Sign in Subscribe
Artificial intelligence

AI Models Show Promise in Predicting Outcomes in Spinal Metastasis

Strategic Insights

— 1 min read

A recent systematic review and meta-analysis published in the Journal of Clinical Medicine has assessed the effectiveness of artificial intelligence (AI) in predicting complications and treatment outcomes for patients with spinal metastases. The study, conducted by researchers from the University College of Medical Sciences, found that AI models showed reasonable accuracy in predicting mortality, ambulatory status, blood loss, and surgical complications in patients with spinal metastases.

Key Takeaways:

  • The study included 47 articles totaling 25,790 patients with spinal metastases.
  • The weighted average Area Under the Curve (AUC) values for training, internal validation, and external validation were 0.762, 0.876, and 0.810, respectively.
  • The Skeletal Oncology Research Group machine learning algorithms (SORG-MLAs) were externally validated the most, producing AUCs of 0.84 for 90-day and 1-year mortality.
  • Models based on radiomics showed promise in preoperative planning, especially for outcomes of radiation and concealed blood loss.
  • Most research focused on breast, lung, and prostate malignancies, limiting its applicability to less common tumors.
  • The study recommends future work on creating multimodal, hybrid models and assessing their practical applications.
  • The authors suggest that wider implementation of AI models in predicting outcomes in spinal metastasis necessitates additional validation, data standardization, and ethical and regulatory framework evaluation.
  • Prachi Dawer, a researcher at the University College of Medical Sciences, is leading the effort to develop AI models for predicting outcomes in spinal metastasis.

Statistics:

  • 47 articles included in the systematic review and meta-analysis.
  • 25,790 patients with spinal metastases included in the analysis.
  • Weighted average AUC values: 0.762 (training), 0.876 (internal validation), and 0.810 (external validation).
  • AUC values for SORG-MLAs: 0.84 (90-day mortality), 0.84 (1-year mortality).

Sources:

  • University College of Medical Sciences, New Delhi, India.
  • Journal of Clinical Medicine, published by Mdpi, Switzerland.
  • NewsRx, a news publication reporting on research and discoveries from around the world.
  • Drug Week, a news publication reporting on pharmaceutical research and developments.

Read more