our initial goal however is more modest. We hope to create a system where a user could just call in and register a health related query, and the system analyzes it and forwards it to the appropriate person/hospital/doctor. The next step would be for it to also dispense minor recommendations, e.g. if the system is aware of a malarial epidemic and someone calls in with a fever and chills, it could give them preliminary advice on malaria and
then connect them to the doctor.
Failure mode: The goal is to have all cases covered. So, if the system is unable to understand what is spoken, it will forward the call to a human operator. The assumption is that if the system makes a mistake and forward the call to the wrong agency, that agency will bounce the call to a human operator who will take correct action.
The objective is to be able to reduce the manpower required to man such information lines, which could be critical in the case of emergencies or epidemics, and would generally be helpful anyway.
Click here .