Technology Category
Machine Learning: OptimizationComputer Vision: Learning & Optimization for CVSearch and Optimization: Learning to Search
Exploring the Feasibility of AI-Assisted Spine MRI Protocol Optimization Using DICOM Image Metadata
🤖 AI Summary
Multicenter MRI equipment heterogeneity and protocol variability lead to inconsistent spinal image quality and inefficient quality control (QC). Method: This study introduces the first AI-driven protocol optimization framework for spinal MRI, leveraging real-world clinical DICOM metadata to formulate a binary image quality classification task (good vs. poor). Five ensemble models—including XGBoost and LightGBM—were trained on ≥292 cases and interpreted via SHAP analysis. Results: Models achieved F1-scores of 77%–93%. Key quality determinants—such as signal-to-noise ratio and motion artifacts—were found to exhibit physically consistent relationships with acquisition parameters (e.g., TR, TE, bandwidth, number of excitations), thereby bridging AI predictions with fundamental MRI physics. The framework has been validated in real-world QC settings, delivering interpretable, clinically actionable protocol optimization guidance for medical physicists.
Technology Category
Application Category
📝 Abstract
Artificial intelligence (AI) is increasingly being utilized to optimize magnetic resonance imaging (MRI) protocols. Given that image details are critical for diagnostic accuracy, optimizing MRI acquisition protocols is essential for enhancing image quality. While medical physicists are responsible for this optimization, the variability in equipment usage and the wide range of MRI protocols in clinical settings pose significant challenges. This study aims to validate the application of AI in optimizing MRI protocols using dynamic data from clinical practice, specifically DICOM metadata. To achieve this, four MRI spine exam databases were created, with the target attribute being the binary classification of image quality (good or bad). Five AI models were trained to identify trends in acquisition parameters that influence image quality, grounded in MRI theory. These trends were analyzed using SHAP graphs. The models achieved F1 performance ranging from 77% to 93% for datasets containing 292 or more instances, with the observed trends aligning with MRI theory. The models effectively reflected the practical realities of clinical MRI settings, offering a valuable tool for medical physicists in quality control tasks. In conclusion, AI has demonstrated its potential to optimize MRI protocols, supporting medical physicists in improving image quality and enhancing the efficiency of quality control in clinical practice.
Problem
Research questions and friction points this paper is trying to address.
Validate AI for MRI protocol optimization
Use DICOM metadata to classify image quality
Support medical physicists in quality control
Innovation
Methods, ideas, or system contributions that make the work stand out.
AI optimizes MRI protocols
Uses DICOM metadata analysis
SHAP graphs interpret AI decisions
🔎 Similar Papers
No similar papers found.
