Abstract: Described herein are systems, methods, and instrumentalities associated with denoising medical images such as fluoroscopic images using deep learning techniques. A first artificial neural network (ANN) is trained to denoise an input medical image in accordance with a provided target noise level. The training of the first ANN is conducted by pairing a noisy input image with target denoised images that include different levels of noise. These target denoised images are generated using a second ANN as intermediate outputs of the second ANN during different training iterations. As such, the first ANN may learn to perform the denoising task in an unsupervised manner without requiring noise-free training images as the ground truth.

Abstract: A method, system, and computer program product that are configured to: receive a Structured Query Language (SQL) statement including placeholders; generate sets of predicted values for the placeholders; generate candidate access paths in a database using the sets of predicted values; receive a query including the SQL statement with actual values instead of the placeholders; select one of the candidate access paths based on determining similarities of the actual values to ones of the sets of predicted values; in response to the selected one of the candidate access paths being acceptable, execute the query using the selected one of the candidate access paths; and in response to the selected one of the candidate access paths not being acceptable, generate a new access path in the database using the actual values, and execute the query using the new access path.

Abstract: A dataset is determined to include skewed data and based on determining that the dataset includes skewed data, one or more sparse indexes for the skewed data of the dataset are built. A sparse index of the one or more sparse indexes is for a range of data of the skewed data, and the building the sparse index includes indicating one pointer for a selected record of the range of data and indicating another pointer for another selected record of the range of data. The sparse index is provided to be used in a query of the dataset.

Abstract: An apparatus for stent visualization includes a hardware processor that is configured to input one or more stent images from a sequence of X-ray images and corresponding balloon marker location data to a cascaded spatial transform network. The background is separated from the one or more stent images using the cascaded spatial transform network and a transformed stent image with a clear background and a non-stent background image is generated. The stent layer and non-stent layer are generated using a neural network without online optimization. A mapping function f maps the inputs, the sequence images and marker coordinates, into the two single image outputs.

Abstract: When a force sensor on a robot arm detects that the force of contact between an end of a calibration device and a calibration plate reaches a threshold, the robot arm stops, and the end of the calibration device performs marking at the contact position between the end of the calibration device and the calibration plate. The robot arm moves upward and stops at a position where the end of the robot arm is at a predetermined height. At this position, a camera at the end of the robot arm photographs marks on the calibration plate, records the coordinates of the marks in the camera coordinate system, and records the coordinates of the end of the calibration device in the robot coordinate system. A calibration transformation matrix is calculated according to the recorded coordinates of at least three marks.

Abstract: Neural network based systems, methods, and instrumentalities may be used to remove motion artifacts from magnetic resonance (MR) images. Such a neural network based system may be trained to perform the motion artifact removal tasks without reference (e.g., without using paired motion-contaminated and motion-free MR images). Various training techniques are described herein including one that feeds the neural network with pairs of MR images with different levels of motion contamination and forces the neural network learn to correct the motion contamination by transforming a first image of a contaminated pair into a second image of the contaminated pair. Other neural network training techniques are also described with an aim to reduce the reliance on training data that is difficult to obtain.

Abstract: Described herein are systems, methods and instrumentalities associated with automatic assessment of aneurysms. An automatic aneurysm assessment system or apparatus may be configured to obtain, e.g., using a pre-trained artificial neural network, strain values associated one or more locations of a human heart and one or more cardiac phases of the human heart and derive a representation (e.g., a 2D matrix) of the strain values across time and/or space. The system or apparatus may determine, based on the derived representation of the strain values, respective strain patterns associated with the one or more locations of the human heart and further determine whether the one or more locations are aneurysm locations by comparing the automatically determined strain patterns with predetermined normal strain patterns of the heart and determining the presence or risk of aneurysms based on the comparison.

Abstract: The application relates to the field of man-machine interaction in artificial intelligence and provides a multi-task processing method. The method includes the following operations: determining a first task based on request information entered by a user; obtaining key information corresponding to the first task and executing the first task, where the key information includes one or more slots and values of the one or more slots; storing task status information of the first task, where the task status information includes the key information; and predicting and initiating a second task based on the task status information of the first task. A man-machine interaction system may predict a next task based on the stored task status information, and actively initiate the predicted task. This improves intelligence and efficiency of multi-task processing by the man-machine interaction system.

Abstract: A computer-implemented method dynamically switches access plans for a query during concurrent query execution. The method includes receiving a first query configured to be processed by a database system. The method also includes generating, for the first query, an access plan for each of identified resource sets. The method includes determining a first set of available resources that represent an available capacity for the database system. The method further includes selecting a first resource set of the one or more resource sets, where the selecting is based on the first set of available resources being closest to the first resource set. The method also includes selecting, based on the first set of available resources, a first access plan of the one or more access plans. The method includes executing the first query and returning results of the first query to a source of the first query.

