RHINOS: A Consultation System for Diagnoses of Headache and Facial Pain
Abstract
How do doctors make diagnoses based on their medical knowledge? In this paper a hypothesis is proposed concerning a diagnostic method used by doctors. The knowledge for the hypothesis has been provided by an authority In the field of medicine, and a diagnostic system for headache and facial pain, named RHINOS, has been implemented using the programming language Prolog, which is operative on an NEC PC-9801 microcomputer. Because it is operative on a microcomputer, RHINOS is portable and readily available. RHINOS has four kinds of rules: two levels of rules perform as forward links from manifestations to diseases. It also has disease-images that act as backward links from manifestations to diseases. It also has disease-images that act as backward links from diseases to manifestations. Through harmonious use of this knowledge, RHINOS makes not only single diagnosis but also makes differential-diagnoses. It also diagnoses cases of complications of two or more diseases. The disease-image link made this possible. RHINOS was used to diagnose 60 patients and 82% of its diagnostic results were equal to those made by a specialist, 16% were close to those made by a specialist. This shows that RHINOS is capable of diagnosing patients in almost the same way as a specialist.
Cite
Text
Kimura et al. "RHINOS: A Consultation System for Diagnoses of Headache and Facial Pain." International Joint Conference on Artificial Intelligence, 1985.Markdown
[Kimura et al. "RHINOS: A Consultation System for Diagnoses of Headache and Facial Pain." International Joint Conference on Artificial Intelligence, 1985.](https://mlanthology.org/ijcai/1985/kimura1985ijcai-rhinos/)BibTeX
@inproceedings{kimura1985ijcai-rhinos,
title = {{RHINOS: A Consultation System for Diagnoses of Headache and Facial Pain}},
author = {Kimura, Michio and Matsumura, Yasushi and Matsunaga, Takashi and Hata, Ryuji and Matsumura, Hiroshi},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {1985},
pages = {393-396},
url = {https://mlanthology.org/ijcai/1985/kimura1985ijcai-rhinos/}
}