dicom_source
Module containing DICOMSource class.
DICOMSource class handles loading of DICOM data.
Classes
DICOMSource
class DICOMSource( path: Union[os.PathLike, str], file_extension: Optional[str] = '.dcm', images_only: bool = False, infer_class_labels_from_filepaths: bool = False, data_cache: Optional[DataPersister] = None, output_path: Optional[Union[os.PathLike, str]] = None, iterable: bool = False, fast_load: bool = False, strict: bool = False, cache_images: bool = True, file_creation_cutoff_date: Optional[Union[Date, DateTD]] = None, file_modification_cutoff_date: Optional[Union[Date, DateTD]] = None, min_file_size: Optional[float] = None, max_file_size: Optional[float] = None, partition_size: int = 100, data_splitter: Optional[DatasetSplitter] = None, seed: Optional[int] = None, modifiers: Optional[Dict[str, DataPathModifiers]] = None, ignore_cols: Optional[Union[str, Sequence[str]]] = None,):
Data source for loading DICOM files.
Arguments
- **
**kwargs
**: Keyword arguments passed to the parent base classes. cache_images
: Whether to cache images in the file system. Defaults to True. This is ignored iffast_load
is True.data_cache
: A DataPersister instance to use if data caching is desired.data_splitter
: Approach used for splitting the data into training, test, validation. Defaults to None.fast_load
: Whether the data will be loaded in fast mode. This is used to determine whether the data will be iterated over during set up for schema generation and splitting (where necessary). Only relevant ifiterable
is True, otherwise it is ignored. Defaults to False.file_creation_cutoff_date
: The oldest possible date to consider for file creation. If None, all files will be considered. Defaults to None.file_extension
: The file extension of the DICOM files. Defaults to '.dcm'.file_modification_cutoff_date
: The oldest possible date to consider for file modification. If None, all files will be considered. Defaults to None.ignore_cols
: Column/list of columns to be ignored from the data. Defaults to None.images_only
: If True, only dicom files containing image data will be loaded. If the file does not contain any image data, or it does but there was an error loading or saving the image(s), the whole file will be skipped.infer_class_labels_from_filepaths
: Whether class labels should be added to the data based on the filepath of the files. Defaults to the first directory withinself.path
, but can go a level deeper if the datasplitter is provided withinfer_data_split_labels
set to trueiterable
: Whether the data source is iterable. This is used to determine whether the data source can be used in a streaming context during a task. Defaults to False.max_file_size
: The maximum file size in megabytes to consider. If None, all files will be considered. Defaults to None.min_file_size
: The minimum file size in megabytes to consider. If None, all files will be considered. Defaults to None.modifiers
: Dictionary used for modifying paths/ extensions in the dataframe. Defaults to None.output_path
: The path where to save intermediary output files. Defaults to 'preprocessed/'.partition_size
: The size of each partition when iterating over the data.path
: The path to the directory containing the DICOM files.seed
: Random number seed. Used for setting random seed for all libraries. Defaults to None.strict
: Whether File loading should be strictly done on files with the explicit file extension provided. If set to True will only load those files in the dataset. Otherwise, it will scan the given path for files of the same type as the provided file extension. Only relevant iffile_extension
is provided. Defaults to False.
Attributes
data
: A Dataframe-type object which contains the data.data_splitter
: Approach used for splitting the data into training, test, validation.seed
: Random number seed. Used for setting random seed for all libraries.
Ancestors
Variables
-
data : pandas.core.frame.DataFrame
- A property containing the underlying dataframe if the data has been loaded.If the datasource is iterable, this will raise an exception.
Raises: IterableDataSourceError: If the datasource is set to iterable. DataNotLoadedError: If the data has not been loaded yet.
file_names
- Returns a cached list of file names in the directory.
-
hash : str
- The hash associated with this BaseSource.This is the hash of the static information regarding the underlying DataFrame, primarily column names and content types but NOT anything content-related itself. It should be consistent across invocations, even if additional data is added, as long as the DataFrame is still compatible in its format.
Returns: The hexdigest of the DataFrame hash.
is_initialised : bool
- Checks ifBaseSource
was initialised.
is_task_running : bool
- Returns True if a task is running.
-
iterable : bool
- Defines whether the data source is iterable.This is defined by the user when instantiating the class.
-
multi_table : bool
- This returns False if the DataSource does not subclassMultiTableSource
.However, this property must be re-implemented in
MultiTableSource
, therefore it is not necessarily True if the DataSource inherits fromMultiTableSource
.
-
path : pathlib.Path
- Resolved absolute path to data.Provides a consistent version of the path provided by the user which should work throughout regardless of operating system and of directory structure.
-
selected_file_names : List[str]
- Returns a list of selected file names.Selected file names are affected by the
selected_file_names_override
andnew_file_names_only
attributes.
-
stale : bool
- Whether the data source is stale.This is defined by whether the data is loaded and the number of files matches the number of rows in the dataframe.
Static methods
get_num_workers
def get_num_workers(file_names: List[str]) ‑> int:
Inherited from:
FileSystemIterableSourceInferrable.get_num_workers :
Gets the number of workers to use for multiprocessing.
Ensures that the number of workers is at least 1 and at most equal to MAX_NUM_MULTIPROCESSING_WORKERS. If the number of files is less than MAX_NUM_MULTIPROCESSING_WORKERS, then we use the number of files as the number of workers. Unless the number of machine cores is also less than MAX_NUM_MULTIPROCESSING_WORKERS, in which case we use the lower of the two.
