Abstract: Systems and methods are provided for optimizing a deep learning model. A multi-site dataset associated with different clinical sites and a deployment dataset associated with a deployment clinical site are received. A deep learning model is trained based on the multi-site dataset. The trained deep learning model is optimized based on the deployment dataset. The optimized trained deep learning model is output.

Abstract: Embodiments of this application provide an unlicensed frequency band-based resource configuration method and an apparatus, to resolve a problem of configuring a fixed frame period FFP for a terminal device operating in an unlicensed frequency band in a current technology. The method includes: A network device sends fixed frame period FFP configuration information to a terminal device, where the FFP configuration information indicates an FFP configuration of the terminal device, an FFP is a period used by the terminal device to transmit a signal, the FFP includes channel occupancy time COT and an idle period Idle period, the channel occupancy time COT is used by the terminal device to transmit the signal, and the idle period is used by the terminal device to perform LBT.

Abstract: A random access method is described that includes a network device determines a first channel access type used by a terminal device to access a first unlicensed frequency band, where the first channel access type is one of channel access types of an unlicensed frequency band, and there are two channel access types of the unlicensed frequency band: frame based equipment (FBE) and load based equipment (LBE). The network device sends a first system message to the terminal device, where the first system message carries first indication information, and the first indication information is used to indicate the first channel access type. The terminal device obtains the first system message from the network device, and accesses the first unlicensed frequency band by using the first channel access type indicated by the first indication information.

Abstract: Embodiments of the present invention provide a bandwidth allocation method and apparatus, user equipment, and a base station. The method includes: receiving, by user equipment, virtual bandwidth configuration information, where the virtual bandwidth configuration information is used for indicating a configuration of a virtual bandwidth; and performing, by the user equipment, signal reception and/or signal processing according to the virtual bandwidth indicated by the virtual bandwidth configuration information, where each virtual bandwidth is a part or all of a downlink maximum available bandwidth or a downlink transmission bandwidth, and the downlink maximum available bandwidth is a maximum bandwidth available for downlink transmission. In the embodiments of the present invention, reduction of overheads and reduction of complexity of signal reception and signal processing are implemented, and overheads and complexity of signal reception and processing feedback can be flexibly controlled.

Abstract: A random access method is described that includes a network device determines a first channel access type used by a terminal device to access a first unlicensed frequency band, where the first channel access type is one of channel access types of an unlicensed frequency band, and there are two channel access types of the unlicensed frequency band: frame based equipment (FBE) and load based equipment (LBE). The network device sends a first system message to the terminal device, where the first system message carries first indication information, and the first indication information is used to indicate the first channel access type. The terminal device obtains the first system message from the network device, and accesses the first unlicensed frequency band by using the first channel access type indicated by the first indication information.

Abstract: In an method for training artificial intelligence entities (AIE) for abnormality detection, medical imaging data of the human organ is provided as training data having training samples, the medical imaging data including imaging results from different types of imaging techniques for each training sample of the training data, a pre-trained or randomly initialized AIE is provided, and the AIE is trained using the provided training samples. The training may include, for at least one training sample, a first loss function for a sub-structure of the AIE is calculated independently of a first spatial region of the human organ, and, for a training sample, a second loss function for a sub-structure of the AIE is calculated independently of a second spatial region of the human organ. The AIE may be trained using the calculated first loss function and the calculated second loss function.

Abstract: Embodiments of the present invention provides a method and an apparatus for channel access for the transmission of a signal. The method includes: counting down, by a first device, an initial value N of a random backoff counter within a contention window; sensing, by the first device, a channel status; determining, by the first device, whether or not to perform transmission of the signal within a periodic window based on the sensed channel status; transmitting, by the first device, the signal after sensing the channel to be idle. Therefore, the channel access method and apparatus adapt to the data and control traffic load in the channel using LBT with random back-off with a contention window of variable size.

Abstract: Similar pre-stored medical datasets are identified by comparison with a current case dataset. A current case dataset is provided and includes radiological data of a patient. A number of pre-stored medical datasets each including radiological data of other patients are provided. Each case dataset is evaluated according to a predefined AI-based method to obtain a number of definitive features for that case dataset. The definitive features of the current case dataset are compared with the definitive features of each pre-stored medical dataset to identify a number of pre-stored medical datasets most similar to the current case dataset. The identified number of most similar pre-stored medical datasets are output.

Abstract: Embodiments of the present invention provide a bandwidth allocation method and apparatus, user equipment, and a base station. The method includes: receiving, by user equipment, virtual bandwidth configuration information, where the virtual bandwidth configuration information is used for indicating a configuration of a virtual bandwidth; and performing, by the user equipment, signal reception and/or signal processing according to the virtual bandwidth indicated by the virtual bandwidth configuration information, where each virtual bandwidth is a part or all of a downlink maximum available bandwidth or a downlink transmission bandwidth, and the downlink maximum available bandwidth is a maximum bandwidth available for downlink transmission. In the embodiments of the present invention, reduction of overheads and reduction of complexity of signal reception and signal processing are implemented, and overheads and complexity of signal reception and processing feedback can be flexibly controlled.

Abstract: The technology of this application relates to a reference signal sending method and apparatus and a reference signal configuration method and apparatus. The reference signal sending method includes a terminal device switching from a first carrier of a first serving cell to a second carrier of a non-serving cell. The terminal device sends a sounding reference signal (SRS) by using the second carrier. According to the method and apparatus in this application, a communication bandwidth is decoupled from a positioning bandwidth through SRS carrier switching of a non-serving cell, thereby meeting a high-precision positioning requirement of an Internet-of-Things (IoT)-type terminal device.

