Abstract: A memory includes S storage blocks, N global bitlines, and a signal amplification circuit. Each of the S storage blocks is connected to the N global bitlines, the N global bitlines are connected to the signal amplification circuit, the signal amplification circuit is configured to amplify electrical signals on the N global bitlines, and each storage block includes N columns of storage units, N local bitlines, and N bitline switches. In each storage block, storage units in an ith column are connected to an ith local bitline, the ith local bitline is connected to an ith global bitline by using an ith bitline switch in the N bitline switches. A memory array is fine-grained, so that ith local bitlines in the S storage blocks can share one global bitline.

Abstract: Embodiments of the disclosure provide systems and methods for biomedical image analysis. A method may include receiving a plurality of unannotated biomedical images, including a first image and a second image. The method may also include determining that the first image is in a first view and the second image is in a second view. The method may further include assigning the first image to a first processing path for the first orientation. The method may additionally include assigning the second image to a second processing path for the second view. The method may also include processing the first image in the first processing path in parallel with processing the second image in the second processing path. The first path may share processing parameters with the second path. The method may further include providing a diagnostic output based on the processing of the first image and the second image.

Abstract: A method, system, and ASIC chip for comparing a bit error rate (BER) to a forward error correction (FEC) threshold to determine a Q margin for a codeblock; wherein the BER corresponds to the number of errors in a given amount of data; where a codeblock of a FEC corresponds to the given amount of data; wherein the FEC threshold corresponds to the maximum amount of errors per codeblock that the FEC is able to remove per given amount of data; wherein the Q margin corresponds to a difference between the BER and the FEC threshold.

Abstract: The present application relates to a safety prevention and control system and control method of a power battery pack for an electric vehicle. The safety prevention and control system of a power battery pack includes a signal acquisition device, a main controller, and a step-by-step prevention and control execution device. The main controller includes a fault diagnosis device, a cell thermal runaway determination device and a battery pack thermal runaway propagation determination device, which are respectively electrically connected to the step-by-step prevention and control execution device to send different control instructions to the step-by-step prevention and control execution device. The step-by-step prevention and control execution device can execute prevention and control actions of different levels according to the different control instructions sent by the fault diagnosis device, the cell thermal runaway determination device and the battery pack thermal runaway propagation determination device.

Abstract: Devices, systems, and methods for hardware-based time synchronization for heterogenous sensors are described. An example method includes generating a plurality of input trigger pulses having a nominal pulse-per-second (PPS) rate, generating, based on timing information derived from the plurality of input trigger pulses, a plurality of output trigger pulses, and transmitting the plurality of output trigger pulses to a sensor of a plurality of sensors, wherein a frequency of the plurality of output trigger pulses corresponds to a target operating frequency of the sensor, wherein, in a case that a navigation system coupled to the synchronization unit is functioning correctly, the plurality of input trigger pulses is generated based on a nominal PPS signal from the navigation unit, and wherein, in a case that the navigation system is not functioning correctly, the plurality of input trigger pulses is generated based on a simulated clock source of the synchronization unit.

Abstract: An autonomous vehicle includes a detection system for identifying the presence changes in wind incident on the autonomous vehicle, particularly wind gusts. The detection system may include one or more wind sensors, particularly those configured to detect wind incident on the vehicle from a direction that is transverse or perpendicular to the direction of motion of the autonomous vehicle. Additionally, systems may be present that correlate the detected wind gusts to changes in the behavior of the autonomous vehicle. The autonomous vehicle may react to the detected wind gusts by altering the vehicle's trajectory, by stopping the vehicle, or by communicating with a control center for further instructions.

Abstract: The disclosure provides a method, device and a computer-readable medium for performing three-dimensional blood vessel reconstruction. The computer-implemented method includes receiving a first two-dimensional image of a blood vessel of a patient, where the first two-dimensional image is a projection image acquired in a first projection direction. The method further includes reconstructing, by a processor, a three-dimensional model of the blood vessel based on at least the first two-dimensional image. The method additional includes adjusting the three-dimensional model of the blood vessel, based on a comparison of a first optical path length determined from a second two-dimensional image of the blood vessel of the patient and a second optical path length determined from the three-dimensional model.

