Correlative Image Analysis for 3D Biopsy

Abstract

The present invention relates to image analysis of pathology images. To improve reliability in image analysis of pathology images, a method is provided for assisting in identifying at least one feature of a tissue sample in a microscopic image. The method includes providing a first image of a first microscopy modality representing a region of a tissue sample; providing a second image of a second microscopy modality representing the region of the tissue sample; generating a first high-intensity image by applying a first high-intensity filter to the first image or generating a first low-intensity image by applying a first low-intensity filter to the first image to obtain first information of at least one feature; generating a second high-intensity image by applying a second high-intensity filter to the second image or generating a second low-intensity image by applying a second low-intensity filter to the second image to obtain second information of the at least one feature; calculating a correlation of an image pair including one of the first high-intensity image and the first low-intensity image and one of the second high-intensity image and the second low-intensity image to correlate the first and second information of the at least one feature; and outputting the calculated correlation to assist in identifying the at least one feature of the tissue sample.

Description

[Technical field]

[0001] The present invention relates to image analysis of pathology images, in particular to methods, data processing apparatus and systems, computer program elements and computer readable media for assisting in identifying at least one feature of a tissue sample in a microscopic image. [Background technology]

[0002] Pathological examination of tissue samples, such as human tissue samples or biopsies, involves the identification of specific features of interest in the tissue sample. For example, 3D biopsy analysis requires the detection of incremental increases in the nuclear architecture of cells or density alterations in the internal morphology of the sample. Various microscopy modalities have been developed to provide 3D or 2D microscopic imaging of, for example, optical absorption, reflection, or scattering contrast in biological tissues.

[0003] US Patent No. 9,224,301 B2 describes image analysis of dark field images. Particles to be distinguished can be identified by fluorescence spectroscopy and removed in the analysis of at least one dark field image. However, due to e.g. image artifacts or simple signal intensity variations across the tissue sample, the features of interest can be inaccurately identified. Summary of the Invention [Problem to be solved by the invention]

[0004] There is a need to improve reliability in image analysis of pathology images. [Means for solving the problem]

[0005] The object of the invention is solved by the subject matter of the independent claims, further embodiments are incorporated in the dependent claims. It is to be noted that the aspects of the invention described below also apply to methods, data processing apparatus, systems, computer program elements and computer readable media.

[0006] A first aspect of the present invention relates to a method for assisting in identifying at least one feature of a tissue sample in a microscopic image. The method includes providing a first image of a first microscopic modality representative of a region of the tissue sample; providing a second image of a second microscopic modality representative of the region of the tissue sample; generating a first high intensity image by applying a first high intensity filter to the first image or generating a first low intensity image by applying a first low intensity filter to the first image to obtain a first information of at least one feature; generating a second high intensity image by applying a second high intensity filter to the second image or generating a second low intensity image by applying a second low intensity filter to the second image to obtain a second information of the at least one feature; calculating a correlation of an image pair including one of the first high intensity image and the first low intensity image and one of the second high intensity image and the second low intensity image to correlate the first information and the second information of the at least one feature; and outputting the calculated correlation to assist in identifying the at least one feature of the tissue sample.

[0007] In other words, in order to improve the detection of certain features of a tissue sample, such as cell nuclei, it is proposed to correlate detection information of the certain feature in an image of a first microscopic modality with detection data of the same feature in a second microscopic modality, i.e. to use the feature information from the second modality as complementary feature information of the first modality.

[0008] The tissue sample is selected from, for example, a liver sample, a kidney sample, a muscle sample, a brain sample, a lung sample, a skin sample, a thymus sample, a spleen sample, a gastrointestinal tract sample, a pancreas sample, a prostate sample, a breast sample, or a thyroid sample. The tissue sample is, for example, from a human sample or an animal, such as a mouse sample, a rat sample, a monkey sample, or a dog sample.

[0009] The first and second microscopy modalities include various optical microscopy modalities that allow the investigation of biological structures. The microscopy modalities include fluorescence microscopy imaging, such as epifluorescence microscopy, total internal reflection fluorescence (TIRF) microscopy, confocal microscopy, or multiphoton excitation microscopy. The microscopy modalities also include absorption-based microscopy modalities, such as bright-field microscopy, stimulated luminescence microscopy, photoacoustic microscopy, or optical projection tomography (OPT). In addition, the microscopy modalities include scattering-based microscopy, such as dark-field microscopy and optical coherence tomography (OCT). As will be described in detail below with reference to the exemplary embodiments of Figures 1 and 2A-2C, the first and second microscopy modalities are dark-field fluorescence microscopy imaging.

