Method and apparatus for tumor purity based on pathological slide image
Inventors
PARK, Ga Hee • OCK, Chan Young • Paeng, Kyung Hyun
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12705733-B2
Publication Date
2026-08-11
Expiration Date
Abstract
Provided is a computing apparatus including: at least one memory; and at least one processor, wherein the at least one processor is configured to: perform a first classification on a plurality of tissues expressed in a pathological slide image by analyzing the pathological slide image, perform a second classification on a plurality of cells expressed in a pathological slide image by analyzing the pathological slide image, and calculate tumor purity including information on noise included in the pathological slide image by combining a first classification result and a second classification result.
Core Innovation
The invention relates to a digital-pathology method and apparatus for predicting tumor purity from pathological whole-slide images by analyzing the pathological slide image to perform a first classification on a plurality of tissues expressed in the pathological slide image and a second classification on a plurality of cells expressed in the pathological slide image. The first classification includes cancer area and other tissue-region information, and the second classification includes tumor cells and additional cell categories. The classifications are combined so that tumor cells are classified among cells included in a cancer area.
The method determines a total number of cells corresponding to the number of tumor cells, the number of lymphocyte cells, and the number of other cells included in the cancer area, a cancer stroma area, and a background area in the pathological slide image. The background area includes an area representing biological noise. Tumor purity is calculated including information about a first ratio of the number of the tumor cells included in the cancer area with respect to the total number of cells.
The tumor purity calculation explicitly incorporates technical noise information calculated based on a size of a degraded area included in the pathological slide image. The disclosed approach further computes indices associated with expected cancer signal and outputs tumor purity and related indices, including an indication based on a preset threshold as to whether additional experiments are likely feasible.
Claims Coverage
The document includes two independent claims, a computing apparatus and a method, and specifies the same core inventive workflow in both: first classify tissue regions, then classify cells, combine results to classify tumor cells in a cancer area, compute cell counts across cancer, stroma, and background areas, and calculate tumor purity with technical-noise information derived from a degraded-area size.
First and second classifications combined for tumor-cell classification in a cancer area
perform a first classification on a plurality of tissues expressed in a pathological slide image by analyzing the pathological slide image; perform a second classification on a plurality of cells expressed in the pathological slide image by analyzing the pathological slide image; classify tumor cells among cells included in a cancer area in the pathological slide image by combining the first classification result and the second classification result
Tumor purity calculation using cell ratios and technical-noise information
calculate tumor purity including information about a first ratio of the number of the tumor cells included in the cancer area with respect to the total number of cells, and information on technical noise included in the pathological slide image
Technical noise calculated based on degraded-area size
wherein the information on the technical noise is calculated based on a size of a degraded area included in the pathological slide image
Cell counting across cancer, cancer stroma, and background areas including biological noise
determine a total number of cells corresponding to a number of the tumor cells, a number of lymphocyte cells, and a number of other cells included in the cancer area, a cancer stroma area, and a background area in the pathological slide image, wherein the background area includes an area representing biological noise
Both independent claims cover interpreting a pathological slide image by combining first and second classifications to classify tumor cells within a cancer area, determining cell counts across cancer, cancer stroma, and background areas, and calculating tumor purity using a tumor-to-total ratio together with technical-noise information derived from a degraded-area size.
Stated Advantages
Enables calculating tumor purity including technical-noise information based on the size of a degraded area.
Documented Applications
Predicting tumor purity from pathological whole-slide images and outputting tumor purity with related indices and decision guidance based on a preset threshold for whether additional experiments are likely feasible.
Interested in licensing this patent?