Method for efficiently determining the density and spatial relationships of multiple cell types in a tissue region - Patents.com

Abstract

To provide efficient methods for identifying biomarkers.SOLUTION: A method can include a step of identifying a tumor area. The method may further include a step of identifying a plurality of regions. The method may also include a step of defining, for each region, a bounding area for the region that encompasses the region. The method may include a step of determining, for each region of a first subset of the plurality of regions, that the region is to be ascribed to the tumor, where the bounding area is fully within the tumor area. The method may further include a step of determining, for each region of a second subset of the plurality of regions, whether or not to ascribe the region to the tumor based on an intersection of the region and the tumor area. The method may also include a step of accessing a metric characterizing a biological observation and a step of generating a result based on the metrics. The result may be used as a biomarker.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of priority to U.S. Application No. 63 / 016,004, filed April 27, 2020, the entire contents of which are incorporated herein by reference for all purposes. [Background technology]

[0002] Immunohistochemistry (IHC) slide image analysis can include computational analysis. For example, multiplex immunofluorescence (MIF) staining of tissue sections allows for simultaneous detection of multiple biomarkers and their co-expression at the single cell level. MIF allows characterization of the immune situation in the tumor microenvironment, which greatly influences the response to immunotherapy. Primary analysis can detect and classify all different kinds of markers / cell phenotypes. After the initial analysis, secondary analysis is a crucial step to generate various features such as density or spatial relationships to correlate clinical studies or trials and predict prognosis or drug action. Methods and systems that efficiently generate useful results of secondary analysis are desired. These and other improvements are addressed by the embodiments described herein. Summary of the Invention

[0003] An embodiment of the present invention allows for efficient determination of density and spatial relationships for multiple cell types in a tissue region. Efficiency is possible in one or more aspects. The same high magnification used for the image is used to analyze features (e.g., cells, phenotypes, epitumors) within the region of interest (e.g., tumor), avoiding calculations for adjustments to different magnifications. Feature shapes can be complex and describable by a large set of pixel coordinates (e.g., corresponding to each point along the perimeter of the feature or a point located at the center of the feature). Thus, performing a spatial evaluation using a set of pixel coordinates (e.g., to determine whether a feature is within a given region of interest, such as a tumor area) can require a large number of calculations. To improve computational efficiency, a bounding region with a simpler shape (e.g., a box) can be defined as an area surrounding the feature. The bounding area speeds up the analysis of whether a feature is within the tumor or another particular region of interest. For example, a feature can be determined to be sufficiently within the region of interest if all four vertices of the bounding box are within the region. A hash table can be used to identify relationships between polygons representing features within the region of interest, tissue, and tumor. Additionally, parameters for secondary analysis are identified as biomarkers that can be useful in characterizing the tumor or other region of interest.

[0004] An embodiment may include a method of identifying a tumor biomarker using an image of a biological sample. The method may include identifying a tumor area in the image. The tumor area may delineate the border of the tumor. The method may further include identifying a plurality of regions of the image. Each region of the plurality of regions corresponds to a tissue block or biological object (e.g., tumor periphery, stroma, cell phenotype). Additionally, the method may include defining, for each region of the plurality of regions, a bounding area (e.g., a bounding box or bounding ellipse) surrounding the region, the bounding area comprising a polygon or an ellipse. The method may include determining, for each region of a first subset of the plurality of regions whose bounding area is entirely within the tumor area, that the region is attributable to the tumor. Additionally, the method may further include determining, for each region of a second subset of the plurality of regions whose bounding area is partially within the tumor area. The method may further include determining, for each region of a second subset of the plurality of regions, whether the region is attributable to the tumor based on an intersection of the region with the tumor area. The method can include accessing, for each region attributable to the tumor, a metric that characterizes the biological observation. The method can include generating a result value based on the accessed metric. The method can further include comparing the result value to a reference value determined by using a region attributable to another tumor or by using a region attributable entirely outside the tumor area. Additionally, the method can include determining whether the result is a biomarker based on the comparison.

[0005] An embodiment may include a method of analyzing an image of a biological sample. The method may include determining a first count or a first density of a cellular phenotype in a first region of interest (e.g., a tumor). The method may further include determining a second count or a second density of a cellular phenotype in a second region of interest (e.g., a stroma). The second region of interest may be within the first region of interest. The method may further include generating an output. The output may include the first count or the first density, as well as the second count or the second density.

[0006] An embodiment may include a method of analyzing an image of a biological sample. The method may include identifying a plurality of first cells within a tumor. Each first cell of the plurality of first cells may have a first phenotype. The method may further include identifying a plurality of second cells within the tumor. Each second cell of the plurality of second cells may have a second phenotype. The method may include calculating, for each first cell of the plurality of first cells, a shortest distance to a second cell of the plurality of second cells using a nearest neighbor search. Additionally, the method may include generating a result based on the calculated shortest distance.

[0007] An embodiment may include a method of analyzing an image of a biological sample. The method may include identifying a plurality of first cells within a tumor. Each first cell of the plurality of first cells may have a first phenotype. The method may further include identifying one or more regions within the tumor, each region of the one or more regions corresponding to a tissue block or biological object (e.g., tumor periphery, stroma, etc.). The method may further include, for each first cell of the plurality of first cells, calculating a shortest distance between the first cell and each region of the one or more regions using a nearest neighbor search. Additionally, the method may include generating a result based on the calculated shortest distance.

[0008] An embodiment may include a method of analyzing an image of a biological sample. The method may include identifying a plurality of regions within a tumor, each region of the plurality of regions corresponding to a tissue block or biological object (e.g., peritumoral, stroma, etc.). The method may further include determining a compression level for each region of the plurality of regions. The method may further include generating a result based on the compression level.

[0009] The nature and advantages of embodiments of the present invention may be better understood with reference to the following detailed description and accompanying drawings. [Brief description of the drawings]

[0010] [Figure 1A] 1 shows analysis of MIF images of tissue. [Figure 1B] 1 shows analysis of MIF images of tissue.

[0011] [Figure 1C] 13 shows a comparison of computation time between using a high magnification and a low magnification in accordance with an embodiment of the present invention.

[0012] [Diagram 2] 1 illustrates a method for analyzing an image of a biological sample according to an embodiment of the present invention.

[0013] [Diagram 3] 1 shows a flow chart for analyzing an image of a biological sample according to an embodiment of the present invention.

