Abstract: A method includes obtaining a first frame of a probed region acquired using a first probe device at a first time, and a first set of control parameters used to acquire the first frame. Processing the first frame includes segmenting features from the first frame after first probe-specific effects are reduced in the first frame based on the first set of control parameters. The method includes, when the first frame contains a respective attribute present in a second frame acquired at a time earlier than the first time using a second set of control parameters, displaying information related to differences in the respective attribute in the first frame and the second frame. The second frame is obtained by processing the one or more features segmented from the second frame after one or more second probe-specific effects are reduced in the second frame.

Abstract: A method includes obtaining a first frame of a probed region acquired using a first probe device at a first time, and a first set of control parameters used to acquire the first frame. Processing the first frame includes segmenting features from the first frame after first probe-specific effects are reduced in the first frame based on the first set of control parameters. The method includes, when the first frame contains a respective attribute present in a second frame acquired at a time earlier than the first time using a second set of control parameters, displaying information related to differences in the respective attribute in the first frame and the second frame. The second frame is obtained by processing the one or more features segmented from the second frame after one or more second probe-specific effects are reduced in the second frame.

Abstract: A computing device receives first imaging data acquired via an ultrasound probe during a first portion of a scan performed by the ultrasound probe. The first imaging data was acquired in accordance with a first set of imaging control parameters, which requires that a first subset of a plurality of transducers of the ultrasound probe are activated during the first portion of the scan. During a second portion of the scan performed by the ultrasound probe, in accordance with a determination that the first imaging data meets a first set of conditions associated with one or more quality requirements for the first, the computing device causes the ultrasound probe to acquire second imaging data in accordance with a second set of imaging control parameters. The second set of imaging control parameters requires that a second subset of the plurality of transducers are activated during the second portion of the scan.

Abstract: A method includes generating a series of two-dimensional (2D) ultrasound images of a tissue volume associated with a plurality of positions, respectively, along a scanning direction of the tissue volume. The method includes, for a pair of consecutive 2D ultrasound images, using a deep neural network to classify a difference between the pair of consecutive 2D ultrasound images of the series of 2D ultrasound images. The deep neural network is trained to classify differences between images based on one of a plurality of predefined classes. The method includes modifying a number of 2D ultrasound images in the series of 2D ultrasound images based on the classification of the difference between the pair of consecutive 2D ultrasound images of the series of 2D ultrasound images, thereby producing a modified series of 2D ultrasound images. The method includes performing a measurement of tissue using the modified series of 2D ultrasound images.

Abstract: A device receives an image. The image is applied to a sequence of encoders that includes a first encoder and one or more subsequent encoders, each including a U network encoder which performs convolution neural network processing. The one or more subsequent encoders receive a down-sampled version of the image from a prior encoder. The sequence of encoders form a first dimension orthogonal to the orthogonal dimension. A result of the sequence of encoders is applied to a sequence of decoders that includes a first decoder and one or more subsequent decoders. A respective decoder comprises a U network decoder that performs convolution neural network processing of an up-sampled version of the image from the first decoder. Probability outputs are produced from paired encoders and decoders in the sequence of encoders and the sequence of decoders. The probability outputs are combined to form a final output.

Abstract: There is provided a method of image processing based on a convolutional neural network (CNN). The method includes: receiving an input image; performing a plurality of feature extraction operations using a plurality of convolution layers, respectively, of the CNN based on the input image to produce a plurality of output feature maps, respectively; and producing an output image for the input image based on the plurality of output feature maps of the plurality of convolution layers.

Abstract: A method for cross-referencing of 2D ultrasound scans of a tissue volume comprises generating first and second 2D representations of a target anatomy, respectively, from first and second series of 2D ultrasound images of the tissue volume generated. The method includes generating first and second simulated 2D representations of the target anatomy, respectively, from the second and first series of 2D ultrasound images. The 2D and simulated 2D representations are processed to at least substantially match pixel positions in the 2D and simulated 2D representations associated with the target anatomy. The method includes determining first and second correspondence transformation matrices from the processed 2D and simulated 2D representations, and using the first and second correspondence transformation matrices to determine, for a location in the second series of 2D ultrasound images associated with the target anatomy, a corresponding location in the first series of 2D ultrasound images.

