Abstract: The present disclosure provides a method, a device, and a storage medium for prior-guided dual-path network (PDNet). The method includes inputting an image into a split-attention network to extract a feature map at each scale and compressing the feature map to form a compressed feature map of each scale, by an image encoder, inputting the compressed feature map and a three-channel image into a prior encoder to generate an attention enhanced feature map of each scale, and outputting the attention enhanced feature map to a decoder; concatenating, by the decoder, an attention enhanced feature map at a current scale, in combination with up-sampled feature maps and/or down-sampled feature maps from other scales, to form a concatenated feature map of the current scale; and attaching a deconvolutional layer to a highest-level scale SA to segment a lesion and predict a RECIST diameter based on concatenated feature maps.

Abstract: Systems and methods are described for managing charging and discharging of battery packs. In one or more aspects, a system and method are provided to minimize overcharging of battery cells of specific battery chemistries while still enabling fast charging cycles. In other aspects, a buck converter may be used to reduce a voltage of power used to charge the cells. In further aspects, a fast overcurrent protection circuit is described to address situations involving internal short circuits of a battery cell or battery pack. In yet further aspects, a bypass circuit is provided in series-connected battery packs to improve the charging of undercharged battery packs while also increasing the efficiency of the overall charging process. In other aspects, a circuit is provided that permits a controller to determine a configuration of battery packs.

Abstract: A semiconductor module includes: a dual-gauge leadframe having thicker and thinner parts, part of the thinner part forming a high voltage lead; a semiconductor die attached to the thicker part; and a molding compound (MC) encapsulating the die. The thicker leadframe part is disposed at a bottom side of the MC. A side face of the MC has a stepped region between the high voltage lead and thicker leadframe part. A first generally vertical part of the stepped region extends from the high voltage lead to the generally horizontal part, a generally horizontal part of the stepped region extends to the second generally vertical part, and a second generally vertical part of the stepped region extends to the bottom side of the MC. A linear dimension of the generally horizontal part as measured from the first generally vertical part to the second generally vertical part is at least 4.5 mm.

Abstract: The present disclosure provides a computer-implemented method, a device, and a computer program product for deep lesion tracker. The method includes inputting a search image into a first three-dimensional DenseFPN (feature pyramid network) of an image encoder and inputting a template image into a second three-dimensional DenseFPN of the image encoder to extract image features; encoding anatomy signals of the search image and the template image as Gaussian heatmaps, and inputting the Gaussian heatmap of the template image into a first anatomy signal encoders (ASE) and inputting the Gaussian heatmap of the search image into a second ASE to extract anatomy features; inputting the image features and the anatomy features into a fast cross-correlation layer to generate correspondence maps, and computing a probability map according to the correspondence maps; and performing supervised learning or self-supervised learning to predict a lesion center in the search image.

Abstract: A method for performing a computer-aided diagnosis (CAD) for universal lesion detection includes: receiving a medical image; processing the medical image to predict lesion proposals and generating cropped feature maps corresponding to the lesion proposals; for each lesion proposal, applying a plurality of lesion detection classifiers to generate a plurality of lesion detection scores, the plurality of lesion detection classifiers including a whole-body classifier and one or more organ-specific classifiers; for each lesion proposal, applying an organ-gating classifier to generate a plurality of weighting coefficients corresponding to the plurality of lesion detection classifiers; and for each lesion proposal, performing weight gating on the plurality of lesion detection scores with the plurality of weighting coefficients to generate a comprehensive lesion detection score.

Abstract: Systems and methods are described for managing charging and discharging of battery packs. In one or more aspects, a system and method are provided to minimize overcharging of battery cells of specific battery chemistries while still enabling fast charging cycles. In other aspects, a buck converter may be used to reduce a voltage of power used to charge the cells. In further aspects, a fast overcurrent protection circuit is described to address situations involving internal short circuits of a battery cell or battery pack. In yet further aspects, a bypass circuit is provided in series-connected battery packs to improve the charging of undercharged battery packs while also increasing the efficiency of the overall charging process. In other aspects, a circuit is provided that permits a controller to determine a configuration of battery packs.

