An eye disease diagnosis method using artificial intelligence may include: collecting, from a database, a first eyeground image of a myopic patient who is not diagnosed with an eye disease and a second eyeground image of a myopic patient who has been diagnosed with the eye disease; learning eyeball change information by degree of myopia based on the first eyeground image, using deep learning; comparing and analyzing the first eyeground image and the second eyeground image based on the eyeball change information by the degree of myopia, and learning eyeball change information by the eye disease using deep learning; and estimating determination criteria of an eyeground image for diagnosis of the eye disease, based on a difference between the eyeball change information by the degree of myopia and the eyeball change information by the eye disease.

A region specification apparatus specifies a region of an object which is included in an input image and which includes a plurality of subclass objects having different properties. The region specification apparatus includes a first discriminator that specifies an object candidate included in the input image. The first discriminator has a component configured to predict at least one of movement or transformation of a plurality of anchors according to the property of the subclass object and specify an object candidate region surrounding the object candidate.

A treatment support device includes an irradiation unit configured to emit treatment light to a medical agent, a fluorescence intensity acquisition unit configured to acquire fluorescence intensity of fluorescence emitted by the fluorescent material of the medical agent excited by the treatment light, and a treatment light imaging unit configured to capture a treatment light image based on the treatment light. The treatment support device is configured such that a first region of interest is capable of being set based on a position of the treatment light in the treatment light image captured by the treatment light imaging unit, the first region of interest being a region for selectively acquiring a temporal change in the fluorescence intensity acquired by the fluorescence intensity acquisition unit.

The present invention proposes an ultrasound system and a method of detecting lung sliding on the basis of a temporal sequence of ultrasound data frames of a first region of interest. The first region of interest includes a pleural interface of a lung. A sub-region identifier (410) is configured to identify, for each of the ultrasound data frames, a sub-region of a scanned region of the ultrasound data frame, the sub-region comprising at least part of the pleural interface; a lung sliding detector (420) is configured to derive a parametric map for the sub-region on the basis of at least two ultrasound data frames of the temporal sequence, parametric values of the parametric map indicating a degree of tissue motion over the at least two ultrasound frames; wherein the lung sliding detector is further configured to extract data of the sub-regions from the at least two ultrasound data frames, and to derive the parametric map on the basis of the extracted data.

Methods include processing image data through a plurality of network stages of a progressively holistically nested convolutional neural network, wherein the processing the image data includes producing a side output from a network stage m, of the network stages, where m>1, based on a progressive combination of an activation output from the network stage m and an activation output from a preceding stage m−1. Image segmentations are produced. Systems include a 3D imaging system operable to obtain 3D imaging data for a patient including a target anatomical body, and a computing system comprising a processor, memory, and software, the computing system operable to process the 3D imaging data through a plurality of progressively holistically nested convolutional neural network stages of a convolutional neural network.

A method for imaging a coronary arterial system of an individual includes releasing, using an actuator, pulses of a radio-opaque dye into a coronary arterial tree of the individual. The method further includes obtaining, using an image capture device, a sequence of invasive coronary x-ray angiogram images over time of the pulses of the radio-opaque dye. The method also includes tracking, using a processor, the pulses through the sequence of invasive coronary x-ray angiogram images and locating the pulses on a three dimensional (3D) structural model of the coronary arterial system to generate a three dimensional (3D) functional model of the coronary arterial system that shows a trajectory of the dye as it flows through different arterial branches.

Systems and methods for improving endoscopy procedures are described that provide not only a conventional real time image of the view obtained by an endoscope, but in addition, a near real time 3D model and/or a 2D flattened image of an interior surface of an organ, which model and image may be processed using AI software to highlight potential tissue abnormalities for closer examination and/or biopsy during the procedure. A navigation module interacts with other system outputs to further assist the endoscopist with navigational indicia, e.g., landmarks and/or directional arrows, that enhance the endoscopists' spatial orientation, and/or may provide navigational guidance to the endoscopist to assist manipulation of the endoscope.

A method for determining healing progress of a tissue disease state includes receiving a thermal image of a target wound area from a thermal imaging system, processing the thermal image to construct an isotherm map of at least one selected area, determining a thermal index value from the isotherm map, correlating the wound thermal index value with a reference thermal index value representative of an injury-free state.

An information processing apparatus includes an obtaining unit, an inference section, and a selection unit. The obtaining unit is configured to obtain a temporal subtraction image between a first medical image captured at a first point of time and a second medical image captured at a second point of time. The inference section includes a plurality of inference units, each for making an inference from the temporal subtraction image. The selection unit is configured to select, based on a region of interest in the obtained temporal subtraction image, at least one inference unit from the plurality of inference units in the inference section. In response to being selected by the selection unit, the at least one inference unit so selected makes the inference from the temporal subtraction image.

A scoring functions is developed and used for identifying patients who might be responsive to a PD-1 axis directed therapy. The scoring functions are obtained by extracting features from multiplex-stained sections, selecting features that correlate with response to the therapy using a feature selection function, and fitting one or more of the selected features to a plurality of candidate scoring functions. A candidate scoring function showing the desired balance between predictive sensitivity and specificity may then selected for incorporation into a scoring system that includes at least an image analysis system.