Discharge Mobility After Osteoporotic Thoracolumbar Fractures Using a 2-Stage Machine Learning Prediction Model: Prospective Cohort Study.
- 1Department for Orthopedics, Trauma and Plastic Surgery, University Hospital Leipzig, Liebigstrasse 20, Leipzig, Saxony, 04103, Germany, 49 03419723000, 49 03419723009.
- 2Innovation Center Computer Assisted Surgery (ICCAS), Medizinische Fakultät, Leipzig University, Leipzig, Saxony, Germany.
Abstract
Background
Discharge mobility after osteoporotic vertebral fractures is heterogeneous and difficult to predict at admission. Machine learning offers structured approaches for multiclass functional outcome prediction, but existing models predominantly address binary end points and report global performance metrics without explicit analysis of the error structure.Objective
This study aimed to develop and evaluate a 2-stage machine learning framework with a predefined decision policy designed to reduce extreme (LOW-HIGH) misclassification for graded discharge mobility prediction in hospitalized patients with osteoporotic thoracolumbar fractures.Methods
This prospective, single-center cohort study enrolled consecutive patients aged 65 years or older admitted for acute osteoporotic thoracic or lumbar fractures (N=158). Discharge mobility was categorized as LOW (<50 m or nonambulatory), MID (50-500 m), or HIGH (>500 m or stair climbing). Clinical associations were analyzed using proportional odds ordinal regression. A predefined 2-stage architecture combined gradient-boosted trees (stage A: HIGH vs non-HIGH) and a random forest classifier (stage L: LOW vs non-LOW). The final 3-class prediction was assigned by a predefined decision policy designed to reduce clinically relevant extreme (LOW-HIGH) misclassification. All preprocessing steps, including fold-wise multiple imputation, were performed within training partitions only. The decision policy was additionally reevaluated in a nested cross-validation with fold-internal threshold selection across 10 repeated partitions.Results
Prefracture HIGH mobility was the only significant independent predictor of discharge mobility (odds ratio [OR] 3.38, 95% CI 1.68-6.80; P<.001); age, comorbidity burden, fracture grade, and treatment were not. In the prespecified analysis with fixed thresholds, performance was moderate (accuracy 0.494; balanced accuracy 0.450; macro-F₁ 0.444). Binary discrimination yielded an area under the curve of 0.693 (HIGH vs non-HIGH) and 0.625 (LOW vs non-LOW). Using the predefined decision policy, major LOW-HIGH errors were less frequent than with the prefracture mobility-only baseline (5 vs 22), and no HIGH-to-LOW misclassification occurred in the prespecified analysis. In the leakage-free nested cross-validation, macro-F₁ was 0.388 (SD 0.031) and major errors had a mean of 8.3 (SD 3.4) vs 22.5 (SD 1.1) for the baseline, with fewer major errors in all 10 partitions. HIGH-to-LOW errors had a mean of 2.6 (SD 2.2) in this analysis.Conclusions
Discharge mobility after osteoporotic thoracolumbar fractures is primarily determined by prefracture functional reserve rather than fracture severity or comorbidity. A 2-stage machine learning framework achieved moderate multiclass discrimination and reduced extreme LOW-HIGH misclassification relative to a mobility-only baseline; this reduction persisted under nested cross-validation, whereas the absence of HIGH-to-LOW errors was specific to the prespecified partition. External validation is required before clinical deployment.