Abstract: Disclosed is a selective field-assisted machining system. The system includes a micron-level high-speed identification module, an in-situ laser assisted module, an ultrasonic vibration module, an energy field loading high-speed control module, and a diamond tool. The micron-level high-speed identification module is used to quickly identify the type of a material substrate of a workpiece to be processed, process the identification information into a corresponding control signal, and send same to the energy field loading high-speed control module to implement selective processing of the workpiece to be processed, i.e. to process brittle particles using in-situ laser assisted machining and to process a soft metal substrate using ultrasonic vibration processing. In the present invention, ultra-precision cutting of brittle particles and a soft metal substrate can be completed at the same time in a single processing process.

Abstract: Described herein are systems, methods, and instrumentalities associated with using a multi-layer perceptron (MLP) neural network to process medical images of an anatomical structure. The processing may include padding an input image in accordance with the training of the MLP neural network, splitting the input image (e.g., the padded input image) into patches of a same size, and processing the patches through the MLP neural network over one or more iterations. During an iteration of the processing, the patches may be processed separately and re-combined into an intermediate image before the intermediate image is shifted to concatenate portions of the image that are derived from different patches. This way, global features of the anatomical structure may be learned and used to improve the quality of the image generated by the MLP neural network, without incurring significant computation or memory costs.

Abstract: A microfluidic apparatus, and its drive circuit and drive method are provided in the present disclosure. The drive circuit includes at least one switch unit, where the switch unit includes a first signal input terminal, a signal output terminal, first and second control signal terminals, a signal terminal, a first module, a second module, and a third module. The first module is electrically connected to the first signal input terminal, the signal output terminal and the second module; and the second module is electrically connected to the first control signal terminal. The first control signal terminal is configured to control the first module and the second module to be in conduction or disconnection, and the second control signal terminal is configured to control the third module to in conduction or disconnection, thereby controlling the signal output terminal to output a first or second signal.

Abstract: A computer-implemented method, including receiving, by a processor set, a query including a query string for a system catalog; identifying, by the processor set, a default index structure of the system catalog; executing, by the processor set, the query based on the default index structure of each index in the system catalog; ranking, by the processor set, a performance of each execution of the query by each index; mapping, by the processor set, a query pattern to a corresponding index of the system catalog; selecting, by the processor set, the index to perform the query using a machine learning (ML) model trained with a knowledge base that includes the ranking and the mapping; executing, by the processor set, the query on the selected index; and in response to executing the query on the selected index, returning, by the processor set, a result of the query.

Abstract: Embodiments of the present application provide a file processing method, a file processing device, an electronic apparatus, and a readable storage medium. The method comprises: receiving an operation request for a file, the file comprising at least two sub-files separately stored in at least two equipments; and correspondingly processing at least one sub-file of the file according to the operation request. By the solutions provided in the embodiments of the present application, since at least two sub-files of a file are separately stored in at least two equipments, original contents of the file cannot be obtained even if a sub-file in one equipment is leaked. By the solutions provided in the embodiments of the present application, the security of files can be effectively improved.

Abstract: Provided is a water-resistance rowing exercise machine, comprising a bracket, a water tank fixed to the bracket, a linkage mechanism, a rotation angle adjustment mechanism. The water tank is internally provided with a first impeller and a second impeller. The first impeller comprises a first plate portion and a plurality of first blades distributed on the periphery of the first plate portion. The second impeller comprises a second plate portion and a plurality of second blades distributed on the periphery of the second plate portion. The rotation angle adjustment mechanism is configured to adjust a relative rotation angle between the first impeller and the second impeller the first blades and the second blades may approach to or depart from each other. The distance between the first blades and the second blades can be adjusted by adjusting the relative rotation angle between the first impeller and the second impeller.

Abstract: Described herein are systems, methods, and instrumentalities associated with reconstruction of multi-contrast magnetic resonance imaging (MRI) images. The reconstruction may be performed based on under-sampled MRI data collected for the multiple contrasts using corresponding sampling patterns. The sampling patterns and the reconstruction operations for the multiple contrasts may be jointly optimized using deep learning techniques implemented through one or more neural networks. An end-to-end reconstruction optimizing framework is provided with which information collected while processing one contrast may be stored and used for another contrast. A differentiable sampler is described for obtaining the under-sampled MRI data from a k-space and a novel holistic recurrent neural network is used to reconstruct MRI images based on the under-sampled MRI data.

Abstract: Mammography may provide multi-view information about the healthy state of a person's breasts, and described herein are deep learning based techniques for obtaining mammographic images associated with multiple views, extracting cross-view features from the images, and automatically determining the existence or non-existence of a medical abnormality based on the extracted features. The cross-view features may be determined using an encoding module of an artificial neural network (ANN) and the encoded features may be decoded using a decoding module of the ANN to generate a prediction (e.g., a classification label, a bounding shape, a segmentation mask, a probability map, etc.) about the medical abnormality.

Abstract: A method includes using fully sampled retro cine data to train an algorithm, and applying the trained algorithm to real time MR cine data to yield reconstructed MR images.