External validation of a machine learning prediction model for massive blood loss during surgery for spinal metastases: a multi-institutional study using 880 patients
To our knowledge, this is the first published external validation of a predictive ML model within orthopedic surgery to demonstrate poor performance. This poor performance might be attributed to overfitting and sampling bias as the development cohort had an insufficient sample size, and distributional shift as our cohort had key differences in predictive variables used by the model. These findings emphasize the importance of extensive validation in different geographical areas and addressing biases and pitfalls of ML model development before clinical implementation, as untested models may do more harm than good.
Authors
- Jorrit‐Jan Verlaan (ORCID: https://orcid.org/0000-0001-8105-6660)
- Darren F. Lui (ORCID: https://orcid.org/0000-0003-2121-6786)
- Diego Agustín Abelleyra Lastoria (ORCID: https://orcid.org/0000-0002-1901-8214)
- R. H. Kuijten
- Chris Taylor (ORCID: https://orcid.org/0000-0001-7867-9533)
- Daniel G. Tobert (ORCID: https://orcid.org/0000-0001-6168-0175)
- Olivier Q. Groot (ORCID: https://orcid.org/0000-0002-5091-3110)
- Priyanshu Saha (ORCID: https://orcid.org/0000-0003-0054-5652)
- Daniël C. de Reus
- A Warr-Esser
Institutions
- Harvard University (US)
- Utrecht University (NL)
- St George’s University Hospitals NHS Foundation Trust (GB)
- University Medical Center Utrecht (NL)
- Massachusetts General Hospital (US)
Publication Details
- Journal
- The Spine Journal
- Published
- 2025-03-29
- DOI
- https://doi.org/10.1016/j.spinee.2025.03.018
- Citations
- 6
- Primary Topic
- Medical Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 4.04
Funders
- Uppsala Monitoring Centre
- Universitair Medisch Centrum Utrecht