Header menu link for other important links
X
Decision Fusion-Based Fetal Ultrasound Image Plane Classification Using Convolutional Neural Networks
Ramarathnam KrishnaKumar
Published in Elsevier USA
2019
PMID: 30826153
Volume: 45
   
Issue: 5
Pages: 1259 - 1273
Abstract
Machine learning for ultrasound image analysis and interpretation can be helpful in automated image classification in large-scale retrospective analyses to objectively derive new indicators of abnormal fetal development that are embedded in ultrasound images. Current approaches to automatic classification are limited to the use of either image patches (cropped images) or the global (whole) image. As many fetal organs have similar visual features, cropped images can misclassify certain structures such as the kidneys and abdomen. Also, the whole image does not encode sufficient local information about structures to identify different structures in different locations. Here we propose a method to automatically classify 14 different fetal structures in 2-D fetal ultrasound images by fusing information from both cropped regions of fetal structures and the whole image. Our method trains two feature extractors by fine-tuning pre-trained convolutional neural networks with the whole ultrasound fetal images and the discriminant regions of the fetal structures found in the whole image. The novelty of our method is in integrating the classification decisions made from the global and local features without relying on priors. In addition, our method can use the classification outcome to localize the fetal structures in the image. Our experiments on a data set of 4074 2-D ultrasound images (training: 3109, test: 965) achieved a mean accuracy of 97.05%, mean precision of 76.47% and mean recall of 75.41%. The Cohen κ of 0.72 revealed the highest agreement between the ground truth and the proposed method. The superiority of the proposed method over the other non-fusion-based methods is statistically significant (p < 0.05). We found that our method is capable of predicting images without ultrasound scanner overlays with a mean accuracy of 92%. The proposed method can be leveraged to retrospectively classify any ultrasound images in clinical research. © 2018 World Federation for Ultrasound in Medicine & Biology
About the journal
JournalData powered by TypesetUltrasound in Medicine and Biology
PublisherData powered by TypesetElsevier USA
ISSN03015629
Open AccessNo
Concepts (49)
  •  related image
    Classification (of information)
  •  related image
    Clinical research
  •  related image
    Convolution
  •  related image
    Learning systems
  •  related image
    Neural networks
  •  related image
    Statistical tests
  •  related image
    Ultrasonics
  •  related image
    Automatic classification
  •  related image
    CLASSIFICATION DECISION
  •  related image
    Convolutional neural network
  •  related image
    DECISION FUSION
  •  related image
    FETAL ULTRASOUND
  •  related image
    FETAL ULTRASOUND IMAGES
  •  related image
    SELECTIVE SEARCH
  •  related image
    ULTRASOUND IMAGE ANALYSIS
  •  related image
    Image classification
  •  related image
    Algorithm
  •  related image
    Article
  •  related image
    Artificial neural network
  •  related image
    Body mass
  •  related image
    Classifier
  •  related image
    Controlled study
  •  related image
    Diagnostic accuracy
  •  related image
    Fetus
  •  related image
    Fetus echography
  •  related image
    Human
  •  related image
    Image processing
  •  related image
    Image segmentation
  •  related image
    INFORMED CONSENT
  •  related image
    Prediction
  •  related image
    Priority journal
  •  related image
    Probability
  •  related image
    Retrospective study
  •  related image
    Scoring system
  •  related image
    Support vector machine
  •  related image
    Validation process
  •  related image
    Computer assisted diagnosis
  •  related image
    Diagnostic imaging
  •  related image
    Female
  •  related image
    INTRAUTERINE GROWTH RETARDATION
  •  related image
    Machine learning
  •  related image
    Pregnancy
  •  related image
    Procedures
  •  related image
    FETAL GROWTH RETARDATION
  •  related image
    Humans
  •  related image
    Image interpretation, computer-assisted
  •  related image
    NEURAL NETWORKS, COMPUTER
  •  related image
    Retrospective studies
  •  related image
    ULTRASONOGRAPHY, PRENATAL