DeepIDC: A Prediction Framework of Injectable Drug Combination Based on Heterogeneous Information and Deep Learning.

Journal: Clinical pharmacokinetics
Published Date:

Abstract

BACKGROUND AND OBJECTIVE: In clinical practice, injectable drug combination (IDC) usually provides good therapeutic effects for patients. Numerous clinical studies have directly indicated that inappropriate IDC generates adverse drug events (ADEs). The clinical application of injections is increasing, and many injections lack relevant combination information. It is still a significant need for experienced clinical pharmacists to participate in evidence-based drug decision making, monitor medication safety, and manage drug interactions. Meanwhile, a large number of injection pairs and dosage combinations limit exhaustive screening. Here, we present a prediction framework, called DeepIDC, that can expediently screen the feasibility of IDCs using heterogeneous information with deep learning. This is the first specific prediction framework to identify IDCs.

Authors

  • Yuhe Yang
    College of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, Sichuan, China.
  • Dong Gao
    College of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, Sichuan, China.
  • Xueqin Xie
    College of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, Sichuan, China.
  • Jiaan Qin
    Beijing Friendship Hospital, Capital Medical University, Beijing, China.
  • Jian Li
    Fujian Key Laboratory of Traditional Chinese Veterinary Medicine and Animal Health, College of Animal Science, Fujian Agriculture and Forestry University, Fuzhou, China.
  • Hao Lin
    Yangtze Delta Region Institute (Huzhou), University of Electronic Science and Technology of China, Huzhou, Zhejiang, China.
  • Dan Yan
    Department of Pharmacy, the First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China.
  • Kejun Deng
    College of Life Science and Technology, University of Electronic Science and Technology of China, Chengdu, Sichuan, China. dengkj@uestc.edu.cn.