System and method for analyzing three-dimensional image data
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11783476-B2
Publication Date
2023-10-10
Expiration Date
Abstract
The present disclosure provides a method for determining a malignancy likelihood score for breast tissue of a patient. The method includes receiving a plurality of two-dimensional images of the breast tissue, the two-dimensional images being derived from a three-dimensional image of the breast tissue, for each two-dimensional image, providing the two-dimensional image to a first model including a first trained neural network, and receiving a number of indicators from the first model, each indicator being associated with a two-dimensional image included in the plurality of two-dimensional images, generating a synthetic two-dimensional image based on the number of indicators and at least one of the plurality of two-dimensional images, providing the synthetic two-dimensional image to a second model including a second trained neural network, receiving a malignancy likelihood score from the second model, and outputting a report including the malignancy likelihood score to at least one of a memory or a display.
Core Innovation
The invention determines a malignancy likelihood score for breast tissue of a patient from multiple two-dimensional images derived from a three-dimensional image of the breast tissue. A plurality of two-dimensional images are received and provided to a first model comprising a first trained neural network trained to identify regions of interest and output indicator data including relevancy scores and arrays of pixels with intensity values.
Based on the first identified regions of interest or the number of indicators, the invention generates a synthetic two-dimensional image by populating a pixel array of the synthetic two-dimensional image. The generating is based on at least the first identified region of interest and the second identified region of interest, including constructing synthetic pixel content from coverage areas associated with indicators, and in some cases pixels not included in the indicators.
The synthetic two-dimensional image is provided to a second model comprising a second trained neural network, and a malignancy likelihood score is received from the second model. The malignancy likelihood score is output in a report to at least one of memory or display, so that the score can be presented for the breast tissue of the patient.
Claims Coverage
The document includes three independent claims. All three independent claims share a common two-stage neural-network pipeline with region-of-interest or indicator extraction from multiple two-dimensional images derived from a three-dimensional image, synthetic two-dimensional image generation, and malignancy likelihood scoring with reporting.
Two-stage neural scoring with synthetic two-dimensional image
Receiving a plurality of two-dimensional images of the breast tissue derived from a three-dimensional image, providing each two-dimensional image to a first model comprising a first trained neural network trained to identify regions of interest, receiving identified regions of interest from the first model, generating a synthetic two-dimensional image by populating a pixel array based on the identified regions of interest, providing the synthetic two-dimensional image to a second model comprising a second trained neural network, receiving a malignancy likelihood score from the second model, and outputting a report including the malignancy likelihood score to at least one of a memory or a display.
Processor-implemented system producing a report from synthetic two-dimensional scoring
A memory configured to store a plurality of two-dimensional images derived from a three-dimensional image of the breast tissue; a processor configured to access the memory and provide each two-dimensional image to a first model comprising a first trained neural network trained to identify regions of interest, receive identified regions of interest corresponding to different images, generate a synthetic two-dimensional image by populating a pixel array based on at least the first and second identified regions of interest, provide the synthetic two-dimensional image to a second model comprising a second trained neural network, determine a malignancy likelihood score using the second model; and a display configured to display a report including the malignancy likelihood score.
Indicator-based synthetic generation using coverage areas and pixel intensities
Receiving a plurality of two-dimensional images derived from a three-dimensional image, providing each two-dimensional image to a first model comprising a first trained neural network and receiving a number of indicators from the first model, generating a synthetic two-dimensional image based on the number of indicators and at least one of the plurality of two-dimensional images, determining coverage areas associated with indicators and determining a pixel intensity value of a pixel included in the synthetic two-dimensional image based on intensity values included in the coverage areas, providing the synthetic two-dimensional image to a second model comprising a second trained neural network, receiving a malignancy likelihood score from the second model, and outputting a report including the malignancy likelihood score to at least one of a memory or a display.
Across the independent claims, the claims cover a method and system that extract regions of interest or indicators from multiple two-dimensional images derived from a three-dimensional breast image, generate a synthetic two-dimensional image by populating a pixel array based on the extracted regions or indicators, and use a second trained neural network to determine and report a malignancy likelihood score.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?