[0010] To extract information of the features, a high or low brightness filter is applied to the first and second images. For example, a particular cell can be detected by using a fluorescence microscopy method, for example by specially staining the cell membrane with a fluorescently labeled antibody. Using a nuclear stain and a fluorescence microscopy image, information about the cell nuclei can also be extracted. A high brightness filter is applied to the fluorescence microscopy image to identify pixels representative of areas with high sample density and zero or near scattering, which indicate sites with many cell nuclei. For example, to extract information about tubules in a fluorescence microscopy image, a low brightness filter is applied to the fluorescence microscopy image to identify pixels representative of areas with low sample density and high scattering, which indicate sites around the tubules in the tissue sample. As a further example, to extract information about cell nuclei in a dark field microscopy image, a low brightness filter is applied to the dark field image to identify pixels representative of areas with high sample density and zero or near scattering, which indicate sites with many cell nuclei. To extract tubule information in dark field microscopy images, a high intensity filter is applied to identify pixels representative of regions of low sample density and high scattering indicative of sites around the tubules in the tissue sample. The high intensity or low intensity filter is a threshold filter. Thresholding identifies pixels having intensity values ​​within a certain range. In one example, thresholding with a high intensity filter identifies pixels above a certain threshold. In another example, thresholding with a low intensity filter identifies pixels below a certain threshold. Various thresholding techniques are employed, including but not limited to global thresholding, local thresholding, histogram shape based thresholding, cluster based thresholding, and object attribute based thresholding. Global thresholding applies the same threshold to all pixels in the entire image. Local thresholding is applied in situations where the background itself in the image varies in intensity. Color images, such as fluorescent images, can also be thresholded. One approach is to specify separate thresholds for each of the RGB components of the image and then combine them using an AND operation.

[0011] It is also noted that the term "generating a high-luminance image or a low-luminance image" refers to obtaining high-luminance image data or low-luminance image data. It is not necessary to display the high-luminance image or the low-luminance image.

[0012] The correlation uses a correlation coefficient as a measure of similarity between two images in an image pair for each position. The image pair includes one of a first high-brightness image and a first low-brightness image and one of a second high-brightness image and a second low-brightness image to correlate the first information and the second information of at least one feature. In other words, the image pair includes at least one of i) a combination of a first high-brightness image and a second high-brightness image, ii) a combination of a first high-brightness image and a second low-brightness image, iii) a combination of a first low-brightness image and a second high-brightness image, and iv) a combination of a first low-brightness image and a second low-brightness image. The combination of images in the image pair depends on the first and second microscopy modalities and the detected features. For example, to extract information of a cell nucleus, the image pair includes a high-brightness fluorescent microscopy image and a low-brightness dark-field microscopy image, since contrast enhancement works differently in these two modalities, as described in Figures 1 and 2A-2C. On the other hand, to extract the information of the cell nuclei, the image pair includes a high-intensity fluorescent microscope image and a high-intensity bright-field image, since contrast enhancement works similarly in these two modalities. The result is maximum for positions where both images have correspondence. These positions are representative of the pixels of the features identified in the first and second images. As will be explained below, especially with reference to the exemplary embodiment of FIG. 1, the correlation is based on Boolean operations or more complex methods. Thus, the identification of the features is performed, for example manually or automatically, using an algorithm and based on the correlation coefficient.

[0013] Since the first image of the first microscopic modality and the second image of the second microscopic modality do not have the same image artifacts or the same signal intensity variations across the tissue sample, the correlation eliminates or reduces their adverse effects. Thus, the identification process becomes more reliable. This is useful for machine-based identification processes that use image segmentation and feature extraction methods. Therefore, these method steps are performed in the image adjustment stage, which is the next stage of image segmentation and feature extraction. In other words, these method steps process the microscopic image to eliminate or reduce the adverse effects of image artifacts and / or signal variations across the tissue sample, so that the microscopic image is more likely to meet the requirements for the next stage of image processing. Thus, the feature of interest is more likely to be correctly identified.

[0014] According to one embodiment of the invention, the first microscopy modality and the second microscopy modality are different modalities selected from at least one of: fluorescence microscopy imaging, dark-field microscopy imaging, and bright-field microscopy imaging.

[0015] According to one embodiment of the present invention, the first microscopy modality is fluorescence microscopy imaging and the second microscopy modality is dark-field microscopy imaging.

[0016] Fluorescence microscopic imaging allows the visualization of structures such as fluorescently labeled cells stained by fluorescently labeled antibodies on cell membranes or by staining of cell nuclei using nuclear stains or so-called intercalating dyes. Dark-field microscopic imaging allows the discrimination of boundaries and internal cavities. Thus, the combination of the two imaging modalities provides additional information about the tissue. The combination of the two imaging modalities also provides a better assessment of, for example, the presence of tubules inside the imaged tissue.

