Unsupervised, deep learning of eye-landmarks in a user-specific eyes' image data by capturing an unlabeled image comprising an eye region of a user, using an initial geometrically regularized loss function, training a plurality of convolutional autoencoders on the unlabeled image comprising the eye region of the user to recover a plurality of user-specific eye landmarks, training a convolutional neural network for autoencoded landmarks-based recovery from the unlabeled image, and where the initial geometrically regularized loss function is represented by the formula L.sub.AE=?.sub.reconL.sub.recon+?.sub.concL.sub.conc+?.sub.sepL.sub.sep+?.sub.eqvL.sub.eqv where L.sub.AE is total AutoEncoder Loss, ?.sub.reconL.sub.recon is ?-weighted reconstruction loss, ?.sub.concL.sub.conce is ?-weighted concentration loss, ?.sub.sepL.sub.sep is ?-weighted separation loss, and ?.sub.eqvL.sub.eqv is ?-weighted equivalence loss.

A method of segmenting an MR organ volume includes performing regional mapping on the MR organ volume using a spatial prior probability map of a location of the organ to create a regionally mapped MR organ volume, and performing boundary refinement on the regionally mapped MR organ volume using a level set framework that employs the spatial prior probability map and a propagated shape constraint to generate a segmented MR organ volume.

Techniques are disclosed for implementing a convolutional neural network that determines an offset field for deforming a kernel to be used in a convolution. The offset field is temporally-based, at least in part, on data generated at an earlier time. Furthermore, techniques are disclosed for using sensor data to train a neural network to learn shapes or configurations of such deformed kernels. The temporal-based deformable convolutions may be used for object identification, object matching, object classification, segmentation, and/or object tracking, in various examples.

A medical scanning system and method for determining scanning parameters based on a scout image, the system includes: a scanned object description module for describing the shape of a scanned object on an initial image; an adjustment module for aligning the shape of the scanned object with the pre-stored average shape; a principal component analysis module for extracting the principal component for the aligned shape of the scanned object; a desired shape acquisition module for imparting weight parameters to said principal component, acquiring a plurality of new shapes, and from said plurality of new shapes, determining the new shape with the maximum cost function value as the desired shape and a scanning parameter setting module for setting scanning parameters based on the desired shape.

An interactive image segmenting apparatus and method are provided. The image segmenting apparatus and corresponding method include a boundary detector, a condition generator, and a boundary modifier. The boundary detector is configured to detect a boundary from an image using an image segmentation process. The feedback receiver is configured to receive information about the detected boundary. The condition generator is configured to generate a constraint for the image segmentation process based on the information. The boundary modifier is configured to modify the detected boundary by applying the generated constraint to the image segmentation process.

An image retrieval apparatus that retrieves a candidate medical image in diagnosis of diffuse lung disease based on a position of an abnormal shadow in an organ region in a target medical image, the apparatus includes a memory, and a processor coupled to the memory and configured to map the organ region in the target medical image to an image having a predetermined shape to make it identifiable whether the abnormal shadow is distributed over a first portion in the organ region or a second portion in the organ region, occurrence portions of the abnormal shadow within the first portion and the second portion are organizationally different, and calculate a position of the abnormal shadow after the mapping in the image having the predetermined shape.

Various aspects of a method and system to localize surgical tools during anatomical surgery are disclosed herein. In accordance with an embodiment of the disclosure, the method is implementable in an image-processing engine, which is communicatively coupled to an image-capturing device that captures one or more video frames. The method includes determination of one or more physical characteristics of one or more surgical tools present in the one or more video frames, based on one or more color and geometric constraints. Thereafter, two-dimensional (2D) masks of the one or more surgical tools are detected, based on the one or more physical characteristics of the one or more surgical tools. Further, poses of the one or more surgical tools are estimated, when the 2D masks of the one or more surgical tools are occluded at tips and/or ends of the one or more surgical tools.

An example method for positioning feature points of a human face edge including fitting a profile edge curve of a human face edge in a human face image according to the human face image; calculating by using a preset human face edge feature point calculation model to obtain feature points of the human face edge in the human face image; calculating by using a preset convergence algorithm to obtain feature information of a convergence feature point corresponding to each feature point; determining respectively whether a distance from each feature point to the profile edge curve is greater than a distance from a respective corresponding convergence feature point to the profile edge curve; and determining whether a magnitude of overall position change of all the feature points of the human face edge in the human face image before and after the above determination is less than a preset threshold.

Various aspects of a method and system to localize surgical tools during anatomical surgery are disclosed herein. In accordance with an embodiment of the disclosure, the method is implementable in an image-processing engine, which is communicatively coupled to an image-capturing device that captures one or more video frames. The method includes determination of one or more physical characteristics of one or more surgical tools present in the one or more video frames, based on one or more color and geometric constraints. Thereafter, two-dimensional (2D) masks of the one or more surgical tools are detected, based on the one or more physical characteristics of the one or more surgical tools. Further, poses of the one or more surgical tools are estimated, when the 2D masks of the one or more surgical tools are occluded at tips and/or ends of the one or more surgical tools.

A method for extracting a blood vessel may include acquiring an image relating to a blood vessel, the image including multiple slices; determining a region of interest in the image; establishing a blood vessel model; and extracting the blood vessel from the region of interest based on the blood vessel model.