Low-dimensional manifold constrained disentanglement network for metal artifact reduction in CT images

Commonly implanted medical devices containing metal parts (i.e., dental fillings, coils, hip replacements) generate streaks in computed tomography (CT) images, thereby impeding diagnosis and interfering with radiation therapy planning. Inventors at RPI created a novel technique to boost the efficacy of neural networks for metal artifact reduction (MAR) in CT images. Currently, deep neural network-based techniques need to be trained on synthetic, paired images. Unfortunately, these images may not accurately reflect clinical reality and technical factors.

CASE - Computer Aided System Editor

Researchers at RPI have conceived a technical article publication system. “CASE” or “Computer-Aided System Editor'', may improve the quality of peer review, reduce publication costs, shorten review times, and facilitate research progress. Whereas centralized publication systems rely on authors to pay “open access” publication fees and volunteer peer reviewers, CASE provides a metric to compensate authors and reviewers as part of a crowdsourced, sustainable publishing ecosystem.

General AI Method for Automatic Personalized Training for Standardized K–12 Math Tests

Researchers at RPI are developing a cognitive logic-enabled AI that can operate on multiple screens and in multiple environments for the K-12 sector. Declining public school math and science test scores have concerned American politicians and educators since the 1980s. This educational failure coincides with the large number of employees unable to fill the growing number of technology and engineering-related jobs. Unfortunately, the COVID-19 pandemic has further complicated the K-12 STEM education crisis.


Advances in the semiconductor industry continue to be desired to address demand for semiconductor devices capable of high performance and low power consumption in a wide variety of applications. In one or more applications, enhanced high-voltage semiconductor devices such as, enhanced Schottky diodes, p-i-n diodes, insulated-gate bipolar transistors (IGBT), bipolar junction transistors (BJTs), etc., may be desired for, for instance, high-speed power switching applications.

Probe-based High Precision Spatial Orientation Control and Assembly of Parts for Microassembly using Computer Vision

This technology relates to visually-guided multiprobe microassembly for assembling micro-electromechanical (MEMS) devices from multiple parts that are assembled rather than using bulk-processes to produce devices monolithically. Current production technologies primarily use a single wafer that is process chemically to produce finished devices. While this is useful for many devices, it results in mechanical regions that exist primarily in the plane and do not have fully spatial mechanisms without significant depth of stacked parts.


This technology relates to detecting radiation by directing an optical beam into a volume of gas, ionizing a portion of the volume of gas with the optical beam to produce a plasma, and then detecting an acoustic signal produced from an interaction of a radiation wave with the plasma. This technology relates to the phenomenon of enhanced acoustic emission from laser-induced plasma under the influence of a single-cycle terahertz pulse. Aspects of this technology bridge the unintentional gap between THz photonics and photo-acoustics.


This technology relates to liquid lenses, which are adaptive optical elements that avoid some of the drawbacks of mechanical optical elements, such as delayed movements and excess weight. This technology provides an oscillating liquid lens that includes a liquid drop with first and second droplet portions, a second liquid, and a drive that oscillates the liquid drop within a channel of a substrate.

A Plasma Diagnostic Method using Terahertz-Wave-Enhanced Fluorescence

This technology relates to a plasma diagnostic method by directing THz radiation into plasma and detecting an emission due to the interaction of the THz radiation with the plasma to characterize the plasma. Terahertz (THz) waves occupy a segment of the electromagnetic spectrum between the infrared and microwave bands. As such, they can be used for imaging and sensing in ways not possible with conventional technologies such as X-ray and microwave.

Method of remote broadband terahertz wave sensing utilizing radiation-enhanced-emission-of-fluorescence from gas plasma

This technology relates to detecting terahertz radiation using an optical beam in a volume of gas. Ionizing a portion of the volume of gas with the optical beam produces a plasma and fluorescence is produced from an interaction of a radiation wave with the plasma. Information contained in the characteristics of the detected fluorescence can be used to characterize the radiation wave. Terahertz (THz) waves occupy a segment of the electromagnetic spectrum between the infrared and microwave bands.