20240027374. Analysis of X-ray Scatterometry Data using Deep Learning simplified abstract (BRUKER TECHNOLOGIES LTD.)

From WikiPatents
Jump to navigation Jump to search

Analysis of X-ray Scatterometry Data using Deep Learning

Organization Name

BRUKER TECHNOLOGIES LTD.

Inventor(s)

Andrei Baranovskiy of Haifa (IL)

Inbar Grinberg of Ramat Yishay (IL)

Michael G. Greene of Migdal HaEmek (IL)

Matthew Wormington of Highlands Ranch CO (US)

Analysis of X-ray Scatterometry Data using Deep Learning - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240027374 titled 'Analysis of X-ray Scatterometry Data using Deep Learning

Simplified Explanation

The abstract describes a method for training a neural network (NN) using a training dataset. The dataset consists of multiple pairs, where each pair includes a diffraction image obtained from directing an incident x-ray beam at a sample and a label indicating properties of the structures formed in the sample and the incident x-ray beam. The NN is trained to produce predefined outputs by applying it to each pair and adjusting its parameters based on the estimated output and the corresponding predefined output.

  • The method involves training a neural network using a dataset of diffraction images and corresponding labels.
  • The diffraction images are obtained by directing an x-ray beam at a sample and capturing the resulting diffraction pattern.
  • The labels include parameters indicating properties of the structures formed in the sample and the incident x-ray beam.
  • The neural network is trained by applying it to each pair of diffraction image and label.
  • The neural network adjusts its parameters based on the estimated output and the corresponding predefined output.
  • The goal of the training is to obtain the predefined outputs for the given pairs.

Potential applications of this technology:

  • Material characterization: The trained neural network can be used to analyze diffraction images and extract information about the structures formed in a sample, such as crystallographic properties or defects.
  • Quality control in manufacturing: The neural network can be employed to identify and classify different types of structures or defects in manufactured products based on their diffraction patterns.
  • Medical imaging: The method can be applied to analyze diffraction images obtained from biological samples, aiding in the diagnosis and understanding of diseases.

Problems solved by this technology:

  • Automated analysis: The method eliminates the need for manual interpretation of diffraction images, enabling faster and more accurate analysis.
  • Complex pattern recognition: The neural network can handle complex diffraction patterns and extract meaningful information from them, which may be challenging for traditional analysis methods.
  • Standardization: The method provides a standardized approach for training neural networks to analyze diffraction images, ensuring consistent and reliable results.

Benefits of this technology:

  • Efficiency: The automated analysis provided by the neural network reduces the time and effort required for analyzing diffraction images, enabling high-throughput analysis.
  • Accuracy: The trained neural network can provide accurate and reliable results, minimizing human errors and subjective interpretations.
  • Versatility: The method can be applied to various types of diffraction images and samples, making it applicable to a wide range of industries and research fields.


Original Abstract Submitted

a method for training a neural network (nn), the method includes: receiving a training dataset including: (a) multiple pairs of: (i) a diffraction image indicative of x-ray photons diffracted from structures formed in a sample responsively to directing an incident x-ray beam at an angle relative to the sample, and (ii) a label, including: a first parameter indicative of at least a first property of the structures, and a second parameter indicative of at least a second property of the incident x-ray beam, and (b) multiple predefined outputs for the multiple pairs, respectively. the nn is trained to obtain the predefined outputs by: (i) applying the nn to at least a given pair of the pairs, and (ii) responsively to receiving from the nn an estimated output of the given pair, providing the nn with a given predefined output of the predefined outputs that corresponds to the given pair.

Retrieved from "http://wikipatents.org/index.php?title=20240027374._Analysis_of_X-ray_Scatterometry_Data_using_Deep_Learning_simplified_abstract_(BRUKER_TECHNOLOGIES_LTD.)&oldid=29720"