TY - JOUR
T1 - Application of a temporal reasoning framework tool in analysis of medical device adverse events.
AU - Clark, Kimberly K.
AU - Sharma, Deepak K.
AU - Chute, Christopher G.
AU - Tao, Cui
PY - 2011
Y1 - 2011
N2 - The Clinical Narrative Temporal Relation Ontology (CNTRO)1 project offers a semantic-web based reasoning framework, which represents temporal events and relationships within clinical narrative texts, and infer new knowledge over them. In this paper, the CNTRO reasoning framework is applied to temporal analysis of medical device adverse event files. One specific adverse event was used as a test case: late stent thrombosis. Adverse event narratives were obtained from the Food and Drug Administration's (FDA) Manufacturing and User Facility Device Experience (MAUDE) database2. 15 adverse event files in which late stent thrombosis was confirmed were randomly selected across multiple drug eluting stent devices. From these files, 81 events and 72 temporal relations were annotated. 73 temporal questions were generated, of which 65 were correctly answered by the CNTRO system. This results in an overall accuracy of 89%. This system should be pursued further to continue assessing its potential benefits in temporal analysis of medical device adverse events.
AB - The Clinical Narrative Temporal Relation Ontology (CNTRO)1 project offers a semantic-web based reasoning framework, which represents temporal events and relationships within clinical narrative texts, and infer new knowledge over them. In this paper, the CNTRO reasoning framework is applied to temporal analysis of medical device adverse event files. One specific adverse event was used as a test case: late stent thrombosis. Adverse event narratives were obtained from the Food and Drug Administration's (FDA) Manufacturing and User Facility Device Experience (MAUDE) database2. 15 adverse event files in which late stent thrombosis was confirmed were randomly selected across multiple drug eluting stent devices. From these files, 81 events and 72 temporal relations were annotated. 73 temporal questions were generated, of which 65 were correctly answered by the CNTRO system. This results in an overall accuracy of 89%. This system should be pursued further to continue assessing its potential benefits in temporal analysis of medical device adverse events.
UR - http://www.scopus.com/inward/record.url?scp=84874209919&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84874209919&partnerID=8YFLogxK
M3 - Article
C2 - 22195199
AN - SCOPUS:84874209919
SN - 1559-4076
VL - 2011
SP - 1366
EP - 1371
JO - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium
JF - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium
ER -