Datasets for Robust Uncertainty Quantification in Biomedical AI Under Data Scarcity

This record contains the simulated and processed datasets used in themanuscript “Robust Uncertainty Quantification in Biomedical AI UnderData Scarcity Using Conformal Prediction and Statistical Randomization”. The pneumonia feature embeddings were derived from the officialPneumoniaMNIST partitions in MedMNIST v2(https://doi.org/10.5281/zenodo.6496656;https://doi.org/10.1038/s41597-022-01721-8). The Chinese herbal medicine electronic-nose dataset was originallycollected and described in:https://doi.org/10.3390/s18092936. The processed breast cancer feature and label matrices were derivedfrom the TCGA-BRCA project:https://portal.gdc.cancer.gov/projects/TCGA-BRCA. The corresponding experimental and analysis code is available at:https://github.com/ZhouYuqi11/Robust-ICP-Biomedical-AI.