[0014] [Figure 4] 4 illustrates an analyzed area of ​​an image according to an embodiment of the present invention.

[0015] [Figure 5A] 1 shows an exemplary image of tumor tissue according to an embodiment of the present invention.

[0016] [Figure 5B] 1 illustrates graphical results of an analysis of cell types within tumor tissue according to an embodiment of the present invention.

[0017] [Figure 6] 1 illustrates graphical results of an analysis of cell types within tumor tissue according to an embodiment of the present invention.

[0018] [Figure 7] 1 illustrates a method for analyzing images of a biological sample for density or count of cell phenotypes according to an embodiment of the present invention.

[0019] [Figure 8A] 1 shows single cell phenotype counting or density analysis according to an embodiment of the present invention.

[0020] [Figure 8B] 1 shows the results of cell phenotyping counting according to an embodiment of the present invention.

[0021] [Figure 9] 1 illustrates a method for analyzing images of a biological sample for phenotypic distance according to an embodiment of the present invention.

[0022] [Figure 10] 4 illustrates distance calculations according to an embodiment of the present invention.

[0023] [Figure 11] 1 illustrates multiple phenotypes within a tumor according to an embodiment of the present invention.

[0024] [Figure 12] 1 illustrates the distances between different phenotypes according to an embodiment of the present invention.

[0025] [Figure 13] 1 shows a table of results regarding distance between phenotypes according to an embodiment of the present invention.

[0026] [Figure 14] 1 shows epitumor and cell contours according to an embodiment of the present invention.

[0027] [Figure 15A] 1 shows an image of a biological sample according to an embodiment of the present invention.

[0028] [Figure 15B] 1 shows a magnified image with a red circle annotating the tumor area according to an embodiment of the present invention.

[0029] [Figure 15C] 13 shows an image with blue polygons representing epitumors and black circles representing phenotypic cells according to an embodiment of the present invention.

[0030] [Figure 16] 1 illustrates the selection of a single epitumor polygon within a tumor according to an embodiment of the present invention.

[0031] [Figure 17] 1 shows a single epitumor surrounded by five contours according to an embodiment of the present invention.

[0032] [Figure 18] 1 shows a table containing cell to epitumor distance results according to an embodiment of the present invention.

[0033] [Figure 19] 1 illustrates a method for analyzing an image of a biological sample for cell-to-region distance according to an embodiment of the present invention.

[0034] [Figure 20] 1 illustrates a method for analyzing images of a biological sample for compression level according to an embodiment of the present invention.

[0035] [Figure 21] 1 illustrates a tumor having regions with different levels of compression in accordance with an embodiment of the present invention.

[0036] [Figure 22] 1 illustrates different compression levels highlighted by polygons according to an embodiment of the present invention.

[0037] [Figure 23] 1 shows a list of results for compression levels according to an embodiment of the present invention.

[0038] [Figure 24] 1 illustrates a computer system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039] Certain features of images of tissue sections, such as density and spatial relationships, can provide useful and important clinical information that can be used to understand the clinical condition of a subject, such as for diagnosis and treatment monitoring.

[0040] Low-magnification mask-based secondary analysis can be used to perform secondary analysis of the image to determine features such as density and spatial relationships. However, the use of low-magnification masks can be computationally intensive.

[0041] Figures 1A and 1B show the analysis of the MIF image of the tissue. Both figures show the magnified section indicated in the blue rectangular area. A low magnification mask using the calculated polygon is generated.

[0042] FIG. 1C shows a table comparing the computation time using the high magnification polygon approach method with the low magnification mask-based method. The computation time in the high magnification polygon-based method is much faster compared to the low magnification mask-based approach according to an embodiment of the present invention. Two downsampling factors are implemented using the low magnification mask-based method, and their performance is shown in FIG. 1C. One has a low magnification that downsamples the original MIF image by a factor of 4. The other has a low magnification downsampling by a factor of 27. For example, in a particular panel, the computation time using the high magnification polygon-based approach is completed in only about 3 minutes, while the slowest low magnification mask approach with a downsampling of 4 is completed in 4.32 hours (260 minutes), which is even longer than 88 times. Moreover, the high magnification polygon-based analysis was generated accurately. I. Boundary Area

[0043] Accurate and computationally efficient results can be achieved in embodiments of the present invention.The use of bounding areas, such as boxes to enclose polygons, aids in the velocity analysis. A. How to Identify Biomarkers

[0044] FIG. 2 illustrates a method 200 for analyzing images of a biological sample to identify tumor biomarkers. In some embodiments, the method 200 can include capturing an image of the biological sample. The images can be from a single slide or from various immunohistochemistry (IHC) slide images, which can include two or more adjacent slides, each with a different biomarker or cell phenotype. The images can be multiplexed immunofluorescence (MIF) stained images of tissue sections. The biological sample can be a single tissue. In some cases, multiple biological samples can be imaged. For example, images from multiple biopsies, multiple tissue blocks, or tissue microarrays can be used for analysis.

[0045] At block 202, a tumor area may be identified in the image. The tumor area may delineate the borders of the tumor. In some embodiments, an area corresponding to a biological structure other than the tumor may be used. The other biological structures may include the tumor, stroma, epitumor, or tumor periphery. The tumor area may be identified manually or by a computer program.

[0046] In block 204, a plurality of regions of the image may be identified. Each region of the plurality of regions may correspond to a tissue block or biological object. The tissue block or biological object may be stroma, epitumor, peritumor, vasculature (e.g., blood), cell phenotype, tumor, necrosis, tissue fold, staining artifact, or red blood cells. The tissue block or biological object may be identified manually or by a computer program. The plurality of regions may cover the entire tumor area and / or the entire image.

[0047] At block 206, a bounding area of ​​the region may be defined for each region of the plurality of regions. The bounding area may encompass the region. In some embodiments, the bounding area may surround the region such that at least one point of the region touches an edge of the bounding area. The bounding area may have an area within a range of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100% greater than the region. The bounding area may include a polygon or an ellipse. For example, the bounding area may be a bounding box or a bounding circle. The bounding area may include a triangle, a rectangle, a pentagon, a hexagon, a heptagon, an octagon, a nonagon, a decagon, or a polygon with 10 or more sides. In some embodiments, the polygon may be a regular polygon.

