TY - GEN
T1 - Signaling drug adverse effects in spontaneous reporting systems
T2 - 3rd IEEE International Conference on Multimedia Processing, Communication and Information Technology, MPCIT 2020
AU - Anup Bhat, B.
AU - Harish, S. V.
AU - Geetha, M.
N1 - Funding Information:
This work was supported by Manipal Academy of Higher Education Dr. T.M.A Pai Research Scholarship under Research Registration No. 170900117.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/12/11
Y1 - 2020/12/11
N2 - Spontaneous Reporting Systems (SRS) are one of the primary sources of the Pharmacovigilance (PV) data that collect and store the reports of drug-adverse effects. In order to signal potential Adverse Drug Reactions (ADRs) from the vast set of reports from the SRS database, Frequent Itemset Mining (FIM) using the Apriori algorithm has been employed successfully. However, FIM fails to accommodate the seriousness of the reported adverse effects and merely counts the support based on the presence or absence of the drug or adverse effect in the database. In this work, Fast High Utility Itemset Mining (FHM) has been employed to accommodate the seriousness factor as a utility measure to arrive at multi-item drug-adverse effects associations. Further, the results indicate the effectiveness of the associations found in terms of Relative Reporting Ratio (RRR) to be significantly better than those of the FIM based Apriori algorithm. A few ADRs enumerated by employing FHM have been summarized which can be taken up for further clinical investigation.
AB - Spontaneous Reporting Systems (SRS) are one of the primary sources of the Pharmacovigilance (PV) data that collect and store the reports of drug-adverse effects. In order to signal potential Adverse Drug Reactions (ADRs) from the vast set of reports from the SRS database, Frequent Itemset Mining (FIM) using the Apriori algorithm has been employed successfully. However, FIM fails to accommodate the seriousness of the reported adverse effects and merely counts the support based on the presence or absence of the drug or adverse effect in the database. In this work, Fast High Utility Itemset Mining (FHM) has been employed to accommodate the seriousness factor as a utility measure to arrive at multi-item drug-adverse effects associations. Further, the results indicate the effectiveness of the associations found in terms of Relative Reporting Ratio (RRR) to be significantly better than those of the FIM based Apriori algorithm. A few ADRs enumerated by employing FHM have been summarized which can be taken up for further clinical investigation.
UR - http://www.scopus.com/inward/record.url?scp=85101683588&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85101683588&partnerID=8YFLogxK
U2 - 10.1109/MPCIT51588.2020.9350404
DO - 10.1109/MPCIT51588.2020.9350404
M3 - Conference contribution
AN - SCOPUS:85101683588
T3 - MPCIT 2020 - Proceedings: IEEE 3rd International Conference on "Multimedia Processing, Communication and Information Technology"
SP - 107
EP - 111
BT - MPCIT 2020 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 11 December 2020 through 12 December 2020
ER -