dementiabank

• Description:

DementiaBank is a medical domain task. It contains 117 people diagnosed with Alzheimer Disease, and 93 healthy people, reading a description of an image, and the task is to classify these groups. This release contains only the audio part of this dataset, without the text features.

• Additional Documentation: Explore on Papers With Code north_east

• Homepage: https://dementia.talkbank.org/

• Source code: tfds.audio.Dementiabank

• Versions:

  • 1.0.0 (default): No release notes.

• Download size: Unknown size

• Dataset size: 17.71 GiB

• Manual download instructions: This dataset requires you to download the source data manually into download_config.manual_dir (defaults to ~/tensorflow_datasets/downloads/manual/):
  manual dir should contain 2 folders with mp3 files:

• dementia/English/Pitt/Control/cookie

• dementia/English/Pitt/Dementia/cookie

Which were downloaded from https://media.talkbank.org/dementia/English/Pitt/ This dataset requires registration for downloading.

• Auto-cached (documentation): No

• Splits:

• Feature structure:

FeaturesDict({
    'audio': Audio(shape=(None,), dtype=int64),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
    'speaker_id': string,
})

• Feature documentation:

• Supervised keys (See as_supervised doc): ('audio', 'label')

• Figure (tfds.show_examples): Not supported.

• Examples (tfds.as_dataframe):

• Citation:

@article{boller2005dementiabank,
  title={Dementiabank database guide},
  author={Boller, Francois and Becker, James},
  journal={University of Pittsburgh},
  year={2005}
}