Abstract: The present disclosure provides a method, a device, and a computer program product using a self-supervised anatomical embedding (SAM) method.

Abstract: The present disclosure provides a computer-implemented method, a device, and a computer program product for deep lesion tracker. The method includes inputting a search image into a first three-dimensional DenseFPN (feature pyramid network) of an image encoder and inputting a template image into a second three-dimensional DenseFPN of the image encoder to extract image features; encoding anatomy signals of the search image and the template image as Gaussian heatmaps, and inputting the Gaussian heatmap of the template image into a first anatomy signal encoders (ASE) and inputting the Gaussian heatmap of the search image into a second ASE to extract anatomy features; inputting the image features and the anatomy features into a fast cross-correlation layer to generate correspondence maps, and computing a probability map according to the correspondence maps; and performing supervised learning or self-supervised learning to predict a lesion center in the search image.

Abstract: A package is disclosed. In one example, the package includes a first main face for mounting a heat sink and an opposing second main face for being mounted on a mounting base. The package comprises a carrier, an electronic component mounted at the carrier, and an encapsulant encapsulating at least part of the electronic component and at least part of the carrier. Electrically insulating material covers electrically conductive material of the carrier at said first main face. The encapsulant comprises at least one step at the first main face.

Abstract: An interconnect clip includes a die attach pad that comp includes rises a die attach surface at an inner side of the interconnect clip, a heat dissipation pad that includes a heat dissipation surface at an outer side of the interconnect clip, and a lead contact pad that includes a lead contact surface at an inner side of the interconnect clip or at an outer side of the interconnect clip. The outer side of the interconnect clip in the lead contact pad faces and is spaced apart from the inner side of the interconnect clip in the heat dissipation pad, and the inner side of the interconnect clip in the lead contact pad faces and is spaced apart from the outer side of the interconnect clip in the die attach pad.

Abstract: A method, apparatus and device for synchronously playing audio are provided. The method includes: acquiring status information of a Bluetooth controller in the first audio playing device and Bluetooth time from the Bluetooth controller, and determining native Bluetooth time of the first audio playing device based on the status information and the Bluetooth time acquired from the Bluetooth controller; estimating, based on the native Bluetooth time of the first audio playing device and offset time of the native Bluetooth time of the first audio playing device relative to native Bluetooth time of a second audio playing device or native Bluetooth time of an audio providing device, absolute time of the second audio playing device or absolute time of the audio providing device; and playing an audio signal provided by the audio providing device synchronously with the second audio playing device based on the estimated absolute time.

Abstract: Systems and methods for characterizing a region of interest (ROI) in a medical image are provided. An exemplary system may include a memory storing instructions and at least one processor communicatively coupled to the memory to execute the instructions which, when executed by the processor, may cause the processor to perform operations. The operations may include detecting one or more candidate ROIs from the medical image using a three-dimensional (3D) machine learning network. The operations may also include determining a key slice for each candidate ROI. The operations may further include selecting a primary ROI from the one or more candidate ROIs based on the respective key slices. In addition, the operations may include classifying the primary ROI into one of a plurality of categories using a texture-based classifier based on the key slice corresponding to the primary ROI.

Abstract: A method, apparatus and device for synchronously playing audio are provided. The method includes: acquiring status information of a Bluetooth controller in the first audio playing device, and determining native Bluetooth time of the first audio playing device based on the status information; estimating, based on the native Bluetooth time of the first audio playing device and offset data of the native Bluetooth time of the first audio playing device relative to native Bluetooth time of a second audio playing device or native Bluetooth time of an audio providing device, absolute time of the second audio playing device or absolute time of the audio providing device; and playing an audio signal provided by the audio providing device synchronously with the second audio playing device based on the estimated absolute time.

Abstract: A clip for a semiconductor package and a semiconductor having a clip is disclosed. In one example, the clip includes a first planar portion, a plurality of first pillars, and a plurality of first solder balls. Each first pillar of the plurality of first pillars is coupled to the first planar portion. Each first solder ball of the plurality of first solder balls is coupled to a corresponding first pillar of the plurality of first pillars.

