Method and system for analyzing pathology image
Inventors
BRATTOLI, Biagio • OCK, Chan-young • JUNG, Wonkyung • Cho, Soo Ick • Paeng, Kyunghyun • Yoo, Dong Geun
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12266196-B2
Publication Date
2025-04-01
Expiration Date
Abstract
Provided is a method for analysing a pathology image, which is performed by at least one processor and includes acquiring a pathology image, inputting the acquired pathology image into a machine learning model and acquiring an analysis result for the pathology image from the machine learning model, and outputting the acquired analysis result, in which the machine learning model is a model trained by using a training data set generated based on a first pathology data set associated with a first domain and a second pathology data set associated with a second domain different from the first domain.
Core Innovation
The invention relates to analysing a pathology image by using a pathology image analysis model. The model generates analysis results by analysing a staining intensity at a first location and at a second location where a staining color is expressed on at least one cell, where the first location and the second location include at least one of cell membrane, cell nucleus, or cytoplasm. Based on the staining intensity amounts at the first location on the first cell and the first location on the second cell, the model classifies a first cell and a second cell in the pathology image.
The invention further segments the pathology image into a first segment corresponding to the first cell and a second segment corresponding to the second cell, and visualizes the first segment in a first color and the second segment in a second color different from the first color. An analysis result is output that includes at least one of the first analysis result or the second analysis result.
The invention includes an information processing system context in which at least one processor, executing instructions stored in memory, acquires the pathology image and generates the analysis result using the pathology image analysis model. The model can determine analysis outputs based on features of the pathology image and can be configured to receive user-provided feature input for output generation. The output can also include compartment-specific distribution information for staining color expression values for at least one of the cell nucleus, cytoplasm, or cell membrane.
Claims Coverage
Independent claims are directed to a method and a corresponding information processing system for analysing pathology images using a pathology image analysis model that analyses staining intensity at multiple subcellular locations, classifies cells, segments them, and visualizes segments in different colors, with locations including at least one of cell membrane, cell nucleus, or cytoplasm. The inventive features are further refined by dependent claims to include model selection, user input conditioning, and additional staining readouts such as positive/negative status, quantified expression values, and distribution statistics.
Analyzing staining intensity at two locations and classifying cells
Generating a first analysis result by analysing a staining intensity at a first location where a staining color is expressed on at least one cell; generating a second analysis result by analysing a staining intensity at a second location where the staining color is expressed on the at least one cell; classifying a first cell and a second cell in the pathology image based on staining intensity amounts at the first location on the first cell and the first location on the second cell in the pathology image; wherein the first location and the second location include at least one of cell membrane, cell nucleus, or cytoplasm.
Segmenting and visualizing cell-specific segments in different colors
Segmenting the pathology image into a first segment corresponding to the first cell and a second segment corresponding to the second cell; visualizing the first segment in a first color and the second segment in a second color different from the first color; and outputting the analysis result including at least one of the first analysis result or the second analysis result.
System executing instructions for model-based pathology image analysis
An information processing system comprising at least one memory storing instructions; and at least one processor operatively connected to the at least one memory and configured to execute the instructions to acquire a pathology image; generate an analysis result using a pathology image analysis model configured to generate first and second analysis results by analysing staining intensity at first and second locations; classify a first cell and a second cell based on staining intensity amounts; segment the pathology image into first and second segments corresponding to the first and second cells; visualize the first segment in a first color and the second segment in a second color different from the first color; and output the analysis result including at least one of the first analysis result or the second analysis result; wherein the first location and the second location include at least one of cell membrane, cell nucleus, or cytoplasm.
Across the independent claims, the core coverage is the combination of analysing staining intensity at a first and second location that correspond to subcellular compartments, classifying cells based on staining intensity amounts, segmenting the image into cell-corresponding segments, and visualizing those segments in different colors, with the outputs including at least one of the first or second analysis results.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?