Abstract: Provided is a method for initiating a mobile voice call from a caller UE to a callee UE having a SIM card associated with a callee phone number, which callee phone number is a E.164 compliant MSISDN. The method includes providing a software function at the callee UE, which software function is associated with a receiving address comprising the callee phone number, and configuring a call forwarding function in a first network, being activated for the callee phone number and configured for forwarding CS voice calls to a forwarding phone number associated with a VOIP service. The method also includes the caller UE calling the callee phone number, the first network performing the call forwarding to the VOIP service, the VOIP service identifying the callee UE, and the VOIP service initiating the call between the caller UE and the software function using the receiving address.

Abstract: Embodiments of the present invention provide a bandwidth allocation method and apparatus, user equipment, and a base station. The method includes: receiving, by user equipment, virtual bandwidth configuration information, where the virtual bandwidth configuration information is used for indicating a configuration of a virtual bandwidth; and performing, by the user equipment, signal reception and/or signal processing according to the virtual bandwidth indicated by the virtual bandwidth configuration information, where each virtual bandwidth is a part or all of a downlink maximum available bandwidth or a downlink transmission bandwidth, and the downlink maximum available bandwidth is a maximum bandwidth available for downlink transmission. In the embodiments of the present invention, reduction of overheads and reduction of complexity of signal reception and signal processing are implemented, and overheads and complexity of signal reception and processing feedback can be flexibly controlled.

Abstract: Systems and methods are provided for classifying an abnormality in a medical image. An input medical image depicting a lesion is received. The lesion is localized in the input medical image using a trained localization network to generate a localization map. The lesion is classified based on the input medical image and the localization map using a trained classification network. The classification of the lesion is output. The trained localization network and the trained classification network are jointly trained.

Abstract: Embodiments of the present invention provide a bandwidth allocation method and apparatus, user equipment, and a base station. The method includes: receiving, by user equipment, virtual bandwidth configuration information, where the virtual bandwidth configuration information is used for indicating a configuration of a virtual bandwidth; and performing, by the user equipment, signal reception and/or signal processing according to the virtual bandwidth indicated by the virtual bandwidth configuration information, where each virtual bandwidth is a part or all of a downlink maximum available bandwidth or a downlink transmission bandwidth, and the downlink maximum available bandwidth is a maximum bandwidth available for downlink transmission. In the embodiments of the present invention, reduction of overheads and reduction of complexity of signal reception and signal processing are implemented, and overheads and complexity of signal reception and processing feedback can be flexibly controlled.

Abstract: A method and an apparatus for transmitting or receiving information are provided. The method includes: receiving, by a terminal device, indication information from a network device, where the indication information indicates a timing adjustment parameter, the timing adjustment parameter is to be used by the terminal device to determine transmit timings on a first uplink carrier and a second uplink carrier, the first uplink carrier is an uplink carrier of a first radio access technology, and the second uplink carrier is an uplink carrier of a second radio access technology; and determining, by the terminal device, the timing adjustment parameter based on the indication information.

Abstract: Systems and methods are provided for classifying an abnormality in a medical image. An input medical image depicting a lesion is received. The lesion is localized in the input medical image using a trained localization network to generate a localization map. The lesion is classified based on the input medical image and the localization map using a trained classification network. The classification of the lesion is output. The trained localization network and the trained classification network are jointly trained.

Abstract: An information receiving method and an apparatus are disclosed. The method includes: receiving, by a terminal device, first power information and second power information from a network device; and determining a first maximum transmission power based on the first power information, and determining a second maximum transmission power based on the second power information, where the first maximum transmission power is a maximum transmission power to be used by the terminal device for transmitting a signal by a first radio access technology, and the second maximum transmission power is a maximum transmission power to be used by the terminal device for transmitting a signal by a second radio access technology.

Abstract: Embodiments of this application provide an unlicensed frequency band-based resource configuration method and an apparatus, to resolve a problem of configuring a fixed frame period FFP for a terminal device operating in an unlicensed frequency band in a current technology. The method includes: A network device sends fixed frame period FFP configuration information to a terminal device, where the FFP configuration information indicates an FFP configuration of the terminal device, an FFP is a period used by the terminal device to transmit a signal, the FFP includes channel occupancy time COT and an idle period Idle period, the channel occupancy time COT is used by the terminal device to transmit the signal, and the idle period is used by the terminal device to perform LBT.

Abstract: Systems and methods are provided for classifying an abnormality in a medical image. An input medical image depicting a lesion is received. The lesion is localized in the input medical image using a trained localization network to generate a localization map. The lesion is classified based on the input medical image and the localization map using a trained classification network. The classification of the lesion is output. The trained localization network and the trained classification network are jointly trained.

Abstract: A contention window maintenance method and a communications apparatus are provided. A contention window maintenance method includes maintaining, by a communications apparatus in a first time window, a contention window value based on responses of N pieces of first reference data sent by the communications apparatus within a first reference time, N being an integer greater than 0, the responses comprising an ACK and/or a NACK, and the contention window value is maintained based on a proportion of ACKs in the received responses of the N pieces of first reference data and/or a proportion of NACKs in the received responses of the N pieces of first reference data.