Related Experiment Video
Updated: Sep 10, 2025
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
Artificial Intelligence Models for Predicting Outcomes in Spinal Metastasis: A Systematic Review and Meta-Analysis
Vivek Sanker1, Prachi Dawer2, Alexander Thaller3
1Department of Neurosurgery, Stanford University, Palo Alto, CA 94305, USA.
Artificial intelligence (AI) models show promise in predicting outcomes for patients with spinal metastases, including mortality and complications. Further validation and standardization are needed for wider clinical implementation.
Area of Science:
- Oncology
- Medical Informatics
- Radiology
Background:
- Spinal metastases significantly impair neurological function and quality of life.
- Personalized clinical decision-making requires accurate prognosis and outcome prediction.
- Assessing the effectiveness of artificial intelligence (AI) in predicting outcomes for spinal metastases is crucial.
Purpose of the Study:
- To systematically assess the effectiveness of AI-based models in predicting complications and treatment outcomes for adult patients with spinal metastases.
- To evaluate the accuracy and generalizability of AI models across different prediction tasks.
Main Methods:
- Comprehensive literature search across major databases (PubMed, Scopus, Web of Science, Embase, Cochrane) up to January 2025.
- Inclusion of studies utilizing AI models for outcome prediction in adult spinal metastases patients, adhering to PRISMA guidelines.
- Meta-analysis of AUC results using a random-effects model and quality assessment via PROBAST.
Main Results:
- Analysis of 47 studies with 25,790 patients revealed weighted average AUCs of 0.762 (training), 0.876 (internal validation), and 0.810 (external validation).
- Skeletal Oncology Research Group machine learning algorithms (SORG-MLAs) demonstrated consistent external validation AUCs > 0.84 for 90-day and 1-year mortality.
- Radiomics models showed potential in preoperative planning for radiation and blood loss prediction; research predominantly focused on breast, lung, and prostate cancers.
Conclusions:
- AI models demonstrate reasonable accuracy in predicting mortality, ambulatory status, blood loss, and surgical complications in spinal metastases.
- Wider implementation requires further external validation, data standardization, and evaluation of ethical/regulatory frameworks.
- Future research should focus on developing multimodal, hybrid AI models and assessing their clinical utility.
More Related Videos
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018