[0048] At block 208, each region of a first subset of the plurality of regions may be determined to be attributable to the tumor if the boundary area of ​​each region of the first subset is entirely within the tumor area. A boundary area may be determined to be entirely within the tumor area if a subset of points of a polygon or ellipse are determined to be entirely within the tumor area. For example, points representing extreme values ​​along two axes in a 2D coordinate system may be compared to the tumor area. As an example, the uppermost, lowermost, leftmost, and / or rightmost points may be determined to be within the tumor area. In some embodiments, vertices of a polygon may be determined to be within the tumor area.

[0049] In some embodiments, a first subset of the plurality of regions may be determined to be outside the exclusion region. The exclusion region may include biological structures (e.g., tumor, peritumor, stroma, necrosis, red blood cells, staining artifacts, or tissue folds). Determining a region to be outside the exclusion region may be similar to determining a region to be within the tumor area. A border area may be determined to be outside the exclusion region if a vertex or extreme point is determined to be outside the exclusion region.

[0050] At block 210, a boundary area of ​​each region of the second subset of the plurality of regions may be determined to be partially within the tumor area. A boundary area may be determined to be partially within the tumor area if at least one of the points representing extreme values, but not all of the points representing extreme values, are each determined to be outside the tumor area. In some embodiments, one or more vertices (but not all vertices) of the boundary polygon may be determined to be outside the tumor area.

[0051] In some embodiments, the boundary area of ​​each region of the third subset of the plurality of regions can be determined to be entirely outside the tumor area. A boundary area can be determined to be entirely outside the tumor area if all points or all vertices representing extrema of a boundary polygon are determined to be outside the tumor area.

[0052] At block 212, for each region of the second subset of the plurality of regions, it may be determined whether the region is attributable to the tumor. The determination is based on the intersection / exclusion of the region with the tumor area. In some embodiments, the intersection is calculated as the area of ​​the region that overlaps the tumor area. The new sub-region that represents only the portion of the region within the tumor area may be used for the analysis.

[0053] In other embodiments, the intersection can be characterized by the area or percentage of the region within the tumor. The intersection can be compared to a threshold (e.g., a minimum area or percentage). If the intersection exceeds the threshold, the region is attributed to the tumor. If the intersection does not exceed the threshold, the region is not attributed to the tumor. The intersection can be calculated using the geometry of the region and the tumor area.

[0054] In some embodiments, a portion of the area can be attributed to the tumor based on the percentage of intersections. For example, the same percentage can be attributed to the tumor as the percentage of intersections. Any metric determined from the area can be scaled by the percentage of intersections.

[0055] Two hash tables can be used to keep track of the relationship between polygons / regions and tumor / tissue areas. The tumor hash table can be updated with information about whether a region / polygon is attributed to a tumor or not. Furthermore, the tumor hash table can also keep track of information about whether a region is partially / fully inside or outside the tumor area and can also update the region / polygon to which the tumor area (tumor number index) belongs. Similarly, the tissue hash table can be updated with information about whether a region / polygon is attributed to a tissue or not. Furthermore, the tumor hash table can also keep track of information about whether a region is partially / fully inside or outside the tissue area and the region to which the tissue area (tissue number index) belongs.

[0056] At block 214, a metric characterizing the biological observations can be accessed for each region attributable to a tumor. The metric can be any metric disclosed herein. For example, the metric can be a count or density of a particular type of cell or phenotype within the region. The count or density of a particular type of cell can be a count or density of a region within the tumor area or a region completely outside the tumor. Characterizing the metric can include referencing a hash table to determine whether the region is attributable to a tumor / tissue.

[0057] As an example, the metric can include locations of the different types of regions. The plurality of regions can include a first region corresponding to a first tissue block or a first biological object. The plurality of regions can further include a second region corresponding to a second tissue block or a second biological object. Each of the first and second regions is determined to be attributable to a tumor. The metric characterizing the biological observation can identify a location associated with the region.

[0058] In some embodiments, the first tissue block or first biological object may be a cell of a first phenotype, which may be a co-localization of a single biomarker or different biomarkers. The second tissue block or second biological object may be a cell of a second phenotype. The metric may include the location of the cell / phenotype.

[0059] In some embodiments, the first region can correspond to cells of a first phenotype and the second region can correspond to a second tissue block. Metrics can include the location of the cells as well as the location of the tissue or other region.

[0060] The metrics can include various features of the region or phenotype, such as absolute intensity values, intensity percentages and distributions, shapes, areas, compression levels, etc. Each region of the plurality of regions can correspond to a tissue block or biological object. The compression level can be of the region corresponding to the tissue block or biological object. The compression level is described as an example feature later in this specification.

[0061] At block 216, a result value can be generated based on the accessed metrics. The result can be generated for the region attributed to the tumor. For example, if the metric includes the location of the region or cells, the result can identify the distance between a first location identified by the metric in a particular region (e.g., a first region) and a second location identified by the metric in another region (e.g., a second region) or cell. The result can include each of the accessed metrics. In some embodiments, the result can include a statistical value of the accessed metrics. For example, the result can include the mean, median, mode, percentile, or standard deviation of the accessed metrics. The result can include a statistical value when one or more of the accessed metrics exceed a threshold.

[0062] In some embodiments, a count or density of a particular cell in a particular tissue block or biological object can be used as a metric. A subset of the plurality of regions can correspond to the tissue block or biological object. The method can include accessing a metric that characterizes a biological observation for each region of the subset of the plurality of regions. Producing a result value can include determining a value of a function that includes the metric for each region attributed to the tumor and each region attributed to the subset of the plurality of regions. For example, the result can be a ratio of a count or density of a particular type of cell in the tissue block to a count or density of a particular type of cell in the tumor area. The result can be any of the outputs described in Figures 7, 8A, and 8B.

[0063] In some embodiments, the distance between the cell types can be used to generate a result value. The plurality of regions can include a first region having a plurality of first cells having a first phenotype. The plurality of regions can include a second region having a plurality of second cells having a second phenotype. The first region and the second region can be determined to be tumor-attributable. A metric characterizing the biological observation can identify a location of each cell of the plurality of cells in the region. The result can be generated by calculating, for each first cell of the plurality of first cells, a shortest distance to a second cell of the plurality of second cells using a nearest neighbor search. The nearest neighbor search can be a K-nearest neighbor (KNN) search. A statistic of the shortest distance can be calculated. The result can include a statistic. In some embodiments, a count of first cells having a shortest distance from a second cell having a range can be determined. The result can be a count. The range can be pre-determined prior to identifying the tumor area in the image. The result can be a result described in FIGS. 9-13.

