Yasintuncer/NIH-CXR14-BiomedCLIP-Features

NIH-CXR14-BiomedCLIP-Features数据集是由NIH Chest X-ray Dataset (NIH-CXR14)派生而来，并使用Microsoft的BiomedCLIP-PubMedBERT_256-vit_base_patch16_224模型进行处理。该数据集包含了从胸部X光图像及其对应的文本发现中提取的图像和文本特征。原始的NIH-CXR14数据集包括112,120张胸部X光图像和30,805名独特患者的疾病标签。处理后的数据集包括512维的图像特征、512维的文本特征、原始疾病标签的多标签格式、每个图像的唯一标识符、X射线的视图位置以及基于发现标签和视图位置生成的语法正确的文本提示。

The NIH-CXR14-BiomedCLIP-Features dataset is derived from the NIH Chest X-ray Dataset (NIH-CXR14) and processed using Microsofts BiomedCLIP-PubMedBERT_256-vit_base_patch16_224 model. It contains image and text features extracted from chest X-ray images and their corresponding textual findings. The original NIH-CXR14 dataset comprises 112,120 chest X-ray images with disease labels from 30,805 unique patients. The processed dataset includes 512-dimensional image features, 512-dimensional text features, multi-label format of the original disease labels, unique identifiers for each image, the view position of the X-ray, and grammatically correct text prompts generated based on the finding labels and view position.