Elekta Limited

Systems and methods are disclosed for performing operations comprising: receiving a new data sample comprising estimated motion; identifying a cluster of training data samples from a plurality of training data samples that corresponds to the new data sample; obtaining motion estimations associated with the cluster of training data samples; determining a reconstruction from the cluster of training data samples to the new data sample; transforming the motion estimations for the cluster of training data samples based on the reconstruction from the cluster of training data samples to the new data sample; and computing a quality factor indicating a measure of similarity between the estimated motion of the new data sample and the motion estimations associated with the cluster of training data samples.

A method for performing deformable image registration of a first volumetric medical image to a second volumetric medical image may comprise estimating a time varying velocity field between the images by encoding coordinates using a time varying positional embedding and using the encoded coordinates and a Neural Field Ordinary Differential Equation (NFODE) to generate a prediction of the rate of change of the deformation field. The method may further comprise integrating the estimated velocity field to generate a deformation field and applying the generated deformation field to the first volumetric medical image to generate a registered volumetric medical image. The NFODE May comprise a non-stationary Neural ODE parameterized by an Implicit Neural Representation.

A method for training an ML model to generate lower dimensional representations of a plurality of Implicit Neural Representations (INRs) can include obtaining a plurality of INRs, each INR representing a medical image. The method can further comprise, for each of the plurality of INRs, repeating, at least twice, identifying an augmentation for application to the INR, applying the identified augmentation to the INR, and inputting the augmented version of the INR to an encoder ML model. The encoder ML model can output a vector representation of the augmented version of the INR. The vector representation is of a lower dimension than the INR. The method can further comprise determining a pairwise similarity between pairs of vector representations output by the encoder ML model and updating trainable parameters of the encoder ML model to minimize a loss function. The loss function rewards increased similarity between vector representations of augmented versions of the same INR.

Systems and methods are disclosed for performing operations comprising: receiving a new data sample comprising estimated motion; identifying a cluster of training data samples from a plurality of training data samples that corresponds to the new data sample; obtaining motion estimations associated with the cluster of training data samples; determining a reconstruction from the cluster of training data samples to the new data sample; transforming the motion estimations for the cluster of training data samples based on the reconstruction from the cluster of training data samples to the new data sample; and computing a quality factor indicating a measure of similarity between the estimated motion of the new data sample and the motion estimations associated with the cluster of training data samples.

A dynamic prototype convolution network (DPCN) can achieve sufficient information interaction between support features from a support image and query features from a query image for performing Few-shot semantic segmentation (FSS). A dynamic convolution module (DCM) can generate dynamic filters from a support foreground. Then, information interaction can be achieved by convolution operations over query features, such as by using these dynamic filters. A support activation module (SAM) and a feature filtering module (FFM) can be used to mine context information from a query feature. The SAM can learn to generate a pseudo mask for a query image. The FFM can refine the pseudo mask to filter background information from a query feature. Thus, information both from query and support can be used to achieve more accurate prediction. The DPCN can be used to perform k-shot segmentation.