This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: University of California. Active years: 2025.
Company Filing History:
Years Active: 2025
Title: Innovations by Artem Goncharov in Neural-Network Based Spectroscopy
Introduction
Artem Goncharov is an innovative inventor based in Los Angeles, CA. He has made significant contributions to the field of spectral analysis through his groundbreaking patent. His work focuses on utilizing deep learning techniques to enhance the capabilities of spectroscopy.
Latest Patents
Artem Goncharov holds a patent for a "Device and method for neural-network based on-chip spectroscopy using a plasmonic encoder." This invention describes a deep learning-based spectral analysis device that employs a spectral encoder chip containing a plurality of nanohole array tiles. Each tile has a unique geometry, resulting in a distinct optical transmission spectrum. The device captures transmitted light using a CMOS image sensor, eliminating the need for lenses, gratings, or other optical components. The spectral reconstruction neural network processes the transmitted intensities to accurately reconstruct the input spectrum. In one embodiment, the system utilized a spectral encoder chip with 252 nanohole array tiles and was trained on 50,352 spectra. It achieved an impressive identification rate of 96.86% of spectral peaks, with minimal localization and height errors.
Career Highlights
Artem Goncharov is affiliated with the University of California, where he continues to advance his research in the field of spectroscopy. His innovative approach combines neural networks with traditional spectroscopic methods, paving the way for more efficient and accurate spectral analysis.
Collaborations
Artem has collaborated with notable colleagues, including Zachary S. Ballard and Aydogan Ozcan. Their combined expertise contributes to the advancement of research in optical technologies and deep learning applications.
Conclusion
Artem Goncharov's contributions to neural-network based spectroscopy represent a significant advancement in the field. His innovative patent showcases the potential of integrating deep learning with optical technologies, promising to enhance spectral analysis in various applications.