An image processing apparatus includes a processor including hardware. The processor is configured to evaluate, based on information including an image which is captured by an endoscope and is sequentially input, a technical level of an operator who operates the endoscope, and measure a transit time of a specific section.

In one embodiment, a patient's brain is evaluated after onset of a stroke by capturing computed tomography angiography (CTA) images of the brain, analyzing the CTA images with a CTA image analysis program to evaluate the patient's brain, and generating results based upon the analysis that provide an assessment of the brain. In some cases, the CTA image analysis program comprises a machine-learning algorithm that has been trained on the results of perfusion imaging analysis.

A method and system for computer-based motion estimation and modeling in a medical image sequence of a patient is disclosed. A medical image sequence of a patient is received. A plurality of frames of the medical image sequence are input to a trained deep neural network. Diffeomorphic deformation fields representing estimated motion between the frames of the medical image sequence input to the trained deep neural network are generated. Future motion, or motion between frames, is predicted from the medical image sequence and at least one predicted next frame is generated using the trained deep neural network. An encoding of the observed motion in the medical image sequence is also generated, which is used for motion classification (e.g., normal or abnormal) or motion synthesis to generate synthetic data.

A method of determining respiratory phase of a living body from a sequence of digitized fluoroscopic images of a living-body region exhibiting respiratory displacement, the method employing programmable computing apparatus and comprising the steps of: (a) in each living-body-region image in the sequence, defining one or more zones with each image having identical image-to-image zone locations, sizes, and shapes; (b) for each image, computing an average pixel intensity for each zone to form a sequence thereof for each zone; (c) for each zone, modifying the average pixel intensities by (i) computing the mean value of the sequence of average pixel intensities for such zone, and (ii) subtracting the mean from each average pixel intensity in the zone; (d) for each zone sequence, computing a figure-of-merit; (e) selecting the zone having the highest figure-of-merit; and (f) using the sequence of pixel intensities of the selected zone to determine respiratory phase.

The present disclosure relates to a method, storage medium, and system for analyzing an image sequence of a periodic physiological activity. In one implementation, the method includes receiving the image sequence acquired by an imaging device, the image sequence having a plurality of frames, and identifying a feature point in a first frame. The method further includes determining motion vectors for the feature point in the frames of the image sequence. Each motion vector for the feature point is determined based on respective locations of corresponding feature points in frames adjacent to the first frame. The method also includes determining a motion magnitude profile based on the determined motion vectors and determining a phase of each frame in the image sequence based on the motion magnitude profile.

The present invention is directed to compositions, methods and systems for identifying, in real time, the position and orientation of the esophagus prior to and during atrial fibrillation ablation procedures, so as to avoid or reduce the incidence of atrioesophageal fistula (AEF). The compositions, methods and systems of the present invention include the identification and visualization of the esophagus, the rapid and accurate integration of the visualized esophagus into an anatomical map together with the posterior wall of the left atrium, in each case presented as a 3-D map, so as to facilitate the accurate identification of those areas of the esophagus that lie in contact with or in near proximity to those areas of the posterior wall of the left atrium that the operator intends to ablate.

A method for determining a sequence of activation of a set of brain networks by an electronic device in the course of a predetermined cognitive task. The method includes: obtaining at least one time series of data on encephalographic activities, according to a predetermined sampling value, the data on activities representing a signal captured on the cranial surface by a capture device, during the execution of the predetermined cognitive task by at least one individual, delivering a set having at least one measurements vector per sample of the time series; for each measurements vector, determining connectivity between cortical sources, the cortical sources being obtained through measurements of a measurements vector, delivering a connectivity network; and grouping connectivity networks according to a resemblance parameter delivering a set of groups of grouped connectivity networks, each representing an activity of a given brain network at a given instant.

In one embodiment, a medical image processing apparatus which analyzes blood flow dynamics in a predetermined region of a subject, the blood flow dynamics being generated from medical images obtained by imaging the predetermined region in time sequence over a plurality of time phases. The medical image processing apparatus includes memory circuitry configured to store a program; and processing circuitry configured to correct pixel values of a second medical image according to an amount of deformation of the second medical image when the second medical image is aligned with a first medical image by executing the program read out from the memory circuitry, the first medical image and the second medical image being among the medical images in the plurality of time phases.

Machine logic (for example, software) for registering multiple medical images, each showing a common lesion, with each other. In performing this registration, registration points are chosen to be both: (i) outside of image portion that is potentially compromised by the lesion (in any of the multiple images); and (ii) as close to the lesion as possible. However, in at least one of the images the extent of the lesion is not knownso, in order to accommodate this uncertainty about the lesion boundaries, lesion predicting machine logic rules are used to predict the size, shape and/or location of the lesion. Machine learning is used to intermittently adjust and improve the lesion predicting machine logic rules.

Systems and methods for performing a medical imaging analysis task are provided. One or more input medical images of a patient are received. The one or more input medical images are encoded into embeddings using a machine learning based encoder network. A medical imaging analysis task is performed based on the embeddings. Results of the medical imaging analysis task are output.