[0064] In some embodiments, the distance between the cell and the tissue block or biological object can be used to generate a result value. The plurality of regions can include a first region having a plurality of first cells having a first phenotype. The plurality of regions can include one or more second regions. Each second region can correspond to a tissue block or biological object. Each of the first region and the one or more second regions can be determined to be attributable to a tumor. The metric characterizing the biological observation can identify at least one location of each region. The at least one location can be a cell location of each of the plurality of first cells in the first region. The at least one location can be a location of a second region of each of the one or more second regions. The location of each second region can be a center of the region or a point on an edge of the second region. The point on the edge of the second region can be a point closest to the first cell. The result value can be generated by calculating, for each cell of the plurality of first cells, a shortest distance to the second region of the one or more regions using a nearest neighbor search. The nearest neighbor search can be a K-nearest neighbor search. The result may include calculating a statistic of the shortest distance, and the result may be a statistic. In some embodiments, the method may determine a count of first cells having a shortest distance within a range. The range may be predetermined prior to identifying the tumor area in the image. The result may be a count. The result may be a result as described in Figures 15A-19.

[0065] The distance between the first cell and the second region may be the closest vertical distance or the closest horizontal distance. The KNN search may determine a first cell among a plurality of first cells that has the closest vertical distance to a given second region. The method may include calculating the distance between the cell with the closest vertical distance and the given second region. The cell with the closest horizontal distance may also be used.

[0066] Results can be reported for each image or each biological sample. In other embodiments, results can be aggregated for multiple images or multiple biological samples. Aggregation can be reporting the results together or providing statistics of a metric for multiple images or biological samples. Any of the results or metrics described herein can be used as described above.

[0067] In block 218, the resulting value can be compared to a reference value determined using an area attributed to another tumor, or determined using an area attributed entirely outside the tumor area.

[0068] A reference value can be determined using a region completely outside the tumor area. For example, for each region of the third subset of the plurality of regions, a boundary area can be determined to be completely outside the tumor area. A metric characterizing a biological observation can be accessed for each region that is completely outside the tumor. A reference value can be generated based on the accessed metric for each region that is completely outside the tumor in the same manner as the result value is generated.

[0069] One or more tumors in one or more reference subjects can be used to determine the reference value. The one or more reference subjects may be known to have a tumor or may be known to have a particular level of cancer.

[0070] At block 220, the result can be determined to be a biomarker for the tumor based on the comparison. If the reference value is obtained from a non-tumor area, the metric can be determined to be a biomarker when the result value is statistically different from the reference value. The non-tumor area can be an area from the same subject or from a reference subject. The reference subject can be known to be cancer-free. In embodiments where the reference value is determined from an area attributed to a tumor, the result can be determined to be a biomarker if the result is not statistically different from the reference value. B. How Tumor Classification is Determined

[0071] The method may include determining a classification of the tumor. The method may include determining a value of a biomarker in an image having a tumor area. The biomarker may be determined by method 200. The value of the biomarker may be compared to a threshold. The threshold may be determined from images of reference subjects having the same tumor classification. The classification of the tumor may be based on the comparison. For example, if the value of the biomarker exceeds the threshold, the tumor may be classified into a more severe category (e.g., malignant or spreading). In some embodiments, the tumor classification may be the presence or severity of cancer. A more severe tumor classification may result in a classification that cancer is present or that the cancer is severe. Examples of cancer classifications include progressive disease and stable disease.

[0072] A classification of the tumor corresponding to the tumor area can be determined using the accessed metrics and / or results. The classification can be any number or other letter associated with a particular characteristic of the tumor area or other area of ​​interest. For example, a "+" sign (or the word "positive") can indicate that the tumor is malignant or growing. The classification can be binary (e.g., positive or negative) or have more classification levels (e.g., a scale of 1-10 or 0-1). The classification can be that the tumor is benign, stable, or shrinking.

[0073] The subject (from whom the biological sample was obtained) can be treated based on the determined classification. For example, a tumor with a classification of malignant or expanding can initiate or increase tumor treatment (e.g., drugs, radiation, therapy, or surgery). A tumor with a classification of benign or shrinking can terminate, decrease, or not initiate tumor treatment. Treatment response classifications can include complete response, partial response, and no response. For these classifications, biomarker values ​​can be obtained before treatment and at one or more time points after treatment. A change (increase or decrease) in the value of the biomarker can indicate a response to treatment.

[0074] Further actions can be determined based on one or more biomarkers, and optionally other clinical information. Further actions can include starting or terminating treatment of the tumor, enrolling or disenrolling the subject with the tumor in a clinical trial, or performing additional diagnostic tests on the subject. The value of the biomarker can be compared to a threshold. The threshold can be determined from a reference subject. The reference subject can include a known healthy individual. In some embodiments, the reference subject can include an individual with a tumor of known stage. The threshold can be a number to establish a statistical difference from the reference subject. For example, the threshold can be one, two, or three standard deviations away from the result value determined from a healthy reference subject. C. Example Bounding Boxes

[0075] 3 illustrates an example of an implementation of method 200. In block 302, manual tumor / exclusion annotations are loaded. The tumor annotations may be the tumor area identified in block 202. The exclusion annotations are for excluding certain regions associated with other biological structures. These excluded regions may include tissue folds, artifacts, metaplasia, necrosis, etc. These excluded regions are regions that may not be used in the analysis.

[0076] In block 304, overlapping tumor and exclusion regions are calculated. A pathologist / user can manually generate / draw tumor areas or exclusion regions that may include overlapping tumor areas or overlapping exclusion regions. Before calculating any further steps, a sanity check can be performed to avoid errors in overlapping tumor or exclusion regions.

[0077] In block 306, the automatically detected polygons are loaded. These polygons may represent epitumors or blood vessels (e.g., blood vessels in tissue that provide nutrition to tumors or immune cells). The polygons may be multiple regions in method 200.

[0078] In block 308, two hash tables are constructed to track the spatial relationship and information between the polygons and the tumor or tissue. The tissue is the largest area and has the tumor and the polygon as a subset of the tissue block.

[0079] A bounding box may be generated at block 310. The bounding box may be the bounding area at block 206 of method 200. Polygons may be identified as being entirely inside the tumor or as intersecting the tumor using the bounding box, similar to blocks 208 and 210 of method 200.

