University Links: Home Page | Site Map
Covenant University Repository

A framework for intelligent speech-based self-care health information and disease diagnosis systems.

Oyelami, O. M. and Uwadia, C. O. and Akinwale, A. A. (2009) A framework for intelligent speech-based self-care health information and disease diagnosis systems. In: South African Institute of Computer Scientists and Information Technologists (SAICSIT 2009), 12-14 October, 2009, South Africa.

[img] PDF (PDF) - Published Version
Download (159kB)


Self-care health information obtained from the Internet is largely delivered in text form, and therefore inadequate for the needs of the blind, the visually impaired and the computer illiterates. Existing speech-based disease screening systems also lack reasoning capability expected of disease diagnosis systems. In this work, we have designed and developed SSDDeS that makes possible, speech-based access to self-care health information. The system also diagnoses diseases using rule-based reasoning technique. Selected number of fever rampant in Africa is diagnosed by the system. Health information for the types of fever diagnosed by the system and information about how these diseases are diagnosed was obtained from the American Medical Library Association recommended websites and doctors. VoiceObjects Desktop for Eclipse was used to develop, test and deploy the prototype application. Voxeo Prophecy was used as the media platform, and server-side Javascript was used to specify the rules using script objects. Logic objects were used to control the logic of the dialog flow. VoiceObjects embedded database was used to store the context of the problem domain, while the inbuilt rule engine within VoiceObjects determines which rule gets fired. X-Lite soft phone was used to call the application. This initiative makes web-based health information available through speech and takes care of the needs of the estimated 180,000,000 visually impaired and blind people worldwide as well as the computer illiterates that cannot access websites. It also incorporates reasoning, which existing systems lack, into speech-based disease screening systems to enable them diagnose more than one disease.

Item Type: Conference or Workshop Item (Paper)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: Dr. Olufemi M. Oyelami
Date Deposited: 08 Oct 2015 16:55
Last Modified: 08 Oct 2015 16:55

Actions (login required)

View Item View Item