Arguments
file_names
: The list of file names to load.
Returns The number of workers to use for multiprocessing.
Methods
clear_file_names_cache
def clear_file_names_cache(self) ‑> None:
Inherited from:
FileSystemIterableSourceInferrable.clear_file_names_cache :
Clears the list of selected file names.
This allows the datasource to pick up any new files that have been added to the directory since the last time it was cached.
get_column
def get_column( self: BaseSource, col_name: str, *args: Any, **kwargs: Any,) ‑> Union[numpy.ndarray, pandas.core.series.Series]:
Inherited from:
FileSystemIterableSourceInferrable.get_column :
Loads and returns single column from the dataset.
Arguments
col_name
: The name of the column which should be loaded.**kwargs
: Additional keyword arguments to pass to theload_data
method if the data is stale.
Returns The column request as a series.
get_column_names
def get_column_names(self, **kwargs: Any) ‑> Iterable[str]:
Inherited from:
FileSystemIterableSourceInferrable.get_column_names :
Get the column names as an iterable.
get_data
def get_data( self, file_names: Optional[List[str]] = None, use_cache: bool = True, **kwargs: Any,) ‑> Optional[pandas.core.frame.DataFrame]:
Inherited from:
FileSystemIterableSourceInferrable.get_data :
Returns data if the datasource is not iterable, otherwise None.
We don't reload the data if it has already been loaded. We are assuming that files are only added, not removed, meaning we can rely on simply matching the number of files detected with the length of the dataframe to ascertain whether the data has already been loaded.
get_dtypes
def get_dtypes(self: BaseSource, *args: Any, **kwargs: Any) ‑> _Dtypes:
Inherited from:
FileSystemIterableSourceInferrable.get_dtypes :
Loads and returns the column names and types of the dataframe.
If fast_load
is set to True and the data is stale, only the first file will
be loaded to get the column names and types. Otherwise, all files will be
loaded. We only perform this optimisation if the data is stale because if it's
not, it's faster to just return the dtypes of the already loaded data.
Arguments
**kwargs
: Additional keyword arguments to pass to theload_data
method if the data is stale.
Returns A mapping from column names to column types.
get_file_creation_date
def get_file_creation_date(self, path: str) ‑> datetime.date:
Inherited from:
FileSystemIterableSourceInferrable.get_file_creation_date :
Get the creation date of a file with consideration for different OSs.
It is not possible to get the creation date of a file on Linux. This method will return the last modification date instead. This will impact filtering of files by date.
Arguments
path
: The path to the file.
Returns The creation date of the file.
get_file_last_modification_date
def get_file_last_modification_date(self, path: str) ‑> datetime.date:
Inherited from:
FileSystemIterableSourceInferrable.get_file_last_modification_date :
Get the last modification date of a file.
Arguments
path
: The path to the file.
Returns The last modification date of the file.
get_values
def get_values(self, col_names: List[str], **kwargs: Any) ‑> Dict[str, Iterable[Any]]:
Inherited from:
FileSystemIterableSourceInferrable.get_values :
Get distinct values from columns in the dataset.
Arguments
col_names
: The list of the columns whose distinct values should be returned.**kwargs
: Additional keyword arguments to pass to theload_data
method if the data is stale.
Returns The distinct values of the requested column as a mapping from col name to a series of distinct values.
load_data
def load_data(self, **kwargs: Any) ‑> None:
Inherited from:
FileSystemIterableSourceInferrable.load_data :
Load the data for the datasource.
Arguments
- **
**kwargs
**: Keyword arguments to be passed to underlying functions.
Raises
TypeError
: If data format is not supported.
partition
def partition(self, iterable: Sequence, partition_size: int = 1) ‑> Iterable:
Inherited from:
FileSystemIterableSourceInferrable.partition :
Takes an iterable and yields partitions of size partition_size
.
The final partition may be less than size partition_size
due to the variable
length of the iterable.
process_sequence_field
def process_sequence_field(self, elem: _DICOMSequenceField) ‑> Optional[Dict[str, Any]]:
Process a sequence field.
This method is called when a sequence field is encountered. It can be overridden by plugins to process specific sequence data.
Override this method in your plugin if you want to process specific sequence data.
Arguments
elem
: The DICOM data element which has its 'VR' set to 'SQ'.
Returns A dictionary containing the processed sequence data or None.
use_file_multiprocessing
def use_file_multiprocessing(self, file_names: List[str]) ‑> bool:
Inherited from:
FileSystemIterableSourceInferrable.use_file_multiprocessing :
Check if file multiprocessing should be used.
Returns True if file multiprocessing has been enabled by the environment variable and the number of workers would be greater than 1, otherwise False. There is no need to use file multiprocessing if we are just going to use one worker - it would be slower than just loading the data in the main process.
Returns True if file multiprocessing should be used, otherwise False.
yield_data
def yield_data( self, file_names: Optional[List[str]] = None, use_cache: bool = True, **kwargs: Any,) ‑> Iterator[pandas.core.frame.DataFrame]:
Inherited from:
FileSystemIterableSourceInferrable.yield_data :
Yields data in batches from files that match the given file names.
Arguments
file_names
: An optional list of file names to use for yielding data. Otherwise, all files that have already been found will be used.file_names
is always provided when this method is called from the Dataset as part of a task.
Raises
ValueError
: If no file names provided and no files have been found.