Rotary motor based transdermal injection device

    公开(公告)号:US12115353B2

    公开(公告)日:2024-10-15

    申请号:US17102958

    申请日:2020-11-24

    摘要: An apparatus for injectate delivery includes a cartridge, a linear actuator, a rotary motor mechanically coupled the actuator, and a controller coupled to the motor. The controller controls a linear motion of the actuator by controlling an electrical input supplied to the motor in a first interval during which the motor is stationary with the linear actuator engaged with the cartridge to displace an injectate in the cartridge, a second interval following the first interval during which the controller accelerates the motor from stationary to a first speed selected to create a jet of the injectate from the cartridge with a velocity sufficient to pierce human tissue to a subcutaneous depth, a third interval during which the controller maintains the motor at or above the first speed, and a fourth interval during which the controller decelerates the motor to a second speed to deliver the injectate at the subcutaneous depth.

    Aggregating security events
    4.
    发明授权

    公开(公告)号:US12095731B2

    公开(公告)日:2024-09-17

    申请号:US17699414

    申请日:2022-03-21

    申请人: Sophos Limited

    IPC分类号: H04L9/40

    摘要: A stream of events is received at a local security agent running on an endpoint at an enterprise network. The local security agent may detect an event of a first event type and may generate an aggregate event with subsequent events of the first event type in the stream. The local security agent may then transmit the aggregate event to a security resource for detecting security threats.

    Malware detection using machine learning

    公开(公告)号:US11818165B2

    公开(公告)日:2023-11-14

    申请号:US17098084

    申请日:2020-11-13

    申请人: Sophos Limited

    发明人: Joseph H. Levy

    IPC分类号: H04L9/40 G06N20/00 G06F21/56

    摘要: Synthetic training sets for machine learning are created by identifying and modifying functional features of code in an existing malware training set. By filtering the resulting synthetic code to measure malware impact and novelty, training sets can be created that predict novel malware and to seek to preemptively exhaust the space of new malware. These synthesized training sets can be used in turn to improve training of machine learning models. Furthermore, by repeating the process of new code generation, filtering and training, an iterative machine learning process may be created that continuously narrows the window of vulnerabilities to new malicious actions.