ABSTRACT
Introduction
Acute poisoning is a significant global health burden, and the causative agent is often unclear. The primary aim of this pilot study was to develop a deep learning algorithm that predicts the most probable agent a poisoned patient was exposed to from a pre-specified list of drugs.
Research design & methods
Data were queried from the National Poison Data System (NPDS) from 2014 through 2018 for eight single-agent poisonings (acetaminophen, diphenhydramine, aspirin, calcium channel blockers, sulfonylureas, benzodiazepines, bupropion, and lithium). Two Deep Neural Networks (PyTorch and Keras) designed for multi-class classification tasks were applied.
Results
There were 201,031 single-agent poisonings included in the analysis. For distinguishing among selected poisonings, PyTorch model had specificity of 97%, accuracy of 83%, precision of 83%, recall of 83%, and a F1-score of 82%. Keras had specificity of 98%, accuracy of 83%, precision of 84%, recall of 83%, and a F1-score of 83%. The best performance was achieved in the diagnosis of single-agent poisoning in diagnosing poisoning by lithium, sulfonylureas, diphenhydramine, calcium channel blockers, then acetaminophen, in PyTorch (F1-score = 99%, 94%, 85%, 83%, and 82%, respectively) and Keras (F1-score = 99%, 94%, 86%, 82%, and 82%, respectively).
Conclusion
Deep neural networks can potentially help in distinguishing the causative agent of acute poisoning. This study used a small list of drugs, with polysubstance ingestions excluded.Reproducible source code and results can be obtained at https://github.com/ashiskb/npds-workspace.git.
Acknowledgments
The authors would like to thank the American Association of Poison Control Centers’ National Poison Data System for providing data for this study.
Declaration of interest
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Reviewer disclosures
Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.
Author contributions
O. Mehrpour and C. Hoyte designed the study. O. Mehrpour, A. Biswas, A. Al Masud, contributed to machine learning coding, C. Hoyte, O. Mehrpour conducted the data collection and contributed to acquisition and interpretation. O. Mehrpour, C. Hoyte, A. Al Masud, A. Biswas, J. Schimmel, S. Nakhaee, M. Sadegh Nasr, H. Delva-Clark, and F. Goss contributed to writing the draft and revising the manuscript. All authors approved the final version of the manuscript.
Disclaimers
The American Association of Poison Control Centers (AAPCC; http://www.aapcc.org) maintains the national database of information logged by the country’s poison centers (PCs). Case records in this database are from self-reported calls: they reflect only information provided when the public or healthcare professionals report an actual or potential exposure to a substance (e.g. ingestion, inhalation, topical exposure, etc.) or request information/educational materials. Exposures do not necessarily represent a poisoning or overdose. The AAPCC is not able to completely verify the accuracy of every report made to member centers. Additional exposures may go unreported to PCs, and data referenced from the AAPCC should not be construed to represent the complete incidence of national exposures to any substance(s).
Data availability statement
Datasets are available from the corresponding author upon any reasonable request with permission of the National Poison Data System (NPDS) administrator https://www.aapcc.org/national-poison-data-system.
Supplementary material
Supplemental data for this article can be accessed online at https://doi.org/10.1080/17425255.2023.2232724.