Enterprise AI Analysis
Clinical validation of an Artificial Intelligence software for bone age assessment based on Greulich and Pyle method in a Portuguese paediatric cohort
This study evaluates BoneAge™, an AI software for bone age (BA) assessment based on the Greulich and Pyle method, in a Portuguese paediatric cohort. Findings show high reliability (r=0.98), strong agreement (MAE/RMSE 4.9 months), and a small underestimation bias (MD -2.55 months), more evident in males and non-adolescents. The AI demonstrates high specificity (96.2%) for ruling out healthy cases but limited sensitivity (67.6%) for detecting pathological cases. It is reliable for BA prediction but needs further multi-centre studies for broader applicability.
Executive Impact Summary
The BoneAge™ AI software offers a reliable method for bone age (BA) assessment in paediatric patients, demonstrating strong correlation with traditional methods and consistent performance. While it shows a slight underestimation bias, particularly in males and non-adolescents, its high specificity is crucial for efficiently identifying healthy cases. However, its lower sensitivity suggests limitations in correctly identifying all pathological cases, indicating a need for caution and further validation in diverse clinical settings. This technology can significantly enhance diagnostic workflow by reducing subjectivity and improving standardization, but its real-world applicability benefits from multi-centre validation to confirm its robustness across varied populations.
Correlation with Ground Truth
Mean Absolute Error
Specificity
Accuracy
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
The AI software demonstrates high reliability for BA assessment with a strong correlation (r=0.98) and consistent performance (MAE/RMSE of 4.9 months). A small underestimation bias (mean difference of -2.55 months) was observed, more evident in males and non-adolescents.
The algorithm achieved 67.6% sensitivity, 96.2% specificity, 92.4% positive predictive value, and 81.4% negative predictive value, with an overall accuracy of 84.6%. It's effective for ruling out healthy cases but has limitations in identifying all pathological cases.
Underestimation bias was present in both sexes (more evident in males, MD -4.06 months) and in non-adolescents (MD -3.32 months). No significant bias was found in adolescents.
The AI tool integrates into existing workflows, automatically processing DICOM images and presenting results as an additional series without on-premise hardware, significantly reducing manual assessment time.
Enterprise Process Flow
Retrospective Radiograph Collection (2022-2023)
→
Expert Rater Consensus (Greulich & Pyle)
→
Exclusion Criteria Applied (Age/Quality)
→
AI Software Analysis (BoneAge™)
→
Statistical Validation & Subgroup Analysis
Small sample size, retrospective single-center design, and imbalanced data limit generalizability. AI performance at age extremes is uncertain. Future research should focus on multi-center prospective studies with larger cohorts, evaluating AI-assisted human performance rather than direct AI vs. human comparisons.
Calculate Your Potential ROI with AI
Estimate the efficiency gains and cost savings AI can bring to your specific operations based on industry benchmarks and operational data.
Annual Cost Savings
$0
Annual Hours Reclaimed
0
Your AI Implementation Roadmap
A typical journey to integrate AI solutions within an enterprise involves strategic planning, careful execution, and continuous optimization.
Discovery & Strategy
Comprehensive analysis of current workflows, identification of AI opportunities, feasibility studies, and development of a tailored AI strategy with clear KPIs.
Pilot & Integration
Development and deployment of a pilot AI solution, integration with existing systems, user training, and initial performance validation against defined benchmarks.
Scaling & Optimization
Full-scale deployment of the AI solution across relevant departments, ongoing monitoring, performance optimization, and continuous improvement based on real-world data.
Ready to Transform Your Enterprise with AI?
Connect with our experts to explore how this AI technology, or a customized solution, can drive significant value and efficiency in your operations.
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