A co-learning framework for maternal and pediatric pharmacotherapy knowledge gap discovery
Journal of the American Medical Informatics Association · 2026
DOI: 10.1093/jamia/ocag130 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, 2026. Verfasst von Xiaofu Liu und 29 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Informatics, College of Medicine, The Ohio State University, Columbus, OH 43210, United States. Hauptthemen laut MEDLINE: Machine Learning, Natural Language Processing, Pharmacokinetics. Die Arbeit nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NIH HHS, Eunice Kennedy Shriver National Institute of Child Health and Human Development, the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health.
Abstract (Originalsprache)
Objective
Pregnant and breastfeeding women, along with children, represent populations persistently underrepresented in clinical research, leading to uncertainty in clinical decision-making. This study investigated the landscape of pharmacology research concerning pharmacokinetics (PK), pharmaco-epidemiology (PE), and clinical trials (CT) within maternal and pediatric patient populations. Materials and Methods A pharmacology research landscape in maternal and pediatric patient populations was analyzed using a novel co-learning framework applied to 38 million PubMed abstracts. It incorporated an evolving annotation guideline for abstract labeling, uncertain sampling, annotation error calibration, and large language model training. The co-learning framework enabled human annotators and language models to iteratively learn from each other, optimizing the classification of PK/PE/CT papers. Analysis was restricted to medications prescribed for maternal and pediatric patients, focusing on publication evidence.
Results
Co-learning significantly increased the F1-scores for classifying the CT, PE, and PK abstracts from (0.81, 0.84, 0.73) to (0.96, 0.94, 0.93), respectively. On the external validation set, the F1-score increased from (0.81, 0.60, 0.42) to (0.95, 0.88, 0.91) for CT, PE, and PK, respectively. The models identified 376 231 and 879 936 abstracts relevant to maternal and pediatric patients, respectively. In pregnant or postpartum women, 22%–66% prescribed drugs lacked either PK or PE/CT evidence. Among pediatric patients, 8%–40% prescribed drugs had no or weak PK or PE/CT evidence.
Discussion
Co-learning is a more effective approach than active learning in obtaining feedback from human annotators for optimizing the language model. With sufficiently labeled samples, a language model such as BioBERT outperformed large language models.
Conclusion
In both maternal and pediatric pharmacology research, the absence of PK evidence was notably more frequent compared to PE and CT. This significant finding calls for more pharmacokinetics studies within pediatric and maternal patient populations.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Biomedical Informatics, College of Medicine, The Ohio State University, Columbus, OH 43210, United States.
- Division of Clinical Pharmacology, Department of Medicine, School of Medicine, Indiana University, Indianapolis, IN 46202, United States.
- Department of Electrical Engineering and Computer Science and Christopher S. Bond Life Sciences Center, University of Missouri, Columbia, MO 65211, United States.
- Department of Pharmacy Practice, College of Pharmacy, Purdue University, West Lafayette, IN 47907, United States.
- Department of Pediatrics, School of Medicine, Indiana University, Indianapolis, IN 46202, United States.
- College of Pharmacy, The Ohio State University, Columbus, OH 43210, United States.
ORCID-Profile: Xiaofu Liu, Lei Wang, Saurabh Rahurkar, Lijun Cheng, Shijun Zhang, Robert Bies, Sara K Quinney, Lang Li
Förderung
- NIH HHS
- Eunice Kennedy Shriver National Institute of Child Health and Human Development
- the Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health
- NLM NIH HHS
- the National Library of Medicine of the National Institutes of Health
- NICHD NIH HHS
Zitieren
APA
Liu, X., O'Kane, A. M., Wang, L., Shendre, A., Ma, H., Zhang, Y., Jiang, H., Chiang, C. W., Zaidi, S. S., Aboshady, O. A., So, G., Tillman, E. M., Kirkpatrick, L. M., Rahurkar, S., Cheng, L., Ouyang, J., Wei, P., Zhu, Z., Li, S., …, & Li, L. (2026). A co-learning framework for maternal and pediatric pharmacotherapy knowledge gap discovery. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag130
Vancouver
Liu X, O'Kane AM, Wang L, Shendre A, Ma H, Zhang Y, et al. A co-learning framework for maternal and pediatric pharmacotherapy knowledge gap discovery. Journal of the American Medical Informatics Association. 2026. doi:10.1093/jamia/ocag130
BibTeX
@article{Liu2026,
author = {Liu, Xiaofu and O'Kane, Aislinn M and Wang, Lei and Shendre, Aditi and Ma, Hanson and Zhang, Yichuan and Jiang, Haoran and Chiang, Chien-Wei and Zaidi, Syed S and Aboshady, Omar A and So, Gerald and Tillman, Emma M and Kirkpatrick, Lindsey M and Rahurkar, Saurabh and Cheng, Lijun and Ouyang, Jiayi and Wei, Ping and Zhu, Zhimo and Li, Shangjia and Huang, Yirui and Wang, Lingling and Cao, Weidan and Liu, Jianing and Oteng, Samuel-Richard and Goodwin, Andrew and Deypalubos, Jiezel Ann Faith and Zhang, Shijun and Bies, Robert and Quinney, Sara K and Li, Lang},
title = {A co-learning framework for maternal and pediatric pharmacotherapy knowledge gap discovery},
journal = {Journal of the American Medical Informatics Association},
year = {2026},
doi = {10.1093/jamia/ocag130}
}
EUVIMED