跳到正文
原文
Nature 健康护理主题·· 14 天前AI 评分30

适应性选择 MICE 算法参数用于肺功能值预测模型

Adaptive selection of MICE algorithm parameters for pulmonary function value prediction models

AI 导读

该研究评估了在随机森林链式插补过程中使用固定迭代预算的可解释初始化规则。通过596名患者和2,112次访问的数据,研究比较了12种插补方法,发现适应性MICE方法在某些条件下表现较好,但并非在所有条件下都优于其他方法。研究结果表明,直接插补准确性和下游预测性能应分别评估。

正文

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

Artificial intelligence-based medical prediction models are often affected by incomplete longitudinal records. We evaluated an interpretable initialization rule with fixed iteration budgets in a random-forest chained imputation procedure for pulmonary-function forecasting. The source cohort comprised 596 patients and 2,112 visits. A controlled cohort of 367 patients and 396 eligible prediction windows was evaluated using patient-disjoint five-fold splits. PFT cells were masked under MCAR and observed-covariate-dependent MAR mechanisms at 5%, 10%, and 20%, and every imputer and scaler was fitted on training rows only. The adaptive configuration used fixed variable-level initialization assignments. A fold-wise sensitivity audit recalculated the rule statistics and CV threshold from training patients only; all five folds retained the same assignments, so the imputed inputs and reported downstream results were unchanged. Twelve imputation approaches were compared by direct masked-cell accuracy and downstream performance using a fixed latest LSTM model. Across 360 latest-LSTM runs, the lowest-RMSE method depended on the condition: mean-initialized MICE ranked first in four conditions, adaptive MICE in one, and LOCF in one. None of 66 paired comparisons against the variable-specific adaptive configuration remained significant after Holm correction. The adaptive procedure is therefore reproducible and interpretable but was not universally superior. Direct imputation fidelity and downstream predictive utility should be evaluated separately.

Funding

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2022-NR072190).

Author information

Authors and Affiliations

  1. Department of Computer Engineering, Daejeon University, Daejeon, 34520, Republic of Korea

    Seongjae Im, Yeonghui Gang & Hongjun Kim

  2. Department of Physical Therapy, Daejeon University, Daejeon, 34520, Republic of Korea

    Myung-Mo Lee

  3. Department of Computer Engineering and Computer Science, California State University Long Beach, Long Beach, CA, 90840, USA

    Jucheol Moon

Authors

  1. Seongjae Im
  2. Yeonghui Gang
  3. Myung-Mo Lee
  4. Jucheol Moon
  5. Hongjun Kim

Corresponding author

Correspondence to Hongjun Kim.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. 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-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Im, S., Gang, Y., Lee, MM. et al. Adaptive selection of MICE algorithm parameters for pulmonary function value prediction models. Sci Rep (2026). https://doi.org/10.1038/s41598-026-78761-2

Download citation

  • Received: 23 June 2026

  • Accepted: 28 October 2026

  • Published: 06 November 2026

  • DOI: https://doi.org/10.1038/s41598-026-78761-2

来源:Nature 健康护理主题 · nature.com