Healthcare faces new challenges with automated medical support systems and artificial intelligence applications for personal care. The incidence of chronic diseases is increasing and monitoring patients in a home environment is inevitable.
Heart failure is a chronic syndrome and a leading cause of death and hospital readmission on developed countries. As it has no cure, patients will have to follow strict medication plans for the rest of their lives. Non-compliance with prescribed medication regimens is a major concern, especially among older people. Thus, conversational systems will be very helpful in managing medication and increasing adherence.
The Personal Medication Advisor aims to be a conversational assistant, capable of interacting with the user through spoken natural language to help him manage information about his prescribed medicines. This patient centered approach to personal healthcare will improve treatment quality and efficacy.
System architecture encompassed the development of three modules: a language parser, a dialog manager and a language generator, integrated with already existing tools for speech recognition and synthesis. All these modules work together and interact with the user through an Android application.
System evaluation was performed through a usability test to assess feasibility, coherence and naturalness of the Personal Medication Advisor. It was concluded that the developed assistant could successfully make users achieve their objectives.
In addition, the system was evaluated as easy to use, coherent and with a pleasant interface. Further improvements can still be done to extend the flexibility and convenience of the Personal Medication Advisor so that an easy integration on senior patient’s daily life can be achieved.
Author: Joana Lobo
Type: MSc thesis
Partner: Faculdade de Engenharia da Universidade do Porto