Benchmarking Foundation Models with Multimodal Public Electronic Health Records
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916852423196672 |
|---|---|
| author | Yu, Kunyu Yang, Rui Liao, Jingchi Li, Siqi Li, Huitao Li, Irene Peng, Yifan Kamaleswaran, Rishikesan Liu, Nan |
| author_facet | Yu, Kunyu Yang, Rui Liao, Jingchi Li, Siqi Li, Huitao Li, Irene Peng, Yifan Kamaleswaran, Rishikesan Liu, Nan |
| contents | Foundation models have emerged as a powerful approach for processing electronic health records (EHRs), offering flexibility to handle diverse medical data modalities. In this study, we present a comprehensive benchmark that evaluates the performance, fairness, and interpretability of foundation models, both as unimodal encoders and as multimodal learners, using the publicly available MIMIC-IV database. To support consistent and reproducible evaluation, we developed a standardized data processing pipeline that harmonizes heterogeneous clinical records into an analysis-ready format. We systematically compared eight foundation models, encompassing both unimodal and multimodal models, as well as domain-specific and general-purpose variants. Our findings demonstrate that incorporating multiple data modalities leads to consistent improvements in predictive performance without introducing additional bias. Through this benchmark, we aim to support the development of effective and trustworthy multimodal artificial intelligence (AI) systems for real-world clinical applications. Our code is available at https://github.com/nliulab/MIMIC-Multimodal. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_14824 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Benchmarking Foundation Models with Multimodal Public Electronic Health Records Yu, Kunyu Yang, Rui Liao, Jingchi Li, Siqi Li, Huitao Li, Irene Peng, Yifan Kamaleswaran, Rishikesan Liu, Nan Machine Learning Artificial Intelligence Foundation models have emerged as a powerful approach for processing electronic health records (EHRs), offering flexibility to handle diverse medical data modalities. In this study, we present a comprehensive benchmark that evaluates the performance, fairness, and interpretability of foundation models, both as unimodal encoders and as multimodal learners, using the publicly available MIMIC-IV database. To support consistent and reproducible evaluation, we developed a standardized data processing pipeline that harmonizes heterogeneous clinical records into an analysis-ready format. We systematically compared eight foundation models, encompassing both unimodal and multimodal models, as well as domain-specific and general-purpose variants. Our findings demonstrate that incorporating multiple data modalities leads to consistent improvements in predictive performance without introducing additional bias. Through this benchmark, we aim to support the development of effective and trustworthy multimodal artificial intelligence (AI) systems for real-world clinical applications. Our code is available at https://github.com/nliulab/MIMIC-Multimodal. |
| title | Benchmarking Foundation Models with Multimodal Public Electronic Health Records |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2507.14824 |