Leakage- and Label-Aware Benchmarking of Class-Imbalance Strategies for Multimodal Chest X-Ray and EHR Classification

This paper benchmarks strategies for handling class imbalance in multimodal AI models that combine chest X-ray images and electronic health records (EHR) for patient classification. The research identifies and mitigates data leakage issues and label artifacts that can artificially inflate model performance, providing validated methods for building more reliable clinical AI systems.

This paper benchmarks strategies for handling class imbalance in multimodal AI models that combine chest X-ray images and electronic health records (EHR) for patient classification. The research identifies and mitigates data leakage issues and label artifacts that can artificially inflate model performance, providing validated methods for building more reliable clinical AI systems. The findings directly apply to healthcare organizations developing diagnostic decision-support tools.

High-quality data pipelines remain the most consistent bottleneck in enterprise AI maturity. QuettaMinds helps clients close the gap between raw data assets and production-ready AI inputs.