MedMNIST+ 2D Dataset Overview for BioFuse Framework

An overview of the MedMNIST+ 2D benchmark datasets used to evaluate the BioFuse embedding fusion framework. The dataset is 9.5 KB in size, authored by Mirza Nasir Hossain, and was last updated on March 18, 2026. It was used to test a framework that fuses embeddings from 9 state-of-the-art foundation models to achieve high performance on biomedical image classification tasks.

Use Cases

• Benchmarking biomedical image classification models based on the MedMNIST+ datasets mentioned in the description.
• Evaluating cross-modal model performance for tasks like radiology and histopathology image analysis.
• Testing embedding fusion frameworks that combine features from multiple pre-trained foundation models.
• Training downstream classifiers like XGBoost on pre-computed, fused biomedical embeddings.

Strengths

• Dataset is small at 9.5 KB, enabling quick download and inspection.
• Associated with a framework that achieved state-of-the-art AUC on 5 out of 12 benchmark datasets.
• Licensed under CC-BY-4.0, permitting open sharing and adaptation.

Limitations

• Row count and column-level documentation are unknown, limiting suitability assessment.
• The 9.5 KB size suggests it is an overview or metadata file, not the primary image datasets themselves.

Provenance

Source

Mirza Nasir Hossain via figshare.

Collection Method

Likely compiled as part of the BioFuse framework evaluation.

Freshness

Last updated 2026-03-18 17:38:48.