Abstract: The present disclosure relates to a key generation method applied to a chip storing an internal key, which includes: acquiring (101) a first and a second random number; generating (102) an initial private key of the chip according to the first random number and the internal key, and generating an update private key of the chip according to the second random number and the internal key; generating (103) an initial public key corresponding to the initial private key according to the initial private key, and generating an update public key corresponding to the update private key according to the update private key; and sending (104) the initial public key, the update public key, and the second random number to a server that sends the second random number to the chip when receiving a private key update request of the chip, to trigger a private key update of the chip.

Abstract: Systems and methods for characterizing a region of interest (ROI) in a medical image are provided. An exemplary system may include a memory storing instructions and at least one processor communicatively coupled to the memory to execute the instructions which, when executed by the processor, may cause the processor to perform operations. The operations may include detecting one or more candidate ROIs from the medical image using a three-dimensional (3D) machine learning network. The operations may also include determining a key slice for each candidate ROI. The operations may further include selecting a primary ROI from the one or more candidate ROIs based on the respective key slices. In addition, the operations may include classifying the primary ROI into one of a plurality of categories using a texture-based classifier based on the key slice corresponding to the primary ROI.

Abstract: A battery system with a large-format Li-ion battery powers attached equipment by discharging battery cells distributed among a plurality of battery packs. The discharging of the battery cells is controlled in an efficient manner while preserving the expected life of the Li-ion battery cells. Each battery pack internally supports a battery management system and may have identical components, thus supporting an architecture that easily scales to higher power/energy. Battery packs may be added or removed without intervention with a user, where one of battery packs serves as a master battery pack and the remaining battery packs serve as slave battery packs. When the master battery pack is removed, one of the slave battery packs becomes the master battery pack. Charging and discharging of the battery cells is coordinated by the master battery pack with the slave battery packs over a communication channel such as a controller area network (CAN) bus.

Abstract: An angling system comprising a wearable band and at least one detector selected from: at least one fish bite detector, wherein the or each fish bite detector comprises a fish bite sensor connected to a transceiver, and wherein the or each fish bite detector is configured to transmit a signal upon the fish bite sensor detecting a fish bite; and a motion detector, wherein the motion detector comprises a motion sensor connected to a transceiver, and wherein the motion detector is configured to transmit a signal upon the motion sensor detecting motion, wherein the wearable band comprises: at least one strap; and a wireless communication unit, wherein the wireless communication unit is attached or attachable to the at least one strap, wherein the wireless communication unit comprises a transceiver, at least one actuator, and at least one of a haptic actuator connected to the transceiver and an alarm connected to the transceiver, wherein the transceiver of the wireless communication unit is pairable with the transc

Abstract: The present disclosure describe embodiments of a laser projection apparatus. The apparatus includes: an apparatus housing, a laser source system, an optical illumination system and a lens system. The laser source system includes: a first laser assembly configured to emit a first laser beam, a second laser assembly configured to emit a second laser beam and a third laser assembly configured to emit a third laser beam; a first laser combining lens configured to reflect the third laser beam; a second laser combining lens configured to transmit the second laser beam and reflect the third laser beam; and a third laser combining lens configured to transmit the first laser beam and reflect the second and third laser beams. The optical illumination system is configured to modulate laser beams to generate a modulated beam. The lens system is configured to project the modulated beam onto a projection screen.

Abstract: A method of harvesting lesion annotations includes conditioning a lesion proposal generator (LPG) based on a first two-dimensional (2D) image set to obtain a conditioned LPG, including adding lesion annotations to the first 2D image set to obtain a revised first 2D image set, forming a three-dimensional (3D) composite image according to the revised first 2D image set, reducing false-positive lesion annotations from the revised first 2D image set according to the 3D composite image to obtain a second-revised first 2D image set, and feeding the second-revised first 2D image set to the LPG to obtain the conditioned LPG, and applying the conditioned LPG to a second 2D image set different than the first 2D image set to harvest lesion annotations.