EUVIMEDOdkrijte EUVIMED
Open access

The ASIA Data Science Challenge: Predicting Functional and Neurological Recovery From Acute ISNCSCI Scores

J Villines, R Stirnimann, L P Lukas, O Taran, M Tuci, Y Li, C R Jutzeler, J L K Kramer, F H Geisler, D Bourbeau, R James Cotton, S C Brüningk and 2 more

Topics in Spinal Cord Injury Rehabilitation · 2026-04-01 · 1-12

DOI: 10.46292/sci25-00137 ↗

At a glance

Journal article in Topics in Spinal Cord Injury Rehabilitation, published on 2026-04-01. Written by J Villines and 13 co-authors. Institution: Department of Quality Management, Craig Hospital, Englewood, Colorado, USA. Major MEDLINE topics: Spinal Cord Injuries, Recovery of Function, Data Science. The full text is freely available in PubMed Central.

PovzetekSubjectsBibliografski podatkiCiteRelated studies

Abstract (original language)

Background

Recovery after spinal cord injury (SCI) is driven by a variety of factors, making SCI recovery prediction particularly suitable for data-driven modelling. The American Spinal Injury Association (ASIA) Engineering and Data Science Committee launched the first open SCI data science challenge focused on recovery prediction.

Objectives

Hosted on Kaggle in 2025, the challenge addressed 2 tasks using data from the Sygen clinical trial ( N = 797): (1) predicting motor status as assessed by the International Standards for Neurological Classification of SCI (ISNCSCI; score: RMSE [root mean square error] between true and predicted motor scores), and (2) predicting walking function on the 8-point modified Benzel scale (score: Spearman's ρ ).

Method

Eight teams participated in the motor status track and 14 in the walking function track. The winning solution (RMSE = 1.0) in the motor status track applied an ensemble of boosting models with multiple imputation and Bayesian hyperparameter optimization, whereas the winning solution in the walking function track (Spearman's ρ = 0.85) harnessed TabPFN, a pretrained transformer for tabular data, combined with feature engineering from early ISNCSCI assessments and auxiliary predictions of future motor scores.

Results

Both approaches demonstrated competitive performance for SCI recovery prediction from acute phase assessments, despite modest dataset size and missing data. Both winning solutions further conducted an interpretability analysis, identifying key predictors consistent with clinical understanding of SCI recovery.

Conclusion

This challenge highlights the value of publicly accessible formats, which encourage broader engagement of diverse audiences, for developing robust predictive models and provides a proof of concept for future initiatives leveraging larger, multicenter, multimodal datasets to improve personalized prognostication in SCI rehabilitation.

Subjects

MeSH terms (MEDLINE)

Spinal Cord InjuriesRecovery of FunctionData ScienceHumansAsiaWalking

Author keywords

ISNCSCIdata science challengefunctional outcomesmachine learningrecovery predictionspinal cord injury

Authors and institutions

  • Department of Quality Management, Craig Hospital, Englewood, Colorado, USA.
  • BMDS Lab, Department of Health Sciences and Technology (D-HEST), ETH Zurich, Zürich, Switzerland.
  • SIB Swiss Institute of Bioinformatics, Lausanne, Switzerland.
  • Spinal Cord Injury Center, University Hospital Balgrist, University of Zurich, Zürich, Switzerland.
  • SCAI Lab, Department of Health Sciences and Technology (D-HEST), ETH Zurich, Zürich, Switzerland.
  • SPF Swiss Paraplegic Research, Nottwil, Switzerland.

Cite

APA

Villines, J., Stirnimann, R., Lukas, L. P., Taran, O., Tuci, M., Li, Y., Jutzeler, C. R., Kramer, J. L. K., Geisler, F. H., Bourbeau, D., Cotton, R. J., Brüningk, S. C., ASIA Engineering and Data Science Committee, & ASIA Engineering and Data Science Committee (2026). The ASIA Data Science Challenge: Predicting Functional and Neurological Recovery From Acute ISNCSCI Scores. Topics in Spinal Cord Injury Rehabilitation, 1-12. https://doi.org/10.46292/sci25-00137

Vancouver

Villines J, Stirnimann R, Lukas LP, Taran O, Tuci M, Li Y, et al. The ASIA Data Science Challenge: Predicting Functional and Neurological Recovery From Acute ISNCSCI Scores. Topics in Spinal Cord Injury Rehabilitation. 2026:1-12. doi:10.46292/sci25-00137

BibTeX
@article{Villines2026,
  author = {Villines, J and Stirnimann, R and Lukas, L P and Taran, O and Tuci, M and Li, Y and Jutzeler, C R and Kramer, J L K and Geisler, F H and Bourbeau, D and Cotton, R James and Brüningk, S C and ASIA Engineering and Data Science Committee, ASIA Engineering and Data Science Committee and ASIA Engineering and Data Science Committee, ASIA Engineering and Data Science Committee},
  title = {The ASIA Data Science Challenge: Predicting Functional and Neurological Recovery From Acute ISNCSCI Scores},
  journal = {Topics in Spinal Cord Injury Rehabilitation},
  year = {2026},
  pages = {1--12},
  doi = {10.46292/sci25-00137}
}
Start for free →