Context

Each DICOM file contains an array representing the pixel values of the image. However, it also contains resourceful information that could help to have a better understanding of the overall data. This dataset is the result of the extraction of all metadata contained in each DICOM file located in both train and test folders.

You can check out the notebook here

Content

train_dicom_metadata.csv : 15000 rows gathering metadata from dicom files located in train folder test_dicom_metadata.csv : 3000 rows gathering metadata from dicom files located in test folder

This dataset presents contiguous axial magnetic resonance imaging (MRI) data captured from the lower left legs of twelve 21-31 year old (young) and twelve 66-79 year old (old) healthy adult male and female participants.Data were collected using a General Electric 1.5T Signa EchoSpeed Plus MRI system, with the following parameters:- Phased-array coil- T1-weighted spin echo sequence- 4mm slice thickness- 400ms repetition time- 11ms echo time- 512 x 512 pixel resolution- 30cm field of viewData are provided as MATLAB binary files produced from the raw DICOM files obtained from the MRI instrument. These files duplicate the DICOM image data but remove any identifying patient data to protect participant privacy. Each file contains a 512 x 512 x ns matrix of 16-bit integers representing grayscale pixel intensities, where ns is the number of slices.Filenames are given in the format XXN_MRI_Data_Combined.mat, where XX is the group identifier (OF: old female; OM: old male; YF: young female; YM: young male) and N is the subject number (1-6).Also provided are output files for each MATLAB binary processed with the software shared with this dataset (https://doi.org/10.6084/m9.figshare.7223357). These include muscle boundary identification, tissue segmentation, and cross-sectional area calculations. The naming format for these files is XXN_MRI_Data_Combined_Output.mat, using the conventions as above. Full details of the variables contained in these files can be found in Output_Variables_Description.pdf.

In the dataset, some of the features are labeled “Not Normal No Lung Opacity”. This extra third class indicates that while pneumonia was determined not to be present, there was nonetheless some type of abnormality on the image and oftentimes this finding may mimic the appearance of true pneumonia.

Dicom original images: - Medical images are stored in a special format called DICOM files (*.dcm). They contain a combination of header metadata as well as underlying raw image arrays for pixel data.

File Description :-

• stage_2_train.csv - the training set. Contains patientIds and bounding box / target information.
• stage_2_sample_submission.csv - a sample submission file in the correct format. Contains patientIds for the test set. Note that the sample submission contains one box per image, but there is no limit to the number of bounding boxes that can be assigned to a given image.
• stage_2_detailed_class_info.csv - provides detailed information about the type of positive or negative class for each image.