Related
- Transfascial Paraspinal Versus Conventional Open Posterior Pedicle Screw Fixation for Thoracolumbar Fractures: A Randomized Controlled Trial
- Procedural Complications and Implant Cutout of the SpineJack®-System in Thoracolumbar Vertebral Fractures: A Retrospective Analysis of 204 Implants from a Level-1 Trauma Center
- Postoperative Risks in Thoracolumbar Fractures Patients with Mental Disorders: A Retrospective Cohort Study Using a Nationwide Inpatient Database
- Association Between Hounsfield Units and Vertebral Body Re-collapse 3 Months After Posterior Fixation for Thoracolumbar Burst Fractures in Working-Age Patients: A Retrospective Register-Based Cohort Study
- Effects of percutaneous kyphoplasty with concomitant postural reduction in the treatment of acute osteoporotic vertebral compression fractures
BibTeX
@article{voelker2026discharge,
title = {{Discharge Mobility After Osteoporotic Thoracolumbar Fractures Using a 2-Stage Machine Learning Prediction Model: Prospective Cohort Study.}},
author = {Anna Voelker and Dina Wilma Wiersbicki and Christoph-Eckhard Heyde and Tim Stelzner and Erik Schreiber and Thomas Neumuth and Juliane Neumann},
journal = {JMIR formative research},
year = {2026},
month = oct,
doi = {10.2196/100486},
url = {https://europepmc.org/article/MED/42955877},
note = {CC BY},
}
RIS
TY - JOUR
TI - Discharge Mobility After Osteoporotic Thoracolumbar Fractures Using a 2-Stage Machine Learning Prediction Model: Prospective Cohort Study.
AU - Anna Voelker
AU - Dina Wilma Wiersbicki
AU - Christoph-Eckhard Heyde
AU - Tim Stelzner
AU - Erik Schreiber
AU - Thomas Neumuth
AU - Juliane Neumann
PY - 2026
DA - 2026/10/29
JO - JMIR formative research
DO - 10.2196/100486
UR - https://europepmc.org/article/MED/42955877
AB - Background Discharge mobility after osteoporotic vertebral fractures is heterogeneous and difficult to predict at admission. Machine learning offers structured approaches for multiclass functional outcome prediction, but existing models predominantly address binary end points and report global performance metrics without explicit analysis of the error structure.Objective This study aimed to develop and evaluate a 2-stage machine learning framework with a predefined decision policy designed to reduce extreme (LOW-HIGH) misclassification for graded discharge mobility prediction in hospitalized patients with osteoporotic thoracolumbar fractures.Methods This prospective, single-center cohort study enrolled consecutive patients aged 65 years or older admitted for acute osteoporotic thoracic or lumbar fractures (N=158). Discharge mobility was categorized as LOW (<50 m or nonambulatory), MID (50-500 m), or HIGH (>500 m or stair climbing). Clinical associations were analyzed using proportional odds ordinal regression. A predefined 2-stage architecture combined gradient-boosted trees (stage A: HIGH vs non-HIGH) and a random forest classifier (stage L: LOW vs non-LOW). The final 3-class prediction was assigned by a predefined decision policy designed to reduce clinically relevant extreme (LOW-HIGH) misclassification. All preprocessing steps, including fold-wise multiple imputation, were performed within training partitions only. The decision policy was additionally reevaluated in a nested cross-validation with fold-internal threshold selection across 10 repeated partitions.Results Prefracture HIGH mobility was the only significant independent predictor of discharge mobility (odds ratio [OR] 3.38, 95% CI 1.68-6.80; P<.001); age, comorbidity burden, fracture grade, and treatment were not. In the prespecified analysis with fixed thresholds, performance was moderate (accuracy 0.494; balanced accuracy 0.450; macro-F₁ 0.444). Binary discrimination yielded an area under the curve of 0.693 (HIGH vs non-HIGH) and 0.625 (LOW vs non-LOW). Using the predefined decision policy, major LOW-HIGH errors were less frequent than with the prefracture mobility-only baseline (5 vs 22), and no HIGH-to-LOW misclassification occurred in the prespecified analysis. In the leakage-free nested cross-validation, macro-F₁ was 0.388 (SD 0.031) and major errors had a mean of 8.3 (SD 3.4) vs 22.5 (SD 1.1) for the baseline, with fewer major errors in all 10 partitions. HIGH-to-LOW errors had a mean of 2.6 (SD 2.2) in this analysis.Conclusions Discharge mobility after osteoporotic thoracolumbar fractures is primarily determined by prefracture functional reserve rather than fracture severity or comorbidity. A 2-stage machine learning framework achieved moderate multiclass discrimination and reduced extreme LOW-HIGH misclassification relative to a mobility-only baseline; this reduction persisted under nested cross-validation, whereas the absence of HIGH-to-LOW errors was specific to the prespecified partition. External validation is required before clinical deployment.
SN - 2561-326X
ER -