|
|
2020 » Papers » Volume 2 » PREECLAMPSIA AND SECONDARY ILLNESSES RECOGNITION BASED ON LAYERED WEIGHTED NETWORKS 1. PREECLAMPSIA AND SECONDARY ILLNESSES RECOGNITION BASED ON LAYERED WEIGHTED NETWORKS Authors: Marin Iuliana, Goga Nicolae Volume 2 | DOI: 10.12753/2066-026X-20-137 | Pages: 398-412 | Download PDF | Abstract
Preeclampsia is a disease which characterizes up to 10% of the pregnancies due to the elevated blood pressure that occurs during the third trimester of labor. As preeclampsia evolves into a severe condition, more factors should be taken into account. The paper proposes a three layered weighted network consisting of illnesses, symptoms and treatments. The network is used to determine the secondary illnesses of the pregnant women. The human disease and the symptom ontologies developed by the Northwestern University from the United States of America, offer information about the first two characteristics and they have been used at the creation phase of the corresponding layers. The treatments given by doctors build the third layer of the weighted network. The illnesses patterns are determined based on the three layer network. Some of the illnesses which characterize the presence of preeclampsia are constipation, cystitis, retinal detachment, lactic acidosis, hepatic coma, and the HELLP syndrome. Based on the analysis, it was proved that many patients have glomerular endotheliosis. In the severe cases of preeclampsia, hemolysis, and low platelets, a sign of HELPP syndrome, are some of the most often illnesses. Medical professionals are able to improve the existent layers of the considered network through the use of an intuitive graphical user interface. This also allows other doctors to learn and apply new principles while treating their patients who develop preeclampsia and further secondary diseases. The proposed solution enhances the illness definition, development and monitoring process, thus improving the lives of many pregnant women and their unborn children. | Keywords
Preeclampsia, Ontology, Layered weighted network, Patterns |
|
|
|