DEVICE, SYSTEM, AND METHOD FOR PASSIVE COLLECTION OF ATMOSPHERIC CARBON DIOXIDE

    公开(公告)号:US20250058266A1

    公开(公告)日:2025-02-20

    申请号:US18815777

    申请日:2024-08-26

    Abstract: A device for passive collection of atmospheric carbon dioxide is disclosed. The device includes a release chamber having an opening and a sorbent regeneration system. The device also includes a capture structure coupled to the release chamber, having at least one collapsible support and a plurality of tiles spaced along the collapsible support. Each tile has a sorbent material. The capture structure is movable between a collection configuration and a release configuration. The collection configuration includes the capture structure extending upward from the release chamber to expose the capture structure to an airflow and allow the sorbent material to capture atmospheric carbon dioxide. The release configuration includes the collapsible support being collapsed and the plurality of tiles being sufficiently enclosed inside the release chamber that the sorbent regeneration system may operate on the plurality of tiles to release captured carbon dioxide from the sorbent material and form an enriched gas.

    Systems, methods, and apparatuses for actively and continually fine-tuning convolutional neural networks to reduce annotation requirements

    公开(公告)号:US12216737B2

    公开(公告)日:2025-02-04

    申请号:US17698805

    申请日:2022-03-18

    Abstract: Described herein are systems, methods, and apparatuses for actively and continually fine-tuning convolutional neural networks to reduce annotation requirements, in which the trained networks are then utilized in the context of medical imaging. The success of convolutional neural networks (CNNs) in computer vision is largely attributable to the availability of massive annotated datasets, such as ImageNet and Places. However, it is tedious, laborious, and time consuming to create large annotated datasets, and demands costly, specialty-oriented skills. A novel method to naturally integrate active learning and transfer learning (fine-tuning) into a single framework is presented to dramatically reduce annotation cost, starting with a pre-trained CNN to seek “worthy” samples for annotation and gradually enhances the (fine-tuned) CNN via continual fine-tuning. The described method was evaluated using three distinct medical imaging applications, demonstrating that it can reduce annotation efforts by at least half compared with random selection.

    LENGTH-SELECTIVE DIELECTROPHORETIC MANIPULATION OF SINGLE-WALLED CARBON NANOTUBES

    公开(公告)号:US20250033971A1

    公开(公告)日:2025-01-30

    申请号:US18913675

    申请日:2024-10-11

    Abstract: Systems & methods for sorting single-walled carbon nanotubes (SWNTs) using an iDEP-based sorting device. The device includes an inlet channel with a constriction and the inlet channel splits into multiple different channels after the constriction—the multiple channels includes a center channel and at least one side channel. A sample is introduced into the iDEP sorting device containing a plurality of SWNTs of different lengths suspended in a fluid. An electrical field is applied to the sample between a first electrode in the center channel and a second electrodes at a proximal end of the inlet channel. The applied electrical field causes longer SWNTs to move towards the side channels while the shorter SWNTs move towards the center channel. Accordingly, a first plurality of shorter SWNTs is then collected from the center channel and a second plurality of longer SWNTs is collected from the at least one side channel.

    FAST TRACKING PLL WITH ANALOG MIXER FOR PHASE DETECTION

    公开(公告)号:US20250023571A1

    公开(公告)日:2025-01-16

    申请号:US18714338

    申请日:2022-11-30

    Abstract: A fast-tracking phase-locked-loop (PLL) with an analog mixer for phase detection and correction is provided. A frequency lock loop architecture as described herein is used for a PLL that can lock the phase of a local oscillator to an input reference signal of arbitrarily high frequency, even if the local oscillator and the input reference signal frequencies are originally very far apart. To accommodate arbitrarily high frequency input reference signals, embodiments use an analog mixer for the phase detector, rather than a phase-frequency detector (PFD). The analog mixer can be designed to operate on input signals with frequencies of 10's of GHz, whereas the PFD is limited to input frequencies of less than 1 GHz. Embodiments use a new architecture utilizing the concept of Hartley Image Rejection receiver architecture in a frequency lock loop, which enables the PLL to adjust the local oscillator frequency to be brought within the frequency locking range of the analog mixer.

    AUTHENTICATION OF IDENTIFIERS BY LIGHT SCATTERING

    公开(公告)号:US20250005308A1

    公开(公告)日:2025-01-02

    申请号:US18701030

    申请日:2022-10-18

    Abstract: Identifying a test pattern includes positioning a test pattern that includes a first multiplicity of particles that reflect or emit light in the field of view of a first imaging device; illuminating the test pattern with light from a first light source; obtaining, with the first imaging device, a first test image of light reflected by the first multiplicity of particles; and comparing the first test image with a first reference image of a reference pattern obtained by a second imaging device.

Patent Agency Ranking