Rendering a time sequence of 2-D images includes obtaining ultrasound image data, applying time variant noise, and raytracing the ultrasound image data to render pixels of a 2-D image for each of a plurality of time steps.

Scan planes in acquisition of ultrasound images and dimensional measurements of objects represented in the images are determined by: defining a body to be examined and a predetermined orientation of the scan plane which intersects said body along which plane at least one ultrasound image is acquired; providing machine learning algorithm for verifying orientation and position of the scan plane, said algorithm trained with database of known cases which correlate image data of an image of a body previously examined and similar to said body with the position and orientation of the scan plane; reporting the scan plane does not correspond to the position and/or orientation of said predetermined scan plane; repeating the scan by changing the position and/or the orientation of the scan plane and repeating the steps as long as the position and orientation of the scan plane do not correspond to those of the predetermined scan plane.

An image generation device generates a plurality of reference in-focus images of an object placed on a surface of an image sensor by using a plurality of images captured by the image sensor using sensor pixels when the object is irradiated with light by a plurality of illuminators. Each of the reference in-focus images is an in-focus image corresponding to one of a plurality of virtual reference focal planes that are located between the image sensor and the plurality of illuminators. The plurality of reference focal planes pass through the object and are spaced apart from one another. The image generation device generates a three-dimensional image of the object by using the reference in-focus images and displays the three-dimensional image on a display screen.

An image generation device generates an image of an object placed on a surface of an image sensor by using a plurality of images each of which is captured by the image sensor when the object is irradiated with a corresponding one of a plurality of illuminators. The object includes first object and one or more second objects included in the first object. The image generation device determines a section of the first object including a largest number of feature points of second objects, generates an in-focus image using the section as a virtual focal plane, and causes the in-focus image to be displayed on a display screen.

A three-dimensional region of interest (ROI) is established with a high degree of accuracy, by a simple method without increasing a burden on the operator, in generating a three-dimensional projected image from medical volume data according to rendering, achieving more efficient interpretation of three-dimensional image and streamlining of diagnostic flow, with the use of the diagnostic image generation apparatus. An energy map is generated on a predetermined tomographic plane, assuming a preset start point as a reference and searching for a path that minimizes the energy, and then the path is set as a boundary of the three-dimensional ROI. The start point may be decided on the basis of the boundary inputted by a user, or the user may set the start point. The user may be allowed to adjust the boundary having been set. The boundary may also be determined on another plane orthogonal to the predetermined tomographic plane.

An image processing method of the invention comprises obtaining three-dimensional image data representing a three-dimensional image of a cell aggregate with a cavity inside obtained by optical coherence tomography imaging, obtaining a thickness distribution representing a thickness of the cell aggregate between an outer surface of the cell aggregate facing an outside space and an inner surface thereof facing the cavity at each position based on the three-dimensional image data; and detecting protrusion parts having a prominent thickness as compared to a surrounding region based on the thickness distribution.

Systems and methods of predicting future medical events are based on the processing of medical image. The prediction of premature birth and estimation of gestational age based on ultrasound images are presented as illustrative examples. The new abilities to estimate the probability of future medical events, before they otherwise could be predicted, provides new avenues for the development of preventative treatments.

A method, and corresponding apparatus, for identifying a feature in an image comprises determining a plurality of characteristic values for each pixel or group of pixels within the image, and determining a confidence value for a target area of the image on the basis of the plurality of characteristic values of the pixels within the target area. Each of the plurality of characteristic values relates to a different characteristic of the feature. The confidence value is indicative of whether the feature is represented by the target area of the image.

An Artificial Intelligence (AI) computational system for generating an embryo viability score from a single image of an embryo to aid selection of an embryo for implantation in an In-Vitro Fertilisation (IVF) procedure is described. The AI model uses a deep learning method applied to images in which the Zona Pellucida region in the image is identified using segmentation, and ground truth labels such as detection of a heartbeat at a six week ultrasound scan.

A computer-implemented method for predicting the likelihood an embryo will produce specific sex offspring by processing video image data derived from video of a target embryo. The method includes receiving image data derived from video of a target embryo taken at substantially real-time frame speed during an embryo observation period of time. The video contains recorded morphokinetic movement of the target embryo occurring during the embryo observation period of time. The movement is represented in the received image data and the received image data is processed using a model generated utilizing machine learning and correlated embryo outcome data to predict the likelihood the target embryo will produce specific sex offspring.