Systems and methods for augmenting a training data set with annotated pseudo images for training machine learning models. The pseudo images are generated from corresponding images of the training data set and provide a realistic model of the interaction of image generating signals with the patient, while also providing a realistic patient model. The pseudo images are of a target imaging modality, which is different than the imaging modality of the training data set, and are generated using algorithms that account for artifacts of the target imaging modality. The pseudo images may include therein the contours and/or features of the anatomical structures contained in corresponding medical images of the training data set. The trained models can be used to generate contours in medical images of a patient of the target imaging modality or to predict an anatomical condition that may be indicative of a disease.

An ultrasound imaging device and a clutter filtering method using the same are disclosed. The clutter filtering method using the ultrasound imaging device according to one embodiment includes obtaining ultrasound data from a field-of-view (FOV) of an object, generating decomposition data including common scale information by performing rank matrix decomposition once on all of the obtained ultrasound data, estimating local characteristic information by reflecting spatial information on each pixel to the common scale information, and extracting a blood flow signal by performing filtering on each pixel based on the estimated local characteristic information.

A workflow is disclosed to accurately register ultrasound imaging to co-modality imaging. The ultrasound imaging is segmented with a convolutional neural network to detect a surface of the object. The ultrasound imaging is calibrated to reflect a variation in propagation speed of the ultrasound waves through the object by minimizing a cost function that sums the differences between the first and second steered frames, and compares the first and second steered frames of the ultrasound imaging with a third frame of the ultrasound imaging that is angled between the first and second steered frames. The ultrasound imaging is temporarily calibrated with respect to a tracking coordinate system by creating a point cloud of the surface and calculating a set of projection values of the point cloud to a vector. The ultrasound imaging, segmented and calibrated, is automatically registered to the co-modality imagine.

According to one embodiment, an analysis apparatus includes processing circuitry. The processing circuitry configured to generate a harmonic signal and a fundamental wave signal based on a reception signal that is collected by an ultrasound probe, the harmonic signal corresponding to a harmonic component of a reflected wave of a ultrasound generated in the subject, the fundamental wave signal corresponding to a fundamental wave component of the reflected wave, calculate a first index value indicating tissue properties of the subject based on the harmonic signal, and calculate a second index value indicating the tissue properties based on the fundamental wave signal, and display an analysis result based on the first index value and the second index value.

Various embodiments are directed to video-based deep learning evaluation of cardiac ultrasound that accurately identify cardiomyopathy and predict ejection fraction, the most common metric of cardiac function. Embodiments include systems and methods for analyzing images obtained from an echocardiogram. Certain embodiments include receiving video from a cardiac ultrasound of a patient illustrating at least one view the patient's heart, segmenting a left ventricle in the video, and estimating ejection fraction of the heart. Certain embodiments include at least one machine learning algorithm.

A defect inspection device includes an ultrasonic probe, an image acquirer, a calculator, and a corrector. The ultrasonic probe acquires an ultrasonic image of an inspection object or a simulated inspection object. The image acquirer acquires an infrared image including a first region of the simulated inspection object or a second region of the inspection object. The calculator calculates a first correction value for correcting a coordinate deviation of the first region in the ultrasonic image and the infrared image with respect to a designed coordinate of the first region, or calculate a second correction value for correcting a coordinate deviation of the second region in the infrared image with respect to a designed coordinate of the second region. The corrector performs coordinate correction with the calculated first or second correction value for the ultrasonic image of the inspection object.

A fingerprint anti-counterfeiting method and an electronic device are provided. The fingerprint anti-counterfeiting method includes: After detecting a fingerprint input action of a user, an electronic device obtains a fingerprint image generated by the fingerprint input action, and obtains a vibration-sound signal generated by the fingerprint input action. The device determines, based on a fingerprint anti-counterfeiting model, whether the fingerprint input action is performed by a true finger. The fingerprint anti-counterfeiting model is a multi-dimensional network model obtained through learning based on fingerprint images for training and corresponding vibration-sound signals. The fingerprint anti-counterfeiting method in embodiments of this application helps improve a protection capability of the electronic device for a fake fingerprint attack.

A non-transitory computer-readable medium (CRM) storing computer program code executed by a computer processor that executes a process, an information processing apparatus, and a model generation method that outputs complication information for a medical treatment. The process includes acquiring a medical image obtained by imaging a lumen organ of a patient before treatment, inputting the acquired medical image into a trained model so as to output complication information on a complication that is likely to occur after the treatment when the medical image is received, and outputting the complication information. Preferably, complication information including a type of the complication that is likely to occur and a probability value indicating an occurrence probability of the complication of the type is output.

A non-transitory computer-readable medium (CRM) storing computer program code executed by a computer processor that executes a process, an information processing apparatus, and model generation method that generates an image of a lumen organ. The process includes acquiring a first image obtained by imaging a lumen organ of a patient based on an ultrasound signal of a first frequency; and generating a second image by inputting the acquired first image into a model, the model being learned to generate, when the first image is input, the second image in which the lumen organ is imaged based on an ultrasound signal of a second frequency. Preferably, the second image, in which a part of an image region of the first image is converted into the second frequency, is generated using the model, and a synthesis image is generated in which the second image is superimposed to the first image.

A computer is caused to perform processing of: acquiring a plurality of medical images generated based on signals detected by a catheter inserted into a lumen organ while the catheter is moving a sensor along a longitudinal direction of the lumen organ, the lumen organ including a main trunk, a side branch branched from the main trunk, and a bifurcated portion of the main trunk and the side branch; and recognizing a main trunk cross-section, a side branch cross-section, and a bifurcated portion cross-section by inputting the acquired medical images into a learning model configured to recognize the main trunk cross-section, the side branch cross-section, and the bifurcated portion cross-section.