Abstract: A varistor is provided having a rectangular configuration defining first and second opposing end surfaces offset in a lengthwise direction. The varistor may include a first terminal adjacent the first opposing end surface and a second terminal adjacent the second opposing end surface. The varistor may include an active electrode layer including a first electrode electrically connected with the first terminal and a second electrode electrically connected with the second terminal. The first electrode may be spaced apart from the second electrode in the lengthwise direction to form an active electrode end gap. The varistor may include a floating electrode layer including a floating electrode. The floating electrode layer may be spaced apart from the active electrode layer in a height-wise direction to form a floating electrode gap. A ratio of the active electrode end gap to the floating electrode gap may be greater than about 2.

Abstract: A radio frequency (RF) generator includes a RF power source configured to output an RF power signal, and a controller coupled to the RF power source. The controller is configured to generate a pulse to modulate the RF power signal of the RF power source. The pulse includes one or more state transitions. The controller is further configured to receive a sync signal indicative of one or more operating characteristics or parameters of another RF generator, and adjust at least one of the state transitions of the pulse to synchronize the state transition with a defined phase of the received sync signal. Other example RF generators, RF power delivery systems including one or more RF generators, and control methods for adjusting a state transition of a pulse to synchronize the state transition with a defined phase of a sync signal are also disclosed.

Abstract: Systems and methods for automated generation and customization of interfaces are disclosed herein. The method includes receiving a request to launch an interface and identifying a subject of the interface and a viewing user of the interface. Based on attributes of one or both of the subject and the viewing user, a page is selected and customized. One or several containers associated with the page are identified. Data associated with the containers is retrieved and customized, and the container is likewise customized. A customized view is generated based on the customized page, the customized one or more containers, and the customized data.

Abstract: Disclosed herein are methods for radiotherapy treatment plan optimization for irradiating one or more target regions using both an internal therapeutic radiation source (ITRS) and an external therapeutic radiation source (ETRS). One variation of a method comprises iterating through ITRS radiation dose values and ETRS radiation dose values to attain a cumulative dose that meets prescribed dose requirements. In some variations, an ITRS is an injectable compound that has a targeting backbone and a radionuclide, and images acquired using an imaging compound that has the same targeting backbone as the injectable compound can be used to calculate the radiation dose deliverable using the injectable ITRS, and also to calculate firing filters for delivering radiation using a biologically-guided radiation therapy (BGRT) system. Image data acquired from a previous treatment session may be used to adapt the dose provided by an ITRS and/or ETRS for a future treatment session.

Abstract: Disclosed herein are systems and methods for adapting and/or updating radiotherapy treatment plans based on biological and/or physiological data and/or anatomical data extracted or calculated from imaging data acquired in real-time (e.g., during a treatment session). Functional imaging data acquired at the time of radiation treatment is used to modify a treatment plan and/or dose delivery instructions to provide a prescribed dose distribution to patient target regions. Also disclosed herein are methods for evaluating treatment plans based on imaging data acquired in real-time.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic biometric identifiers. The method includes generating a scannable code upon receiving a success nonce from a registration server. The registration server can access a user identifier and a hash of at least a signature using the success nonce. The signature can be generated based at least in part upon a biometric identifier of a user. The method includes recreating the hash of the signature stored by the registration server. The method includes generating the scannable code by encrypting the success nonce and the recreated hash. The biometric identifier of the user is generated by feeding a plurality of non-deterministic biometric inputs to a trained machine learning model producing a plurality of feature vectors. The method includes projecting the plurality of feature vectors onto a surface of a unit hyper-sphere and computing a characteristic identity vector representing the user.

Abstract: Described herein are methods for beam station delivery of radiation treatment, where the patient platform is moved to a series of discrete patient platform locations or beam stations that are determined during treatment planning, stopped at each of these locations while the radiation source rotates about the patient delivering radiation to the target regions that intersect the radiation beam path, and then moving to the next location after the prescribed dose of radiation (e.g., in accordance with a calculated fluence map) for that location has been delivered to the patient.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: Described herein are methods for beam station delivery of radiation treatment, where the patient platform is moved to a series of discrete patient platform locations or beam stations that are determined during treatment planning, stopped at each of these locations while the radiation source rotates about the patient delivering radiation to the target regions that intersect the radiation beam path, and then moving to the next location after the prescribed dose of radiation (e.g., in accordance with a calculated fluence map) for that location has been delivered to the patient.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: Disclosed herein are methods for patient setup and patient target region localization for the irradiation of multiple patient target regions in a single treatment session. Virtual localization is a method that can be used to register a patient target region without requiring that the patient is physically moved using the patient platform. Instead, the planned fluence is updated to reflect the current location of the patient target region by selecting a localization reference in the localization image, calculating a localization function based on the localization reference point, and calculating the delivery fluence by convolving the localization function with a shift-invariant firing filter. Mosaic multi-target localization partitions a planned fluence map for multiple patient target regions into sub-regions that can be individually localized.

Abstract: Disclosed are apparatus and a method for providing an integrated multiterminal multilayer ceramic device that has three or more capacitive elements. Two of such capacitive elements may be in series, with a third in parallel. The integrated device may be packaged as an overmolded three leaded component, or can be mounted as SMD (surface mount device). The integrated device may also be combined with a separate varistor in a stacked arrangement of leaded components.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: Described herein are methods for monitoring the radiation delivery during a radiotherapy delivery session and providing a graphical representation of radiation delivery to an operator (e.g., a clinician, a medical physicist, a radiation therapy technologist). The graphics are updated in real-time, as radiation data is collected by the radiotherapy system, and in some variations, can be updated every 15 minutes or less. A variety of graphical representations (“graphics”) can be used to indicate the status of radiation delivery relative to the planned radiation delivery. Methods optionally include calculating a range of acceptable metric values, generating graphics that represent the range of acceptable metrics values, and generating a graphic that depicts the real-time values of those metrics overlaid with the range of acceptable metrics values.

Abstract: The technology disclosed provides systems and methods related to preventing exfiltration of training data by feature reconstruction attacks on model instances trained on the training data during a training job. The system comprises a privacy interface that presents a plurality of modulators for a plurality of training parameters. The modulators are configured to respond to selection commands via the privacy interface to trigger procedural calls. The procedural calls modify corresponding training parameters in the plurality of training parameters for respective training cycles in the training job. The system comprises a trainer configured to execute the training cycles in dependence on the modified training parameters. The trainer can determine a performance accuracy of the model instances for each of the executed training cycles.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: Disclosed herein are methods for patient setup and registration for the irradiation of target tissue regions. A method for positioning a patient for radiation therapy may include acquiring an image of a first patient target region and a second patient target region. A first set of patient position-shift vectors may be calculated based on the acquired image and a treatment planning image of the first patient target region. A second set of patient position-shift vectors may be calculated based on the acquired image, a treatment planning image of the second patient target region, and the first set of patient position-shift vectors. The patient may be positioned according to the first set of patient position-shift vectors in a first location. The patient may be moved to a second location and positioned according to the second set of patient position-shift vectors.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.