Method and device for evaluating quality of pathological slide image
Inventors
PARK, Ga Hee • Paeng, Kyung Hyun • OCK, Chan Young • Song, Sang Hoon • Kim, Suk Jun
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12014502-B2
Publication Date
2024-06-18
Expiration Date
Abstract
A computing device includes at least one memory, and at least one processor configured to analyze at least one object expressed in a pathological slide image, evaluate quality of the pathological slide image based on a result of the analyzing, and perform at least one additional operation according to a result of the evaluating.
Core Innovation
A machine learning model detects a plurality of tissue regions and a cancer region from a pathological slide image. Quantitative information is generated about the detected tissue regions and the cancer region, including an area corresponding to entire tissue regions and an area corresponding to the cancer region, and the quantitative information is compared with a predetermined reference value to support quality evaluation of the pathological slide image.
A quality evaluation test is determined based on the comparison between the quantitative information and the predetermined reference value. A result of the quality evaluation test is used in determining whether to perform downstream analysis on the pathological slide image, and the result indicating whether the pathological slide image has passed the quality evaluation test is displayed on a display device.
The pathological slide image is determined to have failed the quality evaluation test if the area corresponding to entire tissue regions is smaller than a first reference value or the area corresponding to the cancer region is smaller than a second reference value. The framework also includes refining quality evaluation with biomarker-related quantitative information based on tumor cells, lymphocyte cells, and tumor purity, with region classification including a cancer stroma region, a necrosis region, and a background region.
Claims Coverage
The document contains two independent claims, covering a two-part workflow: detecting tissue and cancer regions using a machine learning model and using area-based quantitative comparisons to perform a quality evaluation test whose pass/fail result gates downstream analysis, while also displaying the quality evaluation result.
Machine learning detection of tissue and cancer regions
Use a machine learning model to detect a plurality of tissue regions and cancer region from a pathological slide image.
Generation of quantitative region area information
Generate quantitative information about the detected tissue regions and cancer region, wherein the quantitative information comprises an area corresponding to entire tissue regions and an area corresponding to the cancer region.
Area-based quality evaluation test with predetermined reference values
Compare the quantitative information with a predetermined reference value and determine whether the pathological slide image has passed a quality evaluation test based on the comparison.
Pass/fail quality evaluation gates downstream analysis
Use a result of the quality evaluation test in determining whether to perform downstream analysis on the pathological slide image.
Display quality evaluation pass/fail result
Display on a display device the result of the quality evaluation test indicating whether the pathological slide image has passed the quality evaluation test.
Failure determination by entire-tissue area and cancer-region area thresholds
Determine that the pathological slide image has failed the quality evaluation test if the area corresponding to entire tissue regions is smaller than a first reference value or the area corresponding to the cancer region is smaller than a second reference value.
Across both independent claims, the inventive core is machine learning detection of tissue and cancer regions from a pathological slide image, area-based quantitative information generation, comparison to predetermined reference values, and a quality evaluation test whose pass/fail outcome controls downstream analysis while providing a displayed pass/fail result.
Stated Advantages
Allows determining whether a pathological slide image has passed a quality evaluation test.
Uses the quality evaluation test result to determine whether to perform downstream analysis on the pathological slide image.
Displays the result of the quality evaluation test indicating whether the pathological slide image has passed the quality evaluation test.
Determines failure based on area corresponding to entire tissue regions and/or cancer region relative to reference values.
Documented Applications
Quality evaluation of pathological slide images to control whether downstream analysis is performed, with a displayed pass/fail QC result and a report basis.
Interested in licensing this patent?