duke_ultrasound

  • Description:

DukeUltrasound is an ultrasound dataset collected at Duke University with a Verasonics c52v probe. It contains delay-and-sum (DAS) beamformed data as well as data post-processed with Siemens Dynamic TCE for speckle reduction, contrast enhancement and improvement in conspicuity of anatomical structures. These data were collected with support from the National Institute of Biomedical Imaging and Bioengineering under Grant R01-EB026574 and National Institutes of Health under Grant 5T32GM007171-44. A usage example is available here.

Split Examples
'A' 1,362
'B' 1,194
'MARK' 420
'test' 438
'train' 2,556
'validation' 278
  • Feature structure:
FeaturesDict({
    'das': FeaturesDict({
        'dB': Tensor(shape=(None,), dtype=tf.float32),
        'imag': Tensor(shape=(None,), dtype=tf.float32),
        'real': Tensor(shape=(None,), dtype=tf.float32),
    }),
    'dtce': Tensor(shape=(None,), dtype=tf.float32),
    'f0_hz': tf.float32,
    'final_angle': tf.float32,
    'final_radius': tf.float32,
    'focus_cm': tf.float32,
    'harmonic': tf.bool,
    'height': tf.uint32,
    'initial_angle': tf.float32,
    'initial_radius': tf.float32,
    'probe': tf.string,
    'scanner': tf.string,
    'target': tf.string,
    'timestamp_id': tf.uint32,
    'voltage': tf.float32,
    'width': tf.uint32,
})
  • Feature documentation:
Feature Class Shape Dtype Description
FeaturesDict
das FeaturesDict
das/dB Tensor (None,) tf.float32
das/imag Tensor (None,) tf.float32
das/real Tensor (None,) tf.float32
dtce Tensor (None,) tf.float32
f0_hz Tensor tf.float32
final_angle Tensor tf.float32
final_radius Tensor tf.float32
focus_cm Tensor tf.float32
harmonic Tensor tf.bool
height Tensor tf.uint32
initial_angle Tensor tf.float32
initial_radius Tensor tf.float32
probe Tensor tf.string
scanner Tensor tf.string
target Tensor tf.string
timestamp_id Tensor tf.uint32
voltage Tensor tf.float32
width Tensor tf.uint32
  • Citation:
@article{DBLP:journals/corr/abs-1908-05782,
  author    = {Ouwen Huang and
               Will Long and
               Nick Bottenus and
               Gregg E. Trahey and
               Sina Farsiu and
               Mark L. Palmeri},
  title     = {MimickNet, Matching Clinical Post-Processing Under Realistic Black-Box
               Constraints},
  journal   = {CoRR},
  volume    = {abs/1908.05782},
  year      = {2019},
  url       = {http://arxiv.org/abs/1908.05782},
  archivePrefix = {arXiv},
  eprint    = {1908.05782},
  timestamp = {Mon, 19 Aug 2019 13:21:03 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/abs-1908-05782},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}