Abstract: A device receives an image. The image is applied to a sequence of encoders that includes a first encoder and one or more subsequent encoders, each including a U network encoder which performs convolution neural network processing. The one or more subsequent encoders receive a down-sampled version of the image from a prior encoder. The sequence of encoders form a first dimension orthogonal to the orthogonal dimension. A result of the sequence of encoders is applied to a sequence of decoders that includes a first decoder and one or more subsequent decoders. A respective decoder comprises a U network decoder that performs convolution neural network processing of an up-sampled version of the image from the first decoder. Probability outputs are produced from paired encoders and decoders in the sequence of encoders and the sequence of decoders. The probability outputs are combined to form a final output.

Abstract: A method may include generating a series of two-dimensional (2D) ultrasound images of the tissue volume associated with a plurality of positions along a scanning direction of the tissue volume; estimating, for each pair of consecutive 2D ultrasound images of the series of 2D ultrasound images, a distance between the positions associated with the pair of consecutive 2D ultrasound images based on a classification of a difference image generated from the pair of consecutive 2D ultrasound images using a deep neural network to produce a plurality of estimated distances associated with the plurality of pairs of consecutive 2D ultrasound images, respectively modifying the number of 2D ultrasound images in the series of 2D ultrasound images based on the plurality of estimated distances to produce a modified series of 2D ultrasound images and rendering the 3D ultrasound image of the tissue volume based on the modified series of 2D ultrasound images.

Abstract: A non-transitory computer readable storage medium has instructions executed by a processor to receive an ultrasound image. The ultrasound image is applied to a sequence of encoders where each encoder in the sequence of encoders performs convolution neural network processing of a down-sampled version of the ultrasound image from a prior encoder, the sequence of encoders form a first dimension. The ultrasound image is applied to a transition encoder with an orthogonal dimension to the first dimension. The ultrasound image is applied to a sequence of decoders where each decoder in the sequence of decoders performs convolution neural network processing of an up-sampled version of the ultrasound image from a prior decoder, the sequence of decoders form a second parallel dimension to the first dimension. Encoder and decoder configurations and the first dimension, the orthogonal dimension and the second parallel dimension thereby define a nested U network architecture.

Abstract: A method may include generating a series of two-dimensional (2D) ultrasound images of the tissue volume associated with a plurality of positions along a scanning direction of the tissue volume; estimating, for each pair of consecutive 2D ultrasound images of the series of 2D ultrasound images, a distance between the positions associated with the pair of consecutive 2D ultrasound images based on a classification of a difference image generated from the pair of consecutive 2D ultrasound images using a deep neural network to produce a plurality of estimated distances associated with the plurality of pairs of consecutive 2D ultrasound images, respectively; modifying the number of 2D ultrasound images in the series of 2D ultrasound images based on the plurality of estimated distances to produce a modified series of 2D ultrasound images; and rendering the 3D ultrasound image of the tissue volume based on the modified series of 2D ultrasound images.

Abstract: A method may include generating a series of two-dimensional (2D) ultrasound images of the tissue volume associated with a plurality of positions along a scanning direction of the tissue volume; estimating, for each pair of consecutive 2D ultrasound images of the series of 2D ultrasound images, a distance between the positions associated with the pair of consecutive 2D ultrasound images based on a classification of a difference image generated from the pair of consecutive 2D ultrasound images using a deep neural network to produce a plurality of estimated distances associated with the plurality of pairs of consecutive 2D ultrasound images, respectively; modifying the number of 2D ultrasound images in the series of 2D ultrasound images based on the plurality of estimated distances to produce a modified series of 2D ultrasound images; and rendering the 3D ultrasound image of the tissue volume based on the modified series of 2D ultrasound images.

Abstract: A non-transitory computer readable storage medium has instructions executed by a processor to receive an ultrasound image. The ultrasound image is applied to a sequence of encoders where each encoder in the sequence of encoders performs convolution neural network processing of a down-sampled version of the ultrasound image from a prior encoder, the sequence of encoders form a first dimension. The ultrasound image is applied to a transition encoder with an orthogonal dimension to the first dimension. The ultrasound image is applied to a sequence of decoders where each decoder in the sequence of decoders performs convolution neural network processing of an up-sampled version of the ultrasound image from a prior decoder, the sequence of decoders form a second parallel dimension to the first dimension. Encoder and decoder configurations and the first dimension, the orthogonal dimension and the second parallel dimension thereby define a nested U network architecture.