Ultrasound imaging plane alignment guidance for neural networks and associated devices, systems, and methods

A guidance system for obtaining a medical image, comprising:

• a processor configured to:
• obtain a motion control configuration for repositioning an imaging device configured to be inserted inside a subject's body from a first imaging position to a second imaging position with respect to the subject's body using a control component in communication with the imaging device, wherein the motion control configuration is obtained based on:
• (i) a predictive network comprising at least a first convolutional neural network trained using one or more reference images,
• (ii) an image of the subject's body captured while the imaging device is positioned at the first imaging position, and
• (iii) a target image view including a clinical property, wherein the motion control configuration is obtained based on the predictive network without comparison to a reference image, wherein the first convolutional neural network is configured to compute a probability of reaching the target image view for each of a plurality of candidate motion control configurations and select the candidate motion control configuration with a highest probability of reaching the target image view;
• and wherein the motion control configuration corresponds to a motion vector generated using the predictive network based on the image of the subject's body captured while the imaging device is positioned at the first imaging position and the target image view including a clinical property;
• determine, via at least a second convolutional neural network, whether an image captured while the imaging device is positioned at the second imaging position contains the target image view, wherein the second convolutional neural network is trained to qualify whether an input image includes a target image view or a predetermined or selected clinical property;
• if the image captured while the imaging device is positioned at the second imaging position does not contain the target image view, repeat the obtaining operation to generate a new motion control configuration for repositioning the imaging device to a new imaging position using at least the first convolutional neural network;
• and
• if the image captured while the imaging device is positioned at the second imaging position does contain the target image view:
• generating a plurality of modified motion vectors based on the motion vector generated using the predictive network;
• repositioning the imaging device based on each of the modified motion vectors;
• capturing an image of the target image view including the clinical property at each of the modified motion vectors;
• and
• applying at least a third convolutional neural network to the captured images of the target image view to select an image including the target image view having the highest quality from among the captured images, wherein the third convolutional neural network is trained to compare pairs of input images and select an input image having a higher quality level from the pair of input images;
• a display in communication with the processor and configured to display an instruction, based on at least one of the motion control configuration and the new motion control configuration, for operating the control component such that the imaging device is repositioned to the second imaging position when the instruction is based on the motion control configuration displayed, or the new imaging position when the instruction is based on the new motion control configuration.