Weakly-Supervised Multimodal Learning on MIMIC-CXR
This work addresses label scarcity and multimodal integration for medical applications, but appears incremental as it evaluates an existing method on a new dataset.
The paper tackled the challenges of multimodal data integration and label scarcity in medical machine learning by evaluating the Multimodal Variational Mixture-of-Experts (MMVM) VAE on the MIMIC-CXR dataset, showing it consistently outperforms other multimodal VAEs and fully supervised approaches.
Multimodal data integration and label scarcity pose significant challenges for machine learning in medical settings. To address these issues, we conduct an in-depth evaluation of the newly proposed Multimodal Variational Mixture-of-Experts (MMVM) VAE on the challenging MIMIC-CXR dataset. Our analysis demonstrates that the MMVM VAE consistently outperforms other multimodal VAEs and fully supervised approaches, highlighting its strong potential for real-world medical applications.