SEPHYRES 1: A Symptom Checker based on Semantic Pain Descriptors and Weight Spreading

Authors

  • Ali SANAEIFAR Payam-e-Noor University of Tehran (PNU), Tehran, Iran. http://orcid.org/0000-0001-5164-4533
  • Ahmad FARAAHI Payam-e-Noor University of Tehran (PNU), Tehran, Iran.
  • Mahmood TARA Mashhad University of Medical Science (MUMS), Mashhad, Iran.

Keywords:

Clinical information systems, Clinical decision support, Computer assisted decision making, Knowledge modeling and representation, Telemedicine and telehealth, Computer assisted diagnosis

Abstract

Semantic-enabled medical diagnostic systems, which have exploited an ontology in their internal engines, have failed to perfectly describe disease profiles, especially in complex medical terms having a variant generality level or certainty in the medical literature. The main objective of this paper was to present an ontology with a highly matching grade of proeminent medical concepts able to analyze the patient’s descriptive medical condition. Focusing on semantic pain descriptors and weight spreading techniques, we proposed a semantic-pseudo-fuzzy engine entitled SEPHYRES, with which we tried to present an ontology-based solution using not only a generic semantic reasoner but also complementary domain-heuristic reasoning. Having applied the valid evidence-based references along with local experts, we illustrated how the resilient expressive model represents the complex medical term relations. The twenty test cases were extracted from the MEDSCAPE and PubMed databases and the precision and recall were calculated. Finally, the results were compared against the Isabel symptom checker and performed the Wilcoxon signed-rank test. The recall measures indicated that the accuracy was equal to 75%, if the system was adjusted to only ten results as differential diagnoses. Moreover, the Wilcoxon signed-rank test showed that there was significant difference between SEPHYRES and Isabel symptom checker (P= 0.016) so that this method is sufficiently able to improve semantic expressiveness in both professional medical diagnosis and patient decision aid systems.

Author Biographies

Ali SANAEIFAR, Payam-e-Noor University of Tehran (PNU), Tehran, Iran.

Department of Computer Engineering and Information Technology

Ahmad FARAAHI, Payam-e-Noor University of Tehran (PNU), Tehran, Iran.

Department of Computer Engineering and Information Technology

Mahmood TARA, Mashhad University of Medical Science (MUMS), Mashhad, Iran.

Department of Medical Informatics

Downloads

Published

29.12.2016

How to Cite

1.
SANAEIFAR A, FARAAHI A, TARA M. SEPHYRES 1: A Symptom Checker based on Semantic Pain Descriptors and Weight Spreading. Appl Med Inform [Internet]. 2016 Dec. 29 [cited 2024 Jun. 21];38(3-4):105-16. Available from: https://ami.info.umfcluj.ro/index.php/AMI/article/view/607

Issue

Section

Articles