Benchmarking CXR Foundation Models With Publicly Available MIMIC-CXR and NIH-CXR14 Datasets

arXiv — cs.CVTuesday, December 9, 2025 at 5:00:00 AM
  • Recent research benchmarks two chest X-ray (CXR) embedding models, CXR-Foundation (ELIXR v2.0) and MedImageInsight, using the publicly available MIMIC-CXR and NIH ChestX-ray14 datasets. The study employs a unified preprocessing pipeline and fixed classifiers to ensure reproducibility, revealing that MedImageInsight generally outperforms CXR-Foundation, which shows strong stability across datasets.
  • This benchmarking is significant as it provides insights into the comparative performance of advanced medical imaging models, which can enhance diagnostic accuracy and efficiency in radiology. The findings may influence future developments in medical image representation learning and clinical applications.
  • The study highlights ongoing advancements in AI for medical imaging, including the development of models like S2D-ALIGN for report generation and Multiview Masked Autoencoders for improved representation learning. These innovations reflect a broader trend towards integrating AI in healthcare, addressing challenges such as increasing imaging volumes and the need for efficient diagnostic tools.
— via World Pulse Now AI Editorial System

Was this article worth reading? Share it

Recommended apps based on your readingExplore all apps
Continue Readings
CLARIFID: Improving Radiology Report Generation by Reinforcing Clinically Accurate Impressions and Enforcing Detailed Findings
PositiveArtificial Intelligence
The introduction of CLARIFID represents a significant advancement in the automatic generation of radiology reports, addressing the challenge of delivering clinically reliable conclusions. This framework optimizes diagnostic correctness by mimicking the expert workflow, enhancing the logical flow from findings to impressions and utilizing multiple chest X-ray views for comprehensive analysis.
An Improved Ensemble-Based Machine Learning Model with Feature Optimization for Early Diabetes Prediction
PositiveArtificial Intelligence
A new ensemble-based machine learning model has been developed to enhance early diabetes prediction using the BRFSS dataset, which includes over 253,000 health records. The model employs techniques like SMOTE and Tomek Links to address class imbalance and achieves a strong ROC-AUC score of approximately 0.96 through various algorithms, including Random Forest and XGBoost.
Intrusion Detection on Resource-Constrained IoT Devices with Hardware-Aware ML and DL
PositiveArtificial Intelligence
A new hardware-aware intrusion detection system (IDS) has been proposed for Internet of Things (IoT) and Industrial IoT (IIoT) networks, focusing on optimizing machine learning models and deep neural networks to meet strict resource constraints. The system demonstrates high accuracy with minimal resource usage, achieving 95.3% accuracy with LightGBM and 97.2% with a HW-NAS-optimized CNN.