Stanford AIMI - sporedata/researchdesigneR GitHub Wiki
General description
The Stanford Artificial Intelligence in Medicine & Imaging (Stanford AIMI) Shared Datasets is a collection of de-identified annotated medical imaging data to foster transparent and reproducible collaborative research. It comprises four dataset categories:
- Bone dataset: MURA: MSK Xrays is one of the largest public radiographic image (bone X-rays) datasets.
- Chest dataset: CheXpert: Chest X-rays consists of chest radiographs of 65,240 patients who underwent a radiographic examination between October 2002 and July 2017, with their associated radiology reports. CheXphoto: Chest X-rays comprises a training set of natural photos and synthetic transformations of x-rays from randomly sampled unique patients. CheXplanation is a radiologist-annotated segmentation dataset on chest x-rays and competition for automated pathology segmentation that can also be employed for the evaluation of x-ray interpretation models. COCA- Coronary Calcium and chest CT's constitutes two datasets: 1) gated coronary CT DICOM images with corresponding coronary artery calcium segmentations and scores (xml files) 2) non-gated chest CT DICOM images with coronary artery calcium scores. CT Pulmonary Angiography is a collection of CT pulmonary angiography (CTPA) for patients susceptible to Pulmonary Embolism (PE), with slice-level PE labels and labels for PE location, RV/LV ratio, and PE type. EchoNet-Dynamic Cardiac Ultrasound constitutes over 10k cardiac ultrasound from unique patients.
- Extremity dataset: LERA - Lower Extremity Radiographs consists of images of the ankle, foot, hip, or knee, associated with 182 patients who underwent a radiographic examination between 2003 and 2014. MRNet: Knee MRI's consists of knee MRI exams and labels obtained through manual extraction from clinical reports.
- Head/Brain/Neck dataset: BrainMetShare is designed to generate and test improved methods for the detection and segmentation of brain metastases. It constitutes whole-brain MRI studies, high-resolution, multi-modal pre- and post-contrast sequences in patients with multiple brain metastasis, and ground-truth segmentation by radiologists.