Dark Mode
Healthcare Image Feature Data
Patient Health Records & Digital Health
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Explore the vital DICOM metadata associated with the RSNA ATD Competition Dataset. DICOM (Digital Imaging and Communications in Medicine) serves as the standard format for storing and transmitting medical images, such as CT scans, X-rays, and MRIs, within healthcare environments. This dataset provides a rich collection of structured information beyond the raw image pixels, which is highly valuable for advanced analytical tasks like feature engineering. These fields are essential for correctly interpreting and displaying medical images and connecting them back to patient records and study context.
Columns
The dataset features several critical fields:
- SOP Instance UID: A unique identifier assigned globally to each individual image or instance within a DICOM study.
- Content Date/Time: Indicates the exact date and time when the image or related data was created or acquired.
- Patient ID: The unique identifier used to link various studies and images to a specific patient.
- Slice Thickness: Relevant for 3D modalities like CT scans, defining the thickness of the image slice in millimeters.
- KVP (Kilovolt Peak): The peak voltage of the X-ray machine utilized during image acquisition, influencing image contrast.
- Patient Position: Specifies the patient's position during acquisition (e.g., prone or supine).
- Study Instance UID: A unique identifier for the overall medical examination or procedure, which may encompass multiple series.
- Series Instance UID/Series Number: Identifiers for a specific group of related images within a study.
- Instance Number: The position of the image within its specific series.
- Image Position (Patient) / Image Orientation (Patient): Defines the spatial location and alignment of the image slice relative to the patient’s anatomy using coordinate parameters.
- Frame of Reference UID: Establishes a coordinate system to ensure correct alignment across multi-modality studies.
- Photometric Interpretation: Describes how pixel data should be interpreted for display (e.g., grayscale or RGB).
- Rows/Columns: The dimensions (height and width) of the image.
- Pixel Spacing: The physical size of each pixel in millimeters.
- Bits Allocated/Bits Stored/High Bit: Details relating to the bit depth and storage of pixel values.
- Pixel Representation: Indicates if pixel data is signed (often for CT scans) or unsigned (often for X-ray images).
- Window Center/Window Width: Parameters used for fine-tuning image display and contrast.
- Rescale Intercept/Rescale Slope/Rescale Type: Factors applied when converting pixel values into physical units (like Hounsfield Units).
Distribution
This product is structured as a tabular data file, typically delivered in CSV format. The original data source is derived from the RSNA ATD Competition Dataset metadata. While the full dataset structure includes thirty-two columns, the provided sample file,
test_images_dicom_meta.csv, is small (1.59 kB) and contains only a subset of columns and records for demonstration.Usage
This data product is ideally suited for several analytical applications:
- Feature Engineering: Utilizing the metadata fields as powerful input features for machine learning models.
- 3D Visualization: Creating three-dimensional reconstructions based on scan series information.
- Anomaly Detection: Identifying outliers or inconsistencies in medical imaging acquisition parameters or patient context.
Coverage
The scope of this dataset focuses strictly on medical imaging acquisition details and patient context metadata. Time coverage is indicated by the Content Date and Content Time fields. The patient positioning fields (Patient Position, Image Position, Image Orientation) provide anatomical context relevant to the medical domain.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
Intended users include data scientists, researchers, and healthcare technologists who require detailed insight into medical image acquisition parameters. Typical use cases involve linking imaging characteristics to clinical outcomes, validating imaging protocols, or building robust models that depend on contextual features beyond raw pixel values.
Dataset Name Suggestions
- RSNA ATD Imaging Metadata
- Medical DICOM Parameter Set
- Healthcare Image Feature Data
- Clinical Acquisition Details
Attributes
Original Data Source: Healthcare Image Feature Data
QUALITYLoading...