[0080] Polygon intersections and exclusions can be calculated in block 312. The polygons themselves can be used instead of bounding boxes to determine the intersection of the polygons with the tumor or exclusion regions. A hash table can be updated with the identification of polygons that are completely inside the tumor, intersect with the tumor, or are outside the tumor.

[0081] In blocks 314, 316, 318, and 320, different results can be reported including phenotype counts / densities within the region of interest (e.g., tumor), distances between different phenotypes, polygon counts / densities, and distances between phenotypes and polygons. These results can be evaluated for use as tumor or cancer biomarkers, which can include blocks 218 and 220 of method 200.

[0082] Figure 4 shows the analyzed area of ​​the image. The red line 450 shows the tumor area. The blue line 452 shows one type of polygon. The green line 454 shows the exclusion region. The polygons to be considered are those that are within the tumor but outside the exclusion region. The shaded area 456 shows the polygons that are within the tumor and outside the exclusion region. The metrics of the shaded area 456 can be determined as part of a secondary analysis. D. Exemplary Biomarker Identification

[0083] Cancers can evade immune surveillance and eradication through upregulation of the programmed death 1 (PD-1) and its ligand, the programmed death ligand 1 (PD-L1) pathway on tumor cells and in the tumor microenvironment. Blockade of this pathway with antibodies against PD-1 or PD-L1 has led to remarkable clinical responses in some cancer patients.

[0084] The number of T cells visible in the images of ovarian cancer tissues is analyzed. The T cells tested include CD3, CD8, GZMB, and GZMK. These cells were analyzed in the tumor and stromal regions. The cells were determined to be in either the tumor area or the stoma area (outside the tumor area). The methods used to determine inside or outside the tumor area include those described in FIG. 2 (e.g., blocks 202-212). A total of 17 ovarian cancer tissues were tested. Of these 17 tissues, 9 tissues contained tumors infiltrated with immune cells and 8 tissues contained tumors that excluded immune cells.

[0085] FIG. 5A shows an exemplary multiplex fluorescence assay image of tumor tissue. The top image shows tumor tissue that has excluded immune cells. The top image shows arrows pointing to the co-localization of CD3 and GZMB. The bottom image shows tumor tissue that has infiltrated immune cells. The bottom image shows arrows pointing to specific cells within the tumor. The green arrows point to the co-localization of CD3, GZMB, and PD-1 cells. The yellow arrows point to the location of CD3 and GZMB. FIG. 5A shows that in the excluded tumor, no PD-1 cells were found, but CD3 and GZMB were found. Additionally, FIG. 5A shows that in the infiltrated tissue, PD-1 cells were found along with CD3 and GZMB.

[0086] FIG. 5B graphs the results of the analysis of PD-1 in all 17 tissues. The x-axis shows tumor and stromal tissues, each divided between invaded and excluded tumor tissues. Cells are assigned to these tumor and stromal regions using a method similar to method 200. The y-axis shows the ratio of PD1, GZMB, and CD3 cells to GZMB and CD3. For the tumor area, the percentage of GZMB+ T cells positive for PD-1 is significantly higher in invaded tumors compared to excluded tumors for the tumor area (p-value=0.05). For the stromal region, the percentage of GZMB+ T cells positive for PD-1 is slightly higher for invaded tumors (p-value=0.3). The results support a higher activation state or increased number of GZMB+ T cells in invaded tumors. FIG. 5B shows that biomarkers based on the number of PD1, GZMB, CD3 cells relative to GZMB, CD3 cells in the tumor area can be used to identify invaded or excluded tumor tissue. Invaded tumor tissue can be used to evaluate the work of drugs. In some embodiments, invaded tumor tissue can mean that immune cells are trying to kill tumor cells and drugs can be effective. Furthermore, FIG. 5B shows that the relative number of cells in the stroma is less effective in identifying invaded or excluded tumor tissue than the use of cells in the tumor tissue. Referring to method 200 of FIG. 2, the number of cells can be a metric and the result can be a ratio of the number of cells to other cells. The reference value can be the ratio in excluded tumor tissue or the ratio in the stroma.

[0087] FIG. 6 graphs the results of the analysis including GZMB and GZMK in all 17 tissues. The x-axis shows the tumor tissue excluded and invaded. The y-axis shows the number of cells, either CD8 and GZMK or GZMB cells, relative to the CD8 cells. The graph on the left shows the results for GZMB. The graph on the right shows the results for GZMK. The graph on the right shows that the CD8, GZMK cells relative to the CD8 cells are higher in the excluded samples compared to the invaded samples. The graph on the left shows that there is no significant difference between the invaded and excluded samples for the CD8, GZMB cells relative to the CD8 cells. The number of CD8, GZMK cells relative to the CD8 cells can be used as a biomarker for invaded tumor tissue. II. Cell phenotype density or counting

[0088] 7 shows a method 700 for analyzing an image of a biological sample. The image may be any image described herein.

[0089] At block 702, a first count or a first density of a cellular phenotype having a first region of interest can be determined. The first region of interest can be a tumor or any region of interest described herein. The cellular phenotype can be determined to be within the first region of interest by any method described herein, including method 200.

[0090] At block 704, a second count or a second density of the cellular phenotype within a second region of interest can be determined. The second region of interest can be within the first region of interest. The second region of interest can include the stroma, epitumor, or peritumor. The cellular phenotype can be determined to be within the second region of interest by any method described herein, including method 200.

[0091] A third count or a third density of the cell phenotype in a third region of interest can be determined. The third region of interest can be within the first region of interest. The third region of interest can be stroma, epitumor, or peritumor, and can be different from the second region of interest.

[0092] At block 706, an output may be generated. The output may include the first count or the first density. The output may further include the second count or the second density. The output may also include a third count or the third density.

[0093] The method 700 may further include comparing the second count or the second density to a threshold. A classification of the first region of interest may be determined based on the comparison. For example, if the second count or the second density exceeds a threshold, the classification of the first region of interest may be that disease is present. For example, the classification may be that the tumor is malignant and growing.

[0094] The method 700 can include comparing the first count to a first threshold or comparing the second count to a second threshold. If the first count exceeds the first threshold or the second count exceeds the second threshold, the generated output can include the first density and / or the second density instead of or in addition to the counts.

