Explainable machine learning-based mortality risk stratification for older adults with COVID-19: pinpointing core immunological biomarkers and revealing dose-threshold effects - Takeaways - MDSpire

Explainable machine learning-based mortality risk stratification for older adults with COVID-19: pinpointing core immunological biomarkers and revealing dose-threshold effects

  • By

  • Lin Luo

  • Lin Wang

  • Hao Wang

  • Hui Li

  • Ting Liu

  • Sha Yu

  • May 25, 2026

  • 0 min

Share

  • 1

    A machine learning model was developed to predict mortality risk in elderly COVID-19 patients using routine hematological indicators.

  • 2

    The LGBM model outperformed other algorithms with an AUC of 0.973, demonstrating high accuracy and efficiency in mortality risk prediction.

  • 3

    Key features for mortality risk included basophil percentage, C-reactive protein, and D-dimer, with non-linear associations observed.

  • 4

    A simplified model reduced training time by over 58% while maintaining comparable accuracy and interpretability for clinical use.

  • 5

    The study highlights the potential of explainable machine learning to address unmet medical needs in managing elderly COVID-19 patients.

Original Source(s)

Explainable machine learning-based mortality risk stratification for older adults with COVID-19: pinpointing core immunological biomarkers and revealing dose-threshold effects

  • by Lin Luo, Lin Wang, Hao Wang, Hui Li, Ting Liu, Sha Yu

  • May 25, 2026

Related Content

“Who’s on Backup?”—Exploring Internal Medicine Residency Backup Coverage Systems: Findings from a National Survey

Cough Audio Recognition for Early Detection of Respiratory Diseases: Algorithm Development and Validation Study

Immune remodeling and metabolic reprogramming in chronic fatigue: insights into GPCR signaling and epigenetic regulation