Tatya Hanum Pramudita (1), M. Zainal Arifin (2)
General Background: Microcirculation disorders represent an early marker of chronic health conditions, yet existing detection approaches predominantly rely on invasive and resource-intensive procedures. Specific Background: Recent advances in wearable technology enable noninvasive microcirculation monitoring through Laser Doppler Flowmetry and Fluorescence Spectroscopy signals, which generate complex, nonstationary, and high-dimensional data that challenge conventional analytical methods. Knowledge Gap: Despite the proven capability of Light Gradient Boosting Machine models for wearable physiological data, limited studies have systematically combined feature selection, Bayesian hyperparameter optimization, and cohort-based validation for microcirculation condition detection using LDF-FS data. Aims: This study aims to optimize LightGBM performance for microcirculation condition detection by integrating feature importance–based selection and Bayesian hyperparameter tuning within a Stratified Group K-Fold validation framework. Results: Feature dimensionality was reduced from 34 to 22 informative variables, resulting in improved classification performance, with the optimized model achieving a ROC-AUC of 0.8632, accuracy of 88.04%, and recall of 80.00%. SHAP-based analysis identified age, body mass index, and skin temperature as dominant physiological predictors. Novelty: The study presents an integrated optimization pipeline combining feature selection, Bayesian optimization, and subject-level validation on wearable LDF-FS data. Implications: The findings support the potential of optimized LightGBM models as interpretable and reliable components of noninvasive wearable-based microcirculation monitoring systems.
• Feature selection reduced dimensionality while maintaining robust classification performance• Bayesian optimization improved sensitivity in detecting microcirculation conditions• SHAP analysis revealed dominant demographic and physiological predictors
Bayesian Optimization; Microcirculation Detection; Feature Selection; LightGBM; Wearable LDF-FS
E. V. Zharkikh, Y. I. Loktionova, A. A. Fedorovich, A. Y. Gorshkov, dan A. V. Dunaev, “Assessment of blood microcirculation changes after COVID-19 using wearable laser doppler flowmetry,” Diagnostics, vol. 13, no. 5, p. 920, 2023, doi: 10.3390/diagnostics13050920.
L. Kralj dan H. Lenasi, “Wavelet analysis of laser doppler microcirculatory signals: Current applications and limitations,” Frontiers in Physiology, vol. 13, p. 1076445, 2023, doi: 10.3389/fphys.2023.1076445.
B. Mahesh, “Machine learning algorithms – a review,” International Journal of Science and Research, vol. 9, no. 1, pp. 381–386, 2020, doi: 10.21275/SR20123143423.
M. N. Nguyen et al., “A wearable device dataset for mental health assessment using laser doppler flowmetry and fluorescence spectroscopy sensors,” arXiv preprint, arXiv:2502.00973, 2025, doi: 10.48550/arXiv.2502.00973.
M. A. Zöller dan M. F. Huber, “Benchmark and survey of automated machine learning frameworks,” Journal of Artificial Intelligence Research, vol. 70, pp. 409–472, 2021, doi: 10.1613/jair.1.12207.
J. Liu, Y. Wang, X. Zhang, dan Y. Chen, “A hybrid PSO-LightGBM model for disease prediction and feature selection,” Computational Intelligence and Neuroscience, vol. 2022, pp. 1–11, 2022, doi: 10.1155/2022/1287361.
E. Guldogan dan F. H. Yagin, “Interpretable machine learning for serum-based metabolomics in breast cancer diagnostics: Insights from multi-objective feature selection-driven LightGBM-SHAP models,” Medicina, vol. 61, no. 6, p. 1112, 2025, doi: 10.3390/medicina61061112.
Z. Noroozi, A. Orooji, dan L. Erfannia, “Analyzing the impact of feature selection methods on machine learning algorithms for heart disease prediction,” Scientific Reports, vol. 13, p. 22588, 2023, doi: 10.1038/s41598-023-49962-w.
A. Y. Aliefia dan M. I. Irawan, “Perbandingan metode Extreme Gradient Boosting (XGBoost) dan Light Gradient Boosting Machine (LightGBM) untuk mendeteksi fraud pada data klaim,” Jurnal Sains dan Seni ITS, 2025.
A. R. Raihan, A. W. Wardana, E. P. A. R., dan A. M. Rizki, “Perbandingan algoritma LightGBM dan ANN untuk menentukan kualitas anggur merah,” JATI (Jurnal Mahasiswa Teknik), 2025.
H. K. Hendra et al., “Evaluasi performa Random Forest, XGBoost, dan LightGBM dalam diagnosis dini diabetes mellitus,” Jurnal Penelitian Ilmu, 2025. [Daring]. Tersedia: https://jurnal.polsri.ac.id/index.php/jupiter/article/view/10607
L. Hakim dan A. K. Zyen, “Optimasi model klasifikasi diabetes dengan stacking pada algoritma XGBoost dan LightGBM,” JUPITER: Jurnal Penelitian Ilmu, 2025. [Daring]. Tersedia: https://jurnal.polsri.ac.id/index.php/jupiter/article/view/10499
F. A. Wallad, “Perbandingan algoritma CatBoost dan LightGBM untuk diagnosis glioma: Studi kasus pada data TCGA,” Repository AR-Raniry, 2025. [Daring]. Tersedia: https://repository.ar-raniry.ac.id/id/eprint/46922/
A. N. Royana, Y. V. Via, dan C. A. Putra, “Evaluasi kinerja LightGBM dan CatBoost untuk prediksi churn berdasarkan dataset pelanggan layanan streaming musik,” JATI (Jurnal Mahasiswa Teknik), 2025. [Daring]. Tersedia: https://www.ejournal.itn.ac.id/index.php/jati/article/download/14358/7954
E. E. Pardede, “Optimasi peramalan penjualan menu cafe menggunakan model hybrid Facebook Prophet dan LightGBM,” Repository UPN Jatim, 2025. [Daring]. Tersedia: https://repository.upnjatim.ac.id/43537/
A. P. Kasim et al., “Optimization of LightGBM model with Bayesian optimization for malware detection,” Journal Unublitar, 2025. [Daring]. Tersedia: http://journal.unublitar.ac.id/ilkomnika/index.php/ilkomnika/article/view/722
W. Hussain et al., “Ensemble genetic and CNN model-based image classification by enhancing hyperparameter tuning,” Scientific Reports, 2025.
R. Kochnev et al., “Optuna vs Code Llama: Are LLMs a new paradigm for hyperparameter tuning?” arXiv preprint, 2025. [Daring]. Tersedia: https://arxiv.org/abs/2504.06006
B. P. Lohani, A. Dagur, dan D. Shukla, “An efficient approach for diabetes classification using feature selection and hyperparameter tuning,” Recent Advances in Electrical & Electronic Engineering, 2025, doi: 10.2174/0123520965291885240315051751.
H. N. Fakhouri, S. Alawadi, F. M. Awaysheh, dan F. Hamad, “Novel hybrid success history intelligent optimizer with gaussian transformation: Application in CNN hyperparameter tuning,” Cluster Computing, 2024, doi: 10.1007/s10586-023-04161-0.