EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Use of What-if Scenarios to Help Explain Artificial Intelligence Models for Neonatal Health

Abdullah Mamun, Lawrence D. Devoe, Mark I. Evans, David W. Britt, Judith Klein-Seetharaman, Hassan Ghasemzadeh

ACM Transactions on Computing for Healthcare · 31.10.2026 · 7 (4) · 1-26

DOI: 10.1145/3814951 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Computing for Healthcare, online erschienen am 19.08.2026. Verfasst von Abdullah Mamun und 5 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: WearTech Center, an applied research center owned and operated by the Partnership for Economic Innovation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Early detection of intrapartum risks enables timely interventions to prevent or mitigate adverse labor outcomes such as cerebral palsy. However, accurate automated systems to support clinical decision-making during delivery are currently lacking. To address this gap, we propose Artificial Intelligence for Modeling and Explaining Neonatal Health (AIMEN), a deep learning framework that predicts adverse labor outcomes from maternal, fetal, obstetrical, and intrapartum factors while providing interpretable reasoning behind its predictions. AIMEN reveals how specific modifications to input variables could alter predicted outcomes, enhancing clinical insight. To address class imbalance and limited sample size, AIMEN employs Conditional Tabular Generative Adversarial Network (CTGAN) for data augmentation. This process includes synthetic data generation, and we investigate in detail properties such as relaxing feature bounds for a subset of training points to explore slightly out-of-range physiological values, and applying silhouette-score-based filtering to increase the separability of synthetic samples. AIMEN uses an ensemble of fully connected neural networks for classification and outperforms state-of-the-art models such as XGBoost, TabNet, DANet, and LightGBM, achieving an average F1 score of 0.784 in predicting high-risk deliveries. Moreover, AIMEN generates counterfactual explanations that identify actionable changes involving only two to three attributes on average. Resources: https://github.com/ab9mamun/AIMEN .

Förderung

  • WearTech Center, an applied research center owned and operated by the Partnership for Economic Innovation

Zitieren

APA

Mamun, A., Devoe, L. D., Evans, M. I., Britt, D. W., Klein-Seetharaman, J., & Ghasemzadeh, H. (2026). Use of What-if Scenarios to Help Explain Artificial Intelligence Models for Neonatal Health. ACM Transactions on Computing for Healthcare, 7(4), 1-26. https://doi.org/10.1145/3814951

Vancouver

Mamun A, Devoe LD, Evans MI, Britt DW, Klein-Seetharaman J, Ghasemzadeh H. Use of What-if Scenarios to Help Explain Artificial Intelligence Models for Neonatal Health. ACM Transactions on Computing for Healthcare. 2026;7(4):1-26. doi:10.1145/3814951

BibTeX
@article{Mamun2026,
  author = {Mamun, Abdullah and Devoe, Lawrence D. and Evans, Mark I. and Britt, David W. and Klein-Seetharaman, Judith and Ghasemzadeh, Hassan},
  title = {Use of What-if Scenarios to Help Explain Artificial Intelligence Models for Neonatal Health},
  journal = {ACM Transactions on Computing for Healthcare},
  year = {2026},
  volume = {7},
  number = {4},
  pages = {1--26},
  doi = {10.1145/3814951}
}
Kostenlos starten →