Abstract: Disclosed are a video watermark identification method and apparatus, device, and a storage medium. The method includes: each video frame of multiple video frames of a video is divided into multiple image blocks to obtain an image sequence corresponding to the each video frame; multiple image sequences corresponding to the multiple video frames are input to a target detection model to obtain a classification result of each image block, and a feature vector of the video is obtained according to all classification results of the multiple image blocks; and the feature vector of the video is input to a watermark identification model to obtain a watermark identification probability output by the watermark identification model, and the video containing a watermark is determining in a case where the watermark identification probability is greater than or equal to a probability threshold.

Abstract: Embodiments of the disclosure provide systems and methods for detecting a medical condition of a subject. The system includes a communication interface configured to receive a sequence of images acquired from the subject by an image acquisition device and an end-to-end multi-task learning model. The end-to-end multi-task learning model includes an encoder, a Convolutional Recurrent Neural Network (ConvRNN), and at least one of a decoder and a classifier. The system further includes at least one processor configured to extract feature maps from the images using the encoder, capture contextual information between adjacent images in the sequence using the ConvRNN, and detect medical condition of the subject using the classifier based on the extracted feature maps of the image slices and the contextual information or segment each image slice using the decoder to obtain a region of interest indicative of the medical condition based on the extracted feature maps.

Abstract: A method for controlling an autonomous vehicle includes collecting data about a surrounding of the autonomous vehicle, the data including global information of where the autonomous vehicle is positioned; deriving, at least in part based on the collected data, a perceived environment based on localizing the global information; determining a driving condition of the autonomous vehicle based on the perceived environment; modifying, according to the determined driving condition, a first configuration parameter of a sensor to adjust an amount of output data from the sensor; and simultaneously adjusting, according to the determined driving condition, a second configuration parameter for processing the output data to allow consistent data capturing and processing of the output data. A control system can employ such a method, and an autonomous vehicle can utilize such a control system.

Abstract: The disclosure relates to systems and methods for vessel image analysis. The method includes receiving a set of images along a vessel acquired by a medical imaging device, and determining a sequence of centerline points along the vessel and a sequence of image patches at the respective centerline points based on the set of images. The method further includes detecting plaques based on the sequence of image patches using a first learning network. The first learning network includes an encoder configured to extract feature maps based on the sequence of image patches and a plaque range generator configured to generate a start position and an end position of each plaque based on the extracted feature maps. The method also includes classifying each detected plaque and determining a stenosis degree for the detected plaque, using a second learning network reusing at least part of the parameters of the first learning network and the extracted feature maps.

Abstract: A memory and an electronic device are provided. The memory includes a storage element (10), a first transistor (21), a second transistor (22), a first bit line (BLA), and a second bit line (BLB). The storage element (10) is coupled to the first bit line (BLA) and the second bit line (BLB) by separately using the first transistor (21) and the second transistor (22), and the first transistor (21) and the second transistor (22) are turned on during a write operation. When the foregoing solution is used, compared with providing a required write current by using one transistor, providing the write current by using the two transistors may enable a smaller transistor to meet a requirement, thereby reducing an area required by the entire memory. In addition, the memory in this application can still support a dual-port feature in a read operation.

Abstract: Embodiments of the disclosure provide systems and methods for detecting an intracerebral hemorrhage (ICH). The system includes a communication interface configured to receive a sequence of image slices and an end-to-end multi-task learning model. The sequence of image slices is the head scan images of a subject acquired by an image acquisition device. The end-to-end multi-task learning model includes an encoder, a bi-directional Convolutional Recurrent Neural Network (ConvRNN), a decoder, and a classifier. The system further includes at least one processor configured to extract feature maps from each image slice using the encoder, capture contextual information between adjacent image slices using the bi-directional ConvRNN, and detect the ICH of the subject using the classifier based on the extracted feature maps of the image slices and the contextual information or segment each image slice using the decoder to obtain an ICH region based on the extracted feature maps of the image slice.

