http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Rakha Zharfarizqi Danadibrata,Da Un Jeong,Aroli Marcellinus,Ki Moo Lim 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.11
As part of the Comprehensive in vitro Proarrhythmia Assay initiative, methodologies for predicting the incidence of Torsade de Pointes (TdP) by drugs were recently developed, leading to the importance of assessment using In-silico simulations. We performed the In-silico simulation using the ventricular cell model suggested by Sara Dutta using IC50 and hill coefficient as the input that we got from the In-vitro experiment of drug-response of ion channels. From the In-silico simulation, we obtained the qNet variability according to pace. To increase drug toxicity evaluation performance, we proposed deep CNN model utilized the qNet variability as an input and classify into high-risk, intermediate-risk, and low-risk. We trained the model with 12 drugs and tested it with the remaining 16 drugs. In the high-risk , the proposed CNN model had an AUC of 0.90, 0.75 in the intermediate-risk, and 0.82 in the low-risk.