Shanghai United Imaging Healthcare

The present application relates to a method for training a signal denoising model, a signal denoising method, and a device. According to the present application, the method includes: performing stage-by-stage denoising training on a multi-stage network based on training signal data corresponding to the multi-stage network, to obtain a signal denoising model, obtaining, when a current stage subjected to the stage-by-stage denoising training is not a first stage, a loss value of a current-stage network based on training signal data corresponding to the current-stage network, and after gradient descent is performed individually on the current-stage network based on the loss value, performing, based on the loss value, gradient descent jointly on current multiple stages of the network that have undergone the gradient descent individually.

A method for cross-scattering correction for a dual-source or multi-source CT device may be provided. The dual-source or multi-source CT device at least includes a first source-detector pair and a second source-detector pair. The method comprises obtaining a detection signal curve of a first detector of the first source-detector pair. The method also comprises determining a first convolution region based on the detection signal curve, and determining estimated cross-scattering signal intensities of a plurality of detector channels of the first detector by performing a convolution operation on the first convolution region using a target kernel. The target kernel characterizes scattering contributions from a second source of the second source-detector pair to the first detector. The method further comprises determining corrected signal intensities of the detector channels by performing a cross-scattering correction on the signal intensities of the detector channels based on the estimated cross-scattering signal intensities of the detector channels.

Embodiments of the present disclosure provide a PET system. A PET system comprises multiple detector rings, a first processing card and a second processing card. The multiple detector rings are arranged side by side along an axial direction of the PET system, each of the multiple detector rings includes a plurality of detector modules arranged along a circumferential direction of the detector ring. The first processing card is communicatively connected to the multiple detector rings, wherein the first processing card is configured to receive detection data with a same circumferential index, and determine a target event based on the detection data. The second processing card is communicatively connected to the first processing card, and is configured to determine a coincidence event pair by processing the target event.

A wireless coil applied to an MRI system is provided. The MRI system includes a transmitting coil having a housing cavity. The wireless coil includes: a flexible member and one or more magnetic coupling units, disposed on a surface of or inside the flexible member, wherein the wireless coil and the transmitting coil are respectively separated from each other in a radial direction of the housing cavity, and the wireless coil is capable of resonating under a radio frequency field generated by the transmitting coil of the MRI system to produce a focal field. A coverage area of the focal field is smaller than a coverage area of the radio frequency field.

A method for medical scanning is provided. The method includes: obtaining, via a user terminal, an input of a user specifying scan requirements with respect to a target subject to be scanned; extracting, from the input, scan requirement information; obtaining medical record information of the target subject; and generating a scanning scheme for the target subject by processing the scan requirement information and the medical record information, wherein the processing of the scan requirement information and the medical record information is performed with reference to long-term memory information stored in the at least one storage device, and the long-term memory information comprises at least one of a rule database, a knowledge graph, or a historical scanning scheme database.

The present application provides a transformer and an X-ray system. The transformer includes M primary windings (100, 110, 120, 130) and M*N distributed magnetic cores (200, 210, 220, 230). The M primary windings are supplied with a M-phase alternating current, a first end of each of the M primary windings is supplied with an alternating current of a corresponding phase in the M-phase alternating current, and second ends of the M primary windings are connected to each other after passing through a plurality of distributed magnetic cores in the M*N distributed magnetic cores. The secondary winding (300, 310, 320, 330) on each distributed magnetic core is connected to a corresponding one of a plurality of rectifier units connected in series to output a secondary voltage.

Disclosed is a determination method for adaptive radiotherapy. The method includes: obtaining a current fraction image corresponding to a current radiotherapy fraction of a target subject, and obtaining an initial image configured for determining an initial radiotherapy plan for the target subject; determining a trigger analysis result according to the current fraction image, the initial image, and a trigger analysis model; and determining whether to perform adaptive radiotherapy based on the trigger analysis result.

The present application relates to a cathode and a manufacturing method thereof. The method includes: mixing and reacting cations containing a cathode element with a precipitant in a solution to obtain a gelatinous precipitate; and coating the gelatinous precipitate on a surface of a cathode substrate, which is configured for forming the cathode.

A reconstruction method of an Emission Computed Tomography image, an apparatus, a device, and a storage medium are provided in the present invention. The method includes: acquiring a first ECT image sequence, inputting the first ECT image sequence into a trained intermediate frame prediction network to obtain an estimated second ECT image sequence, and reconstructing a target second ECT image sequence based on the estimated second ECT image sequence serving as a prior image. Temporal resolutions of both the target second ECT image sequence and the estimated second ECT image sequence are higher than that of the first ECT image sequence.

The present disclosure relates to a radiation delivery apparatus (10) and a non-transitory computer-readable storage medium. The radiation delivery apparatus (10) includes a radiation head (11) configured to deliver a radiation beam and rotatable about a rotation axis; and a movement mechanism (12) connected to the radiation head (11). The radiation head (11) is movable along the direction of the rotation axis through the movement mechanism (12) and is movable along the beam direction through the movement mechanism (12).