[0095] Figure 8A shows an analysis of counts or densities of different cell phenotypes. The tumor 802 is outlined in red. The epitumor containing area (804) is outlined in blue and the epitumor excluded area (806) is outlined in green. The stromal region is also an area within the tumor area that is excluded from the epitumor area. Cells of a particular phenotype (phenotype A) are shown outlined in magenta (e.g., cell 808).

[0096] Figure 8B shows the results of cell phenotype counting. Figure 8B shows that there are five cells with phenotype A in the tumor. Of these five, one is in the epitumor and four are in the stroma. The areas of the tumor, epitumor, and stroma are calculated. The areas of the epitumor and stroma are areas within the tumor. The density of phenotype A is then determined by taking the number of cells and dividing by the corresponding area. In some embodiments, the density can be calculated using the area covered by the cells instead of the cell count.

[0097] The results can be used as disclosed in method 200. III. Phenotypic distance

[0098] 9 illustrates a method 900 for analyzing an image of a biological sample. The image may be any image described herein.

[0099] At block 902, a plurality of first cells may be identified within the tumor. Each first cell of the plurality of first cells may have a first phenotype. The plurality of first cells may be determined to be within the tumor by any method described herein, including method 200.

[0100] At block 904, a plurality of second cells may be identified within the tumor. Each second cell of the plurality of second cells may have a second phenotype. The plurality of second cells may be determined to be within the tumor by any method described herein, including method 200.

[0101] In block 906, for each first cell of the plurality of cells, the shortest distance to the second cell of the plurality of second cells can be calculated. The shortest distance is the distance between the first cell and any second cell of the plurality of second cells. The calculation can use nearest neighbor search, including K-nearest neighbor (KNN), kd-tree, vantage point tree, and ball tree, as examples. The commonly used distance metric by calculating two pairs of multiple cells / phenotypes using Euclidean distance causes extremely complex calculation compared to KNN search and other nearest neighbor search. The brute force method on distance is considered to require a lot of calculation. This method can be used to verify whether KNN search or other nearest neighbor search obtains accurate results in small subset examples.

[0102] At block 908, a result based on the calculated shortest distances may be generated. The result may be a list of shortest distances. In some embodiments, a statistic of the statistics (e.g., mean, median, mode, percentile, standard deviation) may be calculated. The result may include the statistic. The result may be used as disclosed in method 200.

[0103] In some embodiments, a count of first cells having a shortest distance within a particular range can be determined. The results can include a count of the first cells. In embodiments, a count of first cells having a shortest distance within several different ranges can be determined. The results can include these counts. These counts can be presented as a histogram.

[0104] Figure 10 shows the calculation of distance from phenotype A to phenotype B. Figure 8 shows only one phenotype A, namely A1. There are four phenotypes B, namely B1, B2, B3, and B4. The shortest distance is between A1 and B3 (shown by the dashed line).

[0105] Figure 11 shows multiple phenotypes A and B. There are three phenotypes A, shown as red circles, in the tumor. There are two phenotypes B, shown as blue Xs, in the tumor.

[0106] FIG. 12 shows the distance between phenotype A and phenotype B. The shortest distance is shown by the dashed line. The shortest distances are found to be 66.0, 173.4, and 201.4. Statistics can then be calculated for the distances. For example, the arithmetic mean shortest distance is 147.0 with a standard deviation of 71.5.

[0107] The arrangement of phenotype A and phenotype B may be more complex than the arrangement shown in FIG. 11. Calculating all distances may be computationally intensive and inefficient. A K-nearest neighbor (KNN) search is used to search for nearest neighbors between two different phenotypes. The KNN search in the configuration of FIG. 11 resulted in the same distance, average, and standard deviation as calculating all distances between the two phenotypes. Compared to brute force methods (e.g., calculating Euclidean distances between two sets of multiple cells / phenotypes), the KNN search is computationally more efficient.

[0108] Figure 13 shows a possible table of results from this analysis. Figure 13 shows the average and standard deviation of the closest distances. Additionally, Figure 13 lists the number of phenotypes A in tumors with a particular range of phenotypes B. In this example, one phenotype A is 66.0 microns from one phenotype B. This one phenotype A corresponds to a value of 1 for the number of phenotypes A within 70 microns, 80 microns, 90 microns, and 100 microns of phenotype B. IV. Cell-to-region distance

[0109] The cell-to-region distance of a phenotype can be useful in clinical analysis. Figure 14 shows the epitumor contour and the phenotype represented as a blue X. The innermost circle 1402 represents the epitumor. Phenotypes have different distances from the epitumor. Phenotypes within the epitumor are considered to have a distance of zero. Phenotype distance to the epitumor can be reported in a histogram with different distance distribution ranges to the epitumor.

[0110] Figure 15A shows an image of a biological sample. Figure 15B shows a magnified image with a red circle annotating the tumor area. Figure 15C shows an image with a blue polygon representing the epitumor and a black circle representing the phenotypic cells. The red outline represents the tumor area.

[0111] Figure 16 shows the selection of a single epitumor polygon within a tumor. A small number of cells of phenotype A are indicated by black circles. Here, the epitumor is selected as an example polygon (similarly, blood vessels and / or fibroblast activation protein stroma could also be selected as polygons).

[0112] Figure 17 shows a single epitumor surrounded by five contours. The five contours are simulated based on distance from the epitumor polygon. The contours are 10, 20, 30, 40, and 50 microns away from the epitumor polygon. Cells of a particular phenotype are shown as red dots. From the contours, Figure 17 shows that three cells are within 10 microns of the epitumor. All four cells are within 40 microns of the epitumor.

[0113] Figure 18 shows a table with the results of the distance from cells to the epitumor of phenotype A. The mean distance and standard deviation of the distance can be determined after calculating each distance from the cells to the epitumor.

[0114] 19 illustrates a method 1900 for analyzing an image of a biological sample. The image may be any image described herein.

[0115] At block 1902, a plurality of first cells can be identified within the tumor. Each first cell of the plurality of first cells can have a first phenotype. The plurality of first cells can be determined to be within the tumor by any method described herein.

[0116] At block 1904, one or more regions within the tumor may be identified. Each of the one or more regions may correspond to a tissue block or biological object. The one or more regions may be determined to be within the tumor by any method described herein.

[0117] In block 1906, the shortest distance between the first cell and each of the one or more regions can be calculated using a nearest neighbor search. Examples of nearest neighbor searches include K-nearest neighbor (KNN), kd-tree, vantage point tree, and ball tree. The KNN search can determine the cell / phenotype with the closest vertical distance to the region. In some embodiments, the distance can be the closest horizontal distance or the distance across another dimension. The distance between the cell / phenotype and the region can be calculated. The calculated distances can be sorted to analyze the distance distribution from the cell / phenotype to the region. Other nearest neighbor searches can be used in a similar manner.

