MultiCaRe synthetically adapted for OCR

Consolidated images, labels, and splits for MultiCaRe-OCR dataset. It is adapted for benchmarking and fine-tuning for OCR (Optical Character Recognition) and contains images with overlaid textual imprints generated from image labels. This dataset is derived from the original MultiCaRe dataset: Nievas Offidani, M. (2025). MultiCaRe: An open-source clinical case dataset for medical image classification and multimodal AI applications (version 2.2) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.10079369.本数据集为MultiCaRe-OCR数据集的整合图像、标签与划分集，适用于光学字符识别（Optical Character Recognition）任务的基准测试与模型微调，数据集内的图像均带有由图像标签生成的叠加文本印记。 本数据集源自原始MultiCaRe数据集：Nievas Offidani, M.（2025）. 《MultiCaRe：面向医学图像分类与多模态人工智能应用的开源临床病例数据集（版本2.2）》[数据集]. Zenodo. https://doi.org/10.5281/zenodo.10079369.