Abstract
Patients with myocardial infarction require continuous monitoring of multiple laboratory parameters after percutaneous coronary intervention (PCI), but assessment of their dynamic changes and of whether they deviate from the typical recovery trajectory still relies largely on clinical experience, and objective methods remain lacking. This study included 183 patients with myocardial infarction who underwent PCI, constructed fixed three-time-point windows, used blinded expert review as the reference, compared dynamic isolation forest (DIF) with other unsupervised methods, and conducted an external supportive prognostic analysis in MIMIC-IV. The results showed that DIF had the best agreement with expert ratings (Spearman’s ρ = 0.585; Kendall’s τ = 0.452), with an AUC of 0.859, and overall outperformed the other methods; the MIMIC-IV analysis showed that higher DIF anomaly scores were associated with an increased risk of death. DIF can be used to identify abnormal recovery windows after PCI that deviate from the typical recovery trajectory and may have potential for risk stratification, although its clinical utility still requires further validation.
Similar content being viewed by others
Funding
This study was supported by the Natural Science Foundation Project of Gansu Province [Grant No. 24JRRA705], Key Research and Development Project of Gansu Province [Grant No. 21YF1FA178], and Scientific Research Project of the Gansu Provincial Health Industry [Grant No. GSWSQN2023-06].
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
According to China’s Measures for the Ethical Review of Life Science and Medical Research Involving Human Beings and the relevant institutional requirements of our hospital, retrospective studies using de-identified data generated during prior routine clinical care may be exempt from ethics review and the requirement for informed consent, provided that the subjects cannot be identified, no additional intervention is involved, and no personal privacy or commercial interests are implicated.External validation data were sourced from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, a publicly available dataset derived from electronic health records at Beth Israel Deaconess Medical Center. The corresponding author, Zhang Yu, completed an online course from the National Institutes of Health and passed the Human Research Participant Protection Training (ID: 66963714). The Institutional Review Board (IRB) of the Massachusetts Institute of Technology approved our use of this database. As our research utilized anonymized data from the database, the requirement for informed consent was waived, and this study adheres to the ethical standards outlined in the Declaration of Helsinki.
AI and AI-assisted technologies in authorship declaration
During the preparation of this manuscript, the authors used ChatGPT for grammar correction and language polishing. After employing this tool, the authors carefully reviewed and edited the content as needed and take full responsibility for the integrity and accuracy of the manuscript.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhang, Y., Gao, S., Xu, H. et al. Dynamic isolation forest for anomaly detection in post-PCI myocardial infarction patients. Sci Rep (2026). https://doi.org/10.1038/s41598-026-54390-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-54390-7