[0118] At block 1908, a result can be generated based on the calculated shortest distance. The result can be similar to FIG. 16. The result can include a count of cells / phenotypes within a particular range. The result can be used as disclosed in method 200.

[0119] The KNN search yielded the results in Figure 16. The KNN search avoids the computation of the perimeter distance contour mask used in the simulation study. Compared to the use of the contour mask, the KNN search obtains the same results faster and more efficiently. V. Compression Level

[0120] 20 illustrates a method 2000 for analyzing an image of a biological sample. The image may be any image disclosed herein.

[0121] In block 2002, a plurality of regions within a tumor may be identified. Each region of the plurality of regions may correspond to a tissue block or a biological object. Block 2002 may be similar to block 1304.

[0122] At block 2004, a compression level for each of the plurality of regions may be determined. The compression level may be calculated from a ratio of area to one or more characteristic dimensions. The characteristic dimension may be the width, length, or major axis length of the region, or the minor axis length of the region. The compression level may be the extent to which the lumen of the blood vessel is free of CD31 cells, which are a biomarker for the blood vessel wall.

[0123] At block 2006, results based on the compression levels may be generated. The results may include a list of compression levels. In some embodiments, the results may include a count of regions within categories of compression levels. The various categories of compression levels may be based on the degree of compression (e.g., low, medium, high). The results may include the area of ​​the regions. The results may further include statistics of the compression levels and / or the area. The results may be used as disclosed in method 200.

[0124] Figure 21 shows a tumor with regions of different compression levels. The tumor is outlined with a red line. The regions are shown as solid green polygons. The polygons at the low compression level are circular. The polygons at the high compression level are long and thin with a large aspect ratio of height to width. The polygons at the medium compression level are not as long and thin as the polygons at the high compression level, nor are they as circular as the polygons at the low compression level.

[0125] Figure 22 shows three polygons with low, medium, and high compression levels, each marked with a pink outline: Polygon 2202 has a low level of compression, Polygon 2204 has a medium level of compression, and Polygon 2206 has a high level of compression.

[0126] Figure 23 shows a possible list of compression level results. The number of polygons in each category, low, medium, and high, is listed. The average compression level is also listed. The polygon area and total polygon area are also reported. VI. Computer Systems

[0127] The image analysis methods described above can be performed by a computer system, which can include computer system 10 of FIG. 24. In some embodiments, the computer system includes a single computer device, and the subsystems can be components of the computer device. In other embodiments, the computer system can include multiple computer devices, each of which is a subsystem with internal components. The computer system can include desktop and laptop computers, tablets, mobile phones, other mobile devices, and cloud-based systems.

[0128] The subsystems shown in FIG. 24 are interconnected via a system bus 75. Further subsystems are shown, such as a printer 74, a keyboard 78, a storage device 79, a monitor 76 (e.g., a display screen such as an LCD or LED) connected to a display adapter 82. Peripherals and input / output (I / O) devices connected to the I / O controller 71 can be connected to the computer system by any of several means known in the art, such as an input / output (I / O) port 77 (e.g., USB). For example, the I / O port 77 or an external interface 81 (e.g., Ethernet, Wi-Fi, etc.) can be used to connect the computer system 10 to a wide area network such as the Internet, a mouse input device, or a scanner. The interconnection via the system bus 75 not only allows for the exchange of information between the subsystems, but also allows the central processor 73 to communicate with each subsystem and control the execution of a plurality of instructions from the system memory 72 or storage device 79 (e.g., a fixed disk such as a hard drive or an optical disk). The system memory 72 and / or the storage device 79 can embody a computer-readable medium. Another subsystem is a data collection device 85, such as a camera, microscope, microphone, accelerometer, etc. The IHC data described herein can be acquired by the data collection device 85. In some embodiments, the image data can be transferred onto a storage device 79 or stored in the system memory 72. Any of the data described herein can be output from one component to another and to a user.

[0129] A computer system may include multiple identical components or subsystems connected to each other, for example, by an external interface 81, an internal interface, or a removable storage device that can be connected and disconnected from one component to another. In some embodiments, computer systems, subsystems, or devices may communicate over a network. In such cases, one computer may be considered a client and another computer a server, each of which may be part of the same computer system. Each of the clients and servers may include multiple systems, subsystems, or components.

[0130] Aspects of the embodiments can be implemented in the form of control logic using hardware circuits (e.g., application specific integrated circuits or field programmable gate arrays) and / or using computer software in conjunction with a general programmable processor in a modular or integrated fashion. As used herein, a processor can include a single-core processor, a multi-core processor on the same integrated chip, or multiple processing units located on a single circuit board or networked, as well as dedicated hardware. Based on the disclosure and teachings provided herein, a person skilled in the art will know and understand other ways and / or methods to implement embodiments of the present invention using hardware and combinations of hardware and software.

[0131] Any of the software components or functions described in this application may be implemented as software code executed by a processor using any suitable computer language, such as, for example, Java, C, C++, C#, Objective-C, Swift, or scripting languages, such as Perl or Python, using traditional or object-oriented techniques. The software code may be stored on a computer-readable medium for storage and / or transmission as a series of instructions or commands. Suitable non-transitory computer-readable media may include random access memory (RAM), read-only memory (ROM), magnetic media such as hard drives or floppy disks, or optical media such as compact disks (CDs), DVDs (digital versatile disks), Blu-ray disks, flash memory, and the like. The computer-readable medium may be any combination of such storage or transmission devices.

[0132] Further, such programs may be encoded and transmitted using carrier signals adapted for transmission over wired, optical, and / or wireless networks conforming to various protocols, including the Internet. Thus, computer-readable media may be generated using data signals encoded with such programs. Computer-readable media encoded with program code may be packaged with a compatible device or provided separately from other devices (e.g., via Internet download). Any such computer-readable media may be located on or within a single computer product (e.g., a hard drive, CD, or an entire computer system), or may be present on or within different computer products within a system or network. A computer system may include a monitor, printer, or other suitable display for providing any of the results described herein to a user.

