Healthcare
Predicting proximal junctional failure in adult spinal deformity patients using machine learning models based on spinal alignment parameters
By Akihiko Hiyama, Daisuke Sakai, Hiroyuki Katoh, Masato Sato & Masahiko Watanabe • Scientific Reports • Published online: 20 November 2025
Executive Impact: AI in Spinal Deformity Prediction
This study evaluated five machine learning (ML) models (Random Forest, Logistic Regression, Support Vector Machine (SVM), Decision Tree, and Naive Bayes) to predict Proximal Junctional Failure (PJF) in Adult Spinal Deformity (ASD) patients after two-stage corrective surgery. Using preoperative and postoperative spinal alignment parameters, Random Forest achieved the highest mean accuracy (78.4%) and Area Under the Curve (AUC = 0.704). The predicted probabilities for PJF were significantly higher in the PJF group (0.306 ± 0.181) compared to non-PJF cases (0.186 ± 0.164, p = 0.0057). The model demonstrated robustness with cross-validation (fivefold: 79.4%, tenfold: 77.3%). The findings suggest Random Forest can be a reliable tool for early PJF risk stratification. Future work includes incorporating bone mineral density, comorbidities, and multicenter validation.
0
Random Forest Accuracy
0
PJF Prediction AUC
0
Patients in Cohort
0
PJF Probability Difference
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
In the healthcare sector, AI can revolutionize diagnostics, treatment planning, and predictive analytics. This study demonstrates how machine learning, specifically Random Forest, can identify patients at high risk for Proximal Junctional Failure (PJF) after adult spinal deformity surgery. By leveraging spinal alignment parameters, AI models can support clinicians in making data-driven decisions, leading to personalized risk stratification and improved patient outcomes.
78.4%
Random Forest Model Accuracy for PJF Prediction
| Model | Mean Accuracy (%) | Key Strengths |
|---|---|---|
| Random Forest | 78.4 |
|
| SVM | 73.4 |
|
| Logistic Regression | 70.0 |
|
| Decision Tree | 67.2 |
|
| Naive Bayes | 64.0 |
|
Enterprise Process Flow
Retrospective Cohort (92 ASD Patients)
→
Radiographic Parameter Measurement (Pre/Post-op)
→
Feature Selection (Univariate t-tests & RF Importance)
→
ML Model Training & Testing (5 Models, 80:20 Split)
→
Performance Evaluation (Accuracy, AUC, Precision, Recall, F1-score)
→
Random Forest for PJF Risk Stratification
Significant (p=0.0057)
Difference in Predicted PJF Probability (PJF vs Non-PJF Groups)
Random Forest in Clinical Decision Support
The Random Forest model demonstrated a higher net benefit compared to 'treat-all' and 'treat-none' strategies across clinically relevant threshold probabilities (0.10 to 0.40). This indicates its potential to guide interventions, identifying high-risk patients for enhanced postoperative surveillance and minimizing unnecessary treatment for low-risk individuals. The model's capacity to distinguish between PJF and non-PJF groups, with significantly higher predicted probabilities for the PJF group, underscores its clinical utility for personalized risk stratification.
Project Your Enterprise ROI
Estimate the potential savings and reclaimed productivity your organization could achieve by integrating advanced AI solutions.
Projected Annual Savings
$0
Annual Hours Reclaimed
0
Your AI Implementation Roadmap
A typical enterprise AI integration follows a structured approach to ensure successful deployment and measurable impact.
Phase 1: Discovery & Strategy
Comprehensive assessment of your current operations, identification of AI opportunities, and development of a tailored AI strategy aligned with your business objectives.
Phase 2: Pilot & Proof-of-Concept
Deployment of a small-scale AI pilot project to validate the technology, demonstrate initial ROI, and gather feedback for optimization. This phase ensures feasibility and builds internal confidence.
Phase 3: Integration & Scaling
Seamless integration of AI solutions into your existing enterprise systems and workflows. Scaling proven pilots across relevant departments and processes for wider impact.
Phase 4: Optimization & Future-Proofing
Continuous monitoring, performance optimization, and iterative improvement of AI models. Staying ahead with ongoing research and development to leverage emerging AI advancements.
Ready to Transform Your Operations?
Book a complimentary strategy session with our AI experts to explore how these insights can drive efficiency and innovation in your enterprise.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat