Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learning estimation networks in a communications system. One of the methods includes: processing first information with ground truth information to generate a first RF signal by altering the first information by channel impairment having at least one channel effect, using a receiver to process the first RF signal to generate second information, training a machine-learning estimation network based on a network architecture, the second information, and the ground truth information, receiving by the receiver a second RF signal transmitted through a communication channel including the at least one channel effect, inferring by the trained estimation network the receiver to estimate an offset of the second RF signal caused by the at least one channel effect, and correcting the offset of the RF signal with the estimated offset to obtain a recovered RF signal.

Abstract: Various examples of power converters including Integrated Capacitor Blocked Transistor (ICBT) cells and methods of control of power converters having ICBT cells are described. In one example, a power converter includes an upper arm including a plurality of upper ICBT cells connected in series to form a series connection path and a lower arm including a plurality of lower ICBT cells connected in series in the series connection path. A controller can be configured to provide a control signal pair to each of the upper ICBT cells and a complementary control signal pair to each of the lower ICBT cells to control the converter output. A capacitor voltage controller can be configured to balance a voltage potential among ICBT capacitors in at least one of the upper arm and the lower arm.

Abstract: A system and method for monitoring the health of dialysis patients with Raman spectroscopy measurements of one or more target analytes is described. The methods include irradiating one or more fluids of interest with light to produce one or more spectrum and detecting the spectrum with a detector. The fluids of interest are preferably those related to dialysis, including hemodialysis and peritoneal dialysis. In a preferred embodiment, the fluids are irradiated with monochromatic light, and one or more Raman spectra are detected as a result of the irradiation. The fluids may be irradiated within the dialysis tubing itself, or removed from the dialysis tubing and irradiated in a separate chamber. The Raman spectra of one or more target analytes of a dialysis patient may be followed over time or compared to one or more reference spectra, thereby providing information on the health of dialysis patients.

Abstract: One or more processors control processing of radio frequency (RF) signals using a machine-learning network. The one or more processors receive as input, to a radio communications apparatus, a first representation of an RF signal, which is processed using one or more radio stages, providing a second representation of the RF signal. Observations about, and metrics of, the second representation of the RF signal are obtained. Past observations and metrics are accessed from storage. Using the observations, metrics and past observations and metrics, parameters of a machine-learning network, which implements policies to process RF signals, are adjusted by controlling the radio stages. In response to the adjustments, actions performed by one or more controllers of the radio stages are updated. A representation of a subsequent input RF signal is processed using the radio stages that are controlled based on actions including the updated one or more actions.

Abstract: The present invention provides systems, methods, and devices for electroporation-based therapies (EBTs). Embodiments provide patient-specific treatment protocols derived by the numerical modeling of 3D reconstructions of target tissue from images taken of the tissue, and optionally accounting for one or more of physical constraints or dynamic tissue properties. The present invention further relates to systems, methods, and devices for delivering bipolar electric pulses for irreversible electroporation exhibiting reduced or no damage to tissue typically associated with an EBT-induced excessive charge delivered to the tissue.

Abstract: The present invention relates to relates to a polymer formulation for three-dimensionally (3D) printing an article by stereolithography, the formulation comprising a functionalized polymer. The invention further relates to lithographic methods to form 3D objects that incorporate the aforementioned polymer formulation.

Abstract: The present invention relates to the field of biomedical engineering and medical treatment of diseases and disorders. Methods, devices, and systems for in vivo treatment of cell proliferative disorders are provided. In embodiments, the methods comprise the delivery of high-frequency bursts of bipolar pulses to achieve the desired modality of cell death. More specifically, embodiments of the invention relate to a device and method for destroying aberrant cells, including tumor tissues, using high-frequency, bipolar electrical pulses having a burst width on the order of microseconds and duration of single polarity on the microsecond to nanosecond scale. In embodiments, the methods rely on conventional electroporation with adjuvant drugs or irreversible electroporation to cause cell death in treated tumors. The invention can be used to treat solid tumors, such as brain tumors.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned compact representations of radio frequency (RF) signals. One of the methods includes: determining a first RF signal to be compressed; using an encoder machine-learning network to process the first RF signal and generate a compressed signal; calculating a measure of compression in the compressed signal; using a decoder machine-learning network to process the compressed signal and generate a second RF signal that represents a reconstruction of the first RF signal; calculating a measure of distance between the second RF signal and the first RF signal; and updating at least one of the encoder machine-learning network or the decoder machine-learning network based on (i) the measure of distance between the second RF signal and the first RF signal, and (ii) the measure of compression in the compressed signal.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and deploying machine-learned classification of radio frequency (RF) signals. One of the methods includes obtaining input data corresponding to the RF spectrum; segmenting the input data into one or more samples; and for each sample of the one or more samples: obtaining information included in the sample, comparing the information to one or more labeled signal classes that are known to the machine-learning network, using results of the comparison, determining whether the information corresponds to the one or more labeled signal classes, and in response, matching, using an identification policy of a plurality of policies available to the machine-learning network, the information to a class of the one or more labeled signal classes, and providing an output that identifies an information signal corresponding to the class matching the information obtained from the sample.

Abstract: The present invention provides a process for making a flow conditioning device that transforms an input flow into a desired output flow. The process includes the steps of inputting into a computer program a set of design constraints representative of the input flow and the output flow. The computer program generates a design representative of a flow-conditioning device that transforms the input flow into the output flow. The process then provides the output design to an additive manufacturing or other suitable production system adapted to form a solid representation of the flow-conditioning device.