[0133] Any of the methods described herein may be performed in whole or in part in a computer system including one or more processors that may be configured to perform the steps. Thus, embodiments may be directed to a computer system configured to perform the steps of any of the methods described herein, possibly including different components performing each step or each group of steps. Although steps of the methods herein are presented as sequenced steps, they may be performed simultaneously, at different times, or in different orders. Furthermore, some of these steps may be used together with some of other steps from other methods. Also, the steps may be optional in whole or in part. Furthermore, any step of any method may be performed in a module, unit, circuit, or other means of a system for performing these steps. These steps may be stored as image analysis code in the system memory 72 or storage device 79.

[0134] The specific details of the particular embodiments may be combined in any suitable manner without departing from the spirit and scope of the embodiments of the invention. However, other embodiments of the invention may be directed to the particular embodiments with respect to each individual aspect, or to particular combinations of these individual aspects.

[0135] The foregoing description of exemplary embodiments of the disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or to limit the disclosure to the precise forms described, many modifications and variations are possible in light of the above teachings.

[0136] The terms "a," "an," or "the" are intended to mean "one or more," unless specifically indicated otherwise. The use of "or" is intended to mean "inclusive or," not "exclusive or," unless specifically indicated otherwise. A reference to a "first" element does not necessarily require that a second element be provided. Furthermore, a reference to a "first" or "second" element does not limit the referenced element to a particular location, unless expressly stated. The term "based on" is intended to mean "based at least in part on."

[0137] All patents, patent applications, publications, and descriptions mentioned herein are incorporated by reference in their entirety for all purposes. None are admitted to be prior art.

Claims

1. 1. A method for identifying a tumor biomarker using an image of a biological sample, comprising: identifying within the image a tumor area indicating a boundary of the tumor; identifying a plurality of regions of the image, each region corresponding to a tissue block or biological object; defining, for each of the plurality of regions, a boundary area surrounding the region, the boundary area including a polygon or an ellipse; for each region of a first subset of the plurality of regions whose boundary area is entirely within the tumor area, determining that the region is attributable to the tumor; determining, for each region of a second subset of the plurality of regions, that the boundary area of ​​the region is partially within the tumor area; for each region of the second subset of the plurality of regions, determining whether the region is attributable to the tumor based on an intersection of the region with the tumor area; accessing, for each region attributable to the tumor, a metric that characterizes a biological observation; generating a result value based on the accessed metrics of the region attributable to the tumor; The method includes:

2. The metric is the count or density of a particular type of cell within the region. The method of claim 1.

3. The method of claim 2, wherein the count or density of the particular type of cells is a count or density of the region within the tumor area or entirely outside the tumor.

4. The method of claim 2 , wherein the count or density of the particular type of cell is the count or density in an area entirely outside the tumor.

5. a third subset of the plurality of regions corresponds to a tissue block or a biological object; The method further includes accessing the metric characterizing the biological observation for each region of the third subset of the plurality of regions; 3. The method of claim 2, wherein generating the value of the result comprises determining a value of a function including the metrics for each region attributable to the tumor and for each region attributable to the third subset of the plurality of regions.

6. The plurality of regions include: a first region corresponding to a first tissue block or a first biological object; a second region corresponding to a second tissue block or a second biological object; Including, each of the first and second regions is determined to be attributable to the tumor; the metric characterizing the biological observation identifies a location relative to the region; The method of claim 1 , wherein the result identifies a distance between a first location identified by the metric for the first region and a second location identified by the metric for the second region.

7. said first tissue block or said first biological object being a cell of a first phenotype; the second tissue block or the second biological object is a cell of a second phenotype; The method according to claim 6.

8. the first region corresponds to a cell of a first phenotype; The second region corresponds to the second tissue block. The method according to claim 6.

9. The plurality of regions include: a first region comprising a plurality of first cells having a first phenotype; a second region comprising a plurality of second cells having a second phenotype; Including, each of the first and second regions is determined to be attributable to the tumor; the metric characterizing the biological observation identifies a cell location of each of a plurality of cells within the region; The value of the result is calculating, for each first cell of the plurality of first cells, a shortest distance to a second cell of the plurality of second cells using a nearest neighbor search; The method of claim 1 , wherein the compound is produced by

10. The method of claim 9 , wherein the nearest neighbor search is a K-nearest neighbor (KNN) search.

11. calculating statistics of said shortest distances Further comprising: The method of claim 9 , wherein the value of the result comprises the statistical value.

12. determining a count of first cells whose shortest distance is within a range Further comprising: The method of claim 9 , wherein the value of the result is the count.

13. The plurality of regions include: a first region comprising a plurality of first cells having a first phenotype; one or more second regions; Including, Each of the one or more second regions corresponds to a tissue block or biological object, and the first region and each of the one or more second regions are determined to be attributable to the tumor; the metric characterizing the biological observation identifies at least one location of each region, the at least one location being the location of each cell of the plurality of first cells within the first region, and the at least one location being the location of each second region of the one or more second regions; The value of the result is calculating, for each first cell of the plurality of first cells, a shortest distance to a second region of the one or more second regions using a nearest neighbor search. The method of claim 1 , wherein the compound is produced by

14. The nearest neighbor search is a K-nearest neighbor (KNN) search; For a given second region of the one or more second regions, the KNN search determines the first cell of the plurality of first cells that has a closest vertical distance to the given second region; The method of claim 13 , further comprising the step of calculating the distance between the cell with the smallest perpendicular distance and the given second region.

15. calculating statistics of said shortest distances Further comprising: The method of claim 13 , wherein the value of the result comprises the statistical value.

16. determining a count of first cells whose shortest distance is within a determined range; Further comprising: The method of claim 13 , wherein the value of the result comprises the count.

17. The method of claim 1 , wherein the metric is a compression level.

18. The method of claim 1 , wherein the intersections of the regions with the tumor area are stored in a hash table that includes a data structure that associates the regions with the tumor area.

19. Capturing the image of the biological sample. The method of claim 1 further comprising:

20. The method of claim 1 , wherein the image is a multiplexed immunofluorescent staining image of a tissue section.

21. The method of claim 1 , wherein the tissue block or biological object is stroma, epitumor, peritumor, blood vessels, cell phenotypes, necrosis, tissue folds, staining artifacts, or red blood cells.

22. A computer program comprising instructions which, when executed, control a computing system to perform a method according to any one of claims 1 to 21.

23. 23. A computer readable storage medium containing the computer program of claim 22.

24. 23. A computing system including the computer program of claim 22.

Images