Medical University of South Carolina

General Electric Company

Nbcuniversal Media, LLC

Yan Tong

Peter Henry Tu

Chen Li

Frederick Wilson Wheeler

Xiaoming Liu

Jilin Tu

Xianming Dai

Fanghao Yang

Li Guan

Yi Yao

Ming-Ching Chang

Fanhao Yang

Microfluidic devices, along with methods of their fabrication, are provided. The microfluidic device can include a substrate defining a microchannel formed between a pair of side walls and a bottom surface and a plurality of nanotips positioned within the microchannel and proximate to each side wall such that a boundary layer is formed along each side wall between the plurality of nanotips and the side wall upon addition of a liquid into the microchannel.

Ultra-efficient two-phase evaporators/boilers enabled by nanotip-induced boundary layers

Microfluidic devices having superhydrophilic bi-porous interfaces are provided, along with their methods of formation. The device can include a substrate defining a microchannel formed between a pair of side walls and a bottom surface and a plurality of nanowires extending from each of the side walls and the bottom surface. For example, the nanowires can be silicon nanowires (e.g., pure silicon, silicon oxide, silicon carbide, etc., or mixtures thereof).

Microfluidic devices for the generation of nano-vapor bubbles and their methods of manufacture and use

Microfluidic devices, along with their methods of formation and use, are provided. The microfluidic device can include a substrate with a main channel and a first auxiliary channel defined in the substrate's surface. The main channel has a main width of about 1000 &#x3bc;m or less. The first auxiliary channel intersects with the main channel at a first aperture defined in a first side wall of the main channel. A second auxiliary channel can intersect with the main channel at a second aperture defined in a second side wall of the main channel. A plurality of main channels and respective auxiliary channel(s) can be included on the surface.

Enhanced flow boiling in microchannels by high frequency microbubble-excited and -modulated oscillations

An apparatus and method, as may be used for predicting solar irradiance variation, are provided. The apparatus may include a solar irradiance predictor processor () configured to process a sequence of images (e.g., sky images). The irradiance predictor processor may include a cloud classifier module () configured to classify respective pixels of an image of a cloud to indicate a solar irradiance-passing characteristic of at least a portion of the cloud. A cloud motion predictor () may be configured to predict motion of the cloud over a time horizon. An event predictor () may be configured to predict over the time horizon occurrence of a solar obscuration event. The prediction of the solar obscuration event may be based on the predicted motion of the cloud. The event predictor may include an irradiance variation prediction for the obscuration event based on the solar irradiance-passing characteristic of the cloud.

Apparatus and method for predicting solar irradiance variation

A system and method for estimating a set of landmarks for a large image ensemble employs only a small number of manually labeled images from the ensemble and avoids labor-intensive and error-prone object detection, tracking and alignment learning task limitations associated with manual image labeling techniques. A semi-supervised least squares congealing approach is employed to minimize an objective function defined on both labeled and unlabeled images. A shape model is learned on-line to constrain the landmark configuration. A partitioning strategy allows coarse-to-fine landmark estimation.

System and method for automatic landmark labeling with minimal supervision

A technique for optimizing object recognition is disclosed. The technique includes receiving at least one image of an object and at least one reference image. The technique further includes identifying at least one performance metric corresponding to an object recognition task. The identified performance metric is optimized to generate the corresponding optimized performance metric by determining an optimal subspace based on a determined objective function corresponding to the object recognition task and a difference between the received image and the corresponding reference image. Subsequently, the technique includes comparing the received image with the reference image based on the optimized performance metric for performing the object recognition task.

Growing community of inventors

Chapin, SC, United States of America

Yan Tong