Development of a Novel Interpretable Transformer‐Based Deep Learning Model for Predicting Postoperative Hypokalemia in Pituitary Adenomas
Journal of Evidence-Based Medicine · 2025
DOI: 10.1111/jebm.70089 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Evidence-Based Medicine, 2025. Verfasst von Zhuoyuan Li und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neurosurgery, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China. Hauptthemen laut MEDLINE: Hypokalemia, Pituitary Neoplasms, Deep Learning, Postoperative Complications. Die Arbeit nennt 32 Literaturangaben. Gefördert von: Clinical Trials from the Affiliated Drum Tower Hospital, Medical School of Nanjing University.
Abstract (Originalsprache)
Aim
Hypokalemia is a prevalent complication following pituitary adenoma (PA) surgery, adversely impacting patient prognosis. Identifying predictors of early postoperative hypokalemia is crucial for managing patients effectively. This study aims to develop an interpretable predictive model to predict postoperative hypokalemia in patients with PA and recognizes individualized significant parameters contributing to the predictive outcomes, thereby facilitating early intervention.
Methods
This retrospective cohort study investigated postoperative hypokalemia in 280 patients with PA. We developed a Transformer‐based predictive model, CliTab‐Transformer, and compared it with an XGBoost‐based model and a multilayer perceptron (MLP)‐based deep learning model. Model performance was evaluated using metrics such as accuracy, precision, recall, F1 score, and ROC/PR curves, based on five‐fold cross‐validation. Model interpretability was assessed using a novel Transformer‐Explainability method to identify significant parameters contributing to individual predictions.
Results
CliTab‐Transformer outperformed XGBoost and MLP in predicting postoperative hypokalemia, showing higher accuracy (0.836 vs. 0.775 vs. 0.771), F1 score (0.845 vs. 0.792 vs. 0.766), sensitivity (0.838 vs. 0.833 vs. 0.771), and AUC (0.835 vs. 0.807 vs. 0.676). The model's interpretability analysis revealed that the preoperative factors, including gender, hypertension, serum potassium concentration, and disease duration, are significantly predictive of postoperative hypokalemia.
Conclusions
Our model outperforms XGBoost in predicting early postoperative hypokalemia in patients with pituitary adenomas. We systematically explore attention mechanisms in clinical tabular data, demonstrating their effectiveness in capturing complex feature interactions, leading to more individualized, interpretable, and clinically meaningful insights.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Neurosurgery, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China.
- Neurosurgical Institute, Nanjing University, Nanjing, China.
Förderung
- Clinical Trials from the Affiliated Drum Tower Hospital, Medical School of Nanjing University
Zitieren
APA
Li, Z., Wang, J., Wang, Y., Peng, Z., Li, X., Chen, C., Yan, H., Jin, W., Lu, Y., Zhuang, Z., Li, W., & Hang, C. (2025). Development of a Novel Interpretable Transformer‐Based Deep Learning Model for Predicting Postoperative Hypokalemia in Pituitary Adenomas. Journal of Evidence-Based Medicine. https://doi.org/10.1111/jebm.70089
Vancouver
Li Z, Wang J, Wang Y, Peng Z, Li X, Chen C, et al. Development of a Novel Interpretable Transformer‐Based Deep Learning Model for Predicting Postoperative Hypokalemia in Pituitary Adenomas. Journal of Evidence-Based Medicine. 2025. doi:10.1111/jebm.70089
BibTeX
@article{Li2025,
author = {Li, Zhuoyuan and Wang, Jie and Wang, Yunfeng and Peng, Zheng and Li, Xiaojian and Chen, Chunlei and Yan, Huiying and Jin, Wei and Lu, Yue and Zhuang, Zong and Li, Wei and Hang, Chunhua},
title = {Development of a Novel Interpretable Transformer‐Based Deep Learning Model for Predicting Postoperative Hypokalemia in Pituitary Adenomas},
journal = {Journal of Evidence-Based Medicine},
year = {2025},
doi = {10.1111/jebm.70089}
}
EUVIMED