Abstract: A power management system is disclosed for managing power in a vehicle. The system may include a power distribution unit (PDU) communicably coupled to a power controller unit located in the vehicle. The power controller unit comprises a microcontroller configured to: receive, from the PDU, a temperature value that indicates a temperature of the PDU, one or more voltage values from the one or more voltage sensors, or one or more current values from the one or more current sensors, perform a determination that the temperature value, the one or more voltage values, or the one or more current values are below one or more of their respective pre-determined threshold values, and send, after the determination, a health status message to a computer located in the vehicle, where the health status message indicates that the PDU is operating in a safe operating condition.

Abstract: The present application relates to a safety prevention and control system and control method of a power battery pack for an electric vehicle. The safety prevention and control system of a power battery pack includes a signal acquisition device, a main controller, and a step-by-step prevention and control execution device. The main controller includes a fault diagnosis device, a cell thermal runaway determination device and a battery pack thermal runaway propagation determination device, which are respectively electrically connected to the step-by-step prevention and control execution device to send different control instructions to the step-by-step prevention and control execution device. The step-by-step prevention and control execution device can execute prevention and control actions of different levels according to the different control instructions sent by the fault diagnosis device, the cell thermal runaway determination device and the battery pack thermal runaway propagation determination device.

Abstract: Embodiments of the present invention provide an index establishment method and device. The method can include determining, according to index status information of a column in a database within a preset time threshold, whether an index is to be established for the column; determining an index type according to the data information of the column; and establishing the index for the column according to the index type based on the determination that the index is to be established for the column.

Abstract: Embodiments of the present application provide a task allocation method and system. The method includes: analyzing at least one query pattern of a target task to acquire expected response time of the query pattern; estimating system cost information and estimated response time according to the query pattern and service description information; estimating node cost information of each processing node of a set of processing nodes in a computing system; selecting a processing node of the set of processing nodes according to the node cost information to allocate subtasks of the target task to the selected processing node; and determining an unallocated subtask in the target task to schedule the unallocated subtask according to the expected response time, the system cost information, and the estimated response time.

Abstract: Query requests for RDF triples are obtained, wherein the query request(s) contain(s) at least one triple pattern; for each triple pattern, the corresponding elementary pattern is determined, and each triple pattern is converted to a weighted elementary pattern. The occurrence frequency of each elementary pattern is computed based on the weighted elementary patterns; at least one elementary pattern is chosen at least according to the occurrence frequency; and the RDF triples corresponding to the chosen at least elementary pattern are prefetched into the buffer. The corresponding apparatus is also provided. With the above method and apparatus, the frequently accessed RDF triples can be determined and prefetched into the buffer, which improves the query efficiency.

Abstract: Embodiments of the disclosure provide systems and methods for biomedical image analysis. A method may include receiving a plurality of unannotated biomedical images, including a first image and a second image. The method may also include determining that the first image is in a first view and the second image is in a second view. The method may further include assigning the first image to a first processing path for the first orientation. The method may additionally include assigning the second image to a second processing path for the second view. The method may also include processing the first image in the first processing path in parallel with processing the second image in the second processing path. The first path may share processing parameters with the second path. The method may further include providing a diagnostic output based on the processing of the first image and the second image.

Abstract: Embodiments of the disclosure provide systems and methods for detecting an intracerebral hemorrhage (ICH). The system includes a communication interface configured to receive a sequence of image slices and an end-to-end multi-task learning model. The sequence of image slices is the head scan images of a subject acquired by an image acquisition device. The end-to-end multi-task learning model includes an encoder, a bi-directional Convolutional Recurrent Neural Network (ConvRNN), a decoder, and a classifier. The system further includes at least one processor configured to extract feature maps from each image slice using the encoder, capture contextual information between adjacent image slices using the bi-directional ConvRNN, and detect the ICH of the subject using the classifier based on the extracted feature maps of the image slices and the contextual information or segment each image slice using the decoder to obtain an ICH region based on the extracted feature maps of the image slice.