Development and validation of a nomogram for predicting ADL outcomes in patients undergoing subacute stroke rehabilitation based on machine learning and standard bedside clinical data: a retrospective cohort study - Takeaways - MDSpire

Development and validation of a nomogram for predicting ADL outcomes in patients undergoing subacute stroke rehabilitation based on machine learning and standard bedside clinical data: a retrospective cohort study

  • By

  • Xinye Chen

  • Juming Liu

  • Jiawei Qin

  • Xi Qin

  • Changyu Ju

  • Suchen Zhao

  • Qianqian Sun

  • June 19, 2026

  • 0 min

Share

  • 1

    A predictive model for ADL recovery in subacute stroke patients was developed using machine learning and standard clinical data.

  • 2

    The study included 270 patients in a training cohort and 165 patients in a validation cohort, focusing on ADL independence.

  • 3

    Key predictors in the final model included the Braden score, baseline Barthel Index score, and age.

  • 4

    The model achieved an AUC of 0.832 in the training cohort and 0.866 in the validation cohort, indicating strong predictive performance.

  • 5

    The resulting nomogram serves as an intuitive tool for rehabilitation physicians to assess ADL recovery potential in stroke patients.

Original Source(s)

Development and validation of a nomogram for predicting ADL outcomes in patients undergoing subacute stroke rehabilitation based on machine learning and standard bedside clinical data: a retrospective cohort study

  • by Xinye Chen, Juming Liu, Jiawei Qin, Xi Qin, Changyu Ju, Suchen Zhao, Qianqian Sun

  • June 19, 2026

Related Content

Nomogram-based prediction of hemorrhagic transformation risk integrating platelet-to-white blood cell ratio in patients with acute ischemic stroke after intravenous thrombolysis

T1 rhizotomy for post-stroke hand flexion spasticity: a preliminary report of 15 cases

Enhancing motor function after stroke: a systematic review and meta-analysis of bioelectrical feedback interventions