Abstract: Polymer resins for the vat photopolymerization of thermoplastics are provided, in particular for the vat photopolymerization of thermoplastics with exception thermal stability and mechanical properties. In some aspects, the polymer resins are prepared by ring opening of an aromatic dianhydride with an alcohol containing an acrylate or methacrylate to produce a photocrosslinkable diacid monomer; conversion of the photocrosslinkable diacid monomer to a photocrosslinkable diacyl chloride; and polymerization of the photocrosslinkable diacyl chloride with an aromatic diamine to produce a photocrosslinkable precursor polymer. Upon crosslinking and drying, a thermal imidization can yield aromatic polyimide polymers with high yield and with micron-scale structural resolution.

Abstract: The invention provides for a system for estimating a 3-dimensional treatment volume for a device that applies treatment energy through a plurality of electrodes defining a treatment area, the system comprising a memory, a display device, a processor coupled to the memory and the display device, and a treatment planning module stored in the memory and executable by the processor. In one embodiment, the treatment planning module is adapted to generate an estimated first 3-dimensional treatment volume for display in the display device based on the ratio of a maximum conductivity of the treatment area to a baseline conductivity of the treatment area. The invention also provides for a method for estimating 3-dimensional treatment volume, the steps of which are executable through the processor. In embodiments, the system and method are based on a numerical model which may be implemented in computer readable code which is executable through a processor.

Abstract: Aspects are described for hybrid modular multilevel converters that include half-bridge submodules. In some embodiments, a hybrid modular multilevel converter can include a direct current (DC) bus and an alternating current (AC) node. A first arm of the hybrid modular multilevel converter includes a first submodule chain link and a first arm inductor and a second arm includes a second submodule chain link and a second arm inductor. A capacitor connects between a first side of the first arm and a first side of the second arm.

Abstract: This disclosure describes the methods, devices, and systems of treating diseased tissue with integrated nanosecond pulse irreversible electroporation. Methods and systems as disclosed provide MRI compatible shielded electrodes and electrode leads to prevent emanating radiofrequency noise and improve image quality, disconnecting the electrode from the cable linkage to the pulse generator reduce electromagnetic interference and image artifacts, placing electrodes strategically within a guide cannula to minimize distortion from heterogeneities or maximize ablation within the tissue, utilizing conductive fluids, innate or external, such as cerebral spinal fluid or grounding pads to provide a pathway for current return, and for timing of the electrical waveforms with inherent brain electrical activity.

Abstract: A method of making an article comprising one or more layers of plasmonic nanoparticles located between opposing layers of dielectric materials and an article comprising one or more layers of plasmonic nanoparticles located between opposing layers of dielectric materials.

Abstract: Various methods, systems and apparatus are provided for imaging and sensing using interferometry. In one example, a system includes an interferometer; a light source that can provide light to the interferometer at multiple wavelengths (?i); and optical path delay (OPD) modifying optics that can enhance contrast in an interferometer output associated with a sample. The light can be directed to the sample by optics of the interferometer. The interferometer output can be captured by a detector (e.g., a camera) at each of the multiple wavelengths (?i). In another example, an apparatus includes an add-on unit containing OPD that can enhance contrast in an interferometer output associated with a sample illuminated by light at a defined wavelength (?i). A detector can be attached to the add-on unit to record the interferometer output at the defined wavelength (?i).

Abstract: Aspects of series resonator DC-to-DC converters are described. A series resonator DC-to-DC converter can include a first half-bridge circuit comprising a first high-side switch and a first low-side switch, a second half-bridge circuit comprising a second high-side switch and a second low-side switch, and a resonator in series between the first high-side switch and the first low-side switch. The circuit design and switching controller can be relied upon to impart soft-switching.

Abstract: A voltage balancing circuit for use in a power converter is described. In one example, a power converter includes series-connected switching transistors for power conversion, and a voltage balancing control loop. The voltage balancing control loop includes a measurement circuit electrically coupled to a transistor in the pair of series-connected switching transistors. The measurement circuit is electrically coupled to measure a body voltage reference of the transistor. The voltage balancing control loop also includes a balancing circuit configured to generate a balancing pulse signal for adjusting a voltage across the transistor using the body voltage reference, and a circuit configured to combine the balancing pulse signal with a gate drive pulse signal for the transistor, to form a balanced gate drive pulse signal for the transistor. The balanced gate drive pulse signal helps to equalize the body diode voltages of the series-connected switching transistors, particularly during “off” periods.

Abstract: Aspects of an efficient, wide voltage range, power converter system are described. In one example, a power converter system includes a first power converter, a second power converter, and a controller for the power converter. An input of the first power converter and an input of the second power converter are connected in series across an input voltage for the power converter system, and an output of the first power converter and an output of the second power converter are connected in parallel at an output of the power converter system. The controller is configured to regulate the second power converter and to determine whether or not to regulate the first power converter based on the input voltage for the power converter system and an output voltage of the power converter system, among other factors, for greater efficiency of the power converter system over wider input and output voltage ranges.

Abstract: The present invention provides stationary CT architecture for imaging at a faster temporal resolution and lower radiation dose. In embodiments, the architecture features stationary distributed x-ray sources and rotating x-ray detectors. Provided is a stationary source computed tomography (CT) architecture comprising: a detector disposed on a rotatable gantry; an x-ray source disposed on a fixed ring; wherein the detector is disposed on the gantry in a manner such that the detector is capable of rotating around a subject and of receiving a signal from the x-ray source. Embodiments of the invention include a CT-MRI scanner comprising the stationary CT architecture.