[0017] According to one embodiment of the present invention, the image pair includes a first high brightness image and a second low brightness image, or a first low brightness image and a second high brightness image.

[0018] This is advantageous for situations where the contrast enhancement of the two microscopy modalities works differently. In one example, an absorption-based imaging modality, e.g. bright field, is combined with a scattering-based imaging modality, e.g. dark field or OCT. In another example, a fluorescence imaging modality is combined with a scattering-based imaging modality. For example, the high-intensity dark field image and the low-intensity fluorescence image represent areas of low specimen density and high scattering, e.g. around tubes or tubules in the sample. On the other hand, the low-intensity dark field image and the high-intensity fluorescence image represent areas of high specimen density and zero or near-zero scattering with many cell nuclei, indicating pathological tissue activity.

[0019] According to one embodiment of the present disclosure, the correlation is calculated based on a Boolean operation.

[0020] For example, pixel-wise Boolean multiplication is performed to obtain the cross-correlation matrix. More sophisticated methods are also used and are described below with particular reference to the exemplary embodiment of FIG.

[0021] A second aspect of the invention relates to a system for assisting in identifying at least one feature of a tissue sample in a microscopic image, the system comprising a data processing device as described above and below and a display, the display being configured to display at least one of the first image and the second image and a calculated correlation between the first extracted information and the second extracted information of the at least one feature output from the data processing device.

[0022] A third aspect of the present invention relates to a data processing device for assisting in identifying at least one feature of a sample in a microscopic image. The data processing device comprises an input unit, an information extraction unit, a correlation unit, and an output unit. The input unit is configured to receive a first image of a first microscopic modality representing a region of the tissue sample and a second image of a second microscopic modality representing said region of the tissue sample. The information extraction unit is configured to apply a first high intensity filter to the first image to generate a first high intensity image or a first low intensity filter to the first image to generate a first low intensity image to obtain a first information of the at least one feature, and to apply a second high intensity filter to the second image to generate a second high intensity image or a second low intensity filter to the second image to obtain a second information of the at least one feature. The correlation unit is configured to calculate a correlation of an image pair including one of the first high-brightness image and the first low-brightness image and one of the second high-brightness image and the second low-brightness image to correlate the first information and the second information of the at least one feature, and the output unit is configured to output the calculated correlation to assist in identifying the at least one feature of the tissue sample.

[0023] For the data processing device the same comments as outlined above for the method apply: the data processing device thus provides image conditioning to remove image artifacts and signal variations across the tissue sample in the microscopic image, thereby making the microscopic image suitable and reliable for further processing in machine-based image processing methods, e.g. image segmentation and feature extraction.

[0024] As used herein, the term "unit" refers to, is part of, or includes an application specific integrated circuit (ASIC), an electronic circuit, a processor (dedicated, shared, or a group of) and / or a memory device (dedicated, shared, or a group of) that executes one or more software or firmware programs, a combinatorial logic circuit, and / or other suitable component that provides the described functionality.

[0025] According to an embodiment of the invention, the data processing apparatus comprises a feature identification unit configured to identify at least one feature based on at least one of the first and second images and the calculated correlation.

[0026] In other words, the data processor performs post-conditioning feature identification, and since the post-conditioning microscope image is more reliable, the likelihood of correctly identifying features of interest in the tissue sample is increased.

[0027] According to one embodiment of the invention, the first microscopy modality and the second microscopy modality are different modalities selected from at least one of: fluorescence microscopy imaging, dark-field microscopy imaging, and bright-field microscopy imaging.

[0028] According to one embodiment of the present invention, the first microscopy modality is fluorescence microscopy imaging and the second microscopy modality is dark-field microscopy imaging.

[0029] Thus, the boundary and the internal cavity can be visualized with dark-field microscopy imaging, and the fluorescent, signature structures (e.g., cancer cells) can be visualized with fluorescence microscopy imaging. In this way, both structures can be combined in one 3D rendering.

[0030] According to one embodiment of the present invention, the image pair includes a first high brightness image and a second low brightness image, or a first low brightness image and a second high brightness image.

[0031] According to one embodiment of the present invention, the correlation unit is adapted to calculate the correlation based on a Boolean operation.

[0032] A fourth aspect of the invention relates to a computer program element for instructing an apparatus as described above and below, which is adapted to perform the steps of the methods as described above and below when executed by a processing unit.

[0033] A fifth aspect of the invention relates to a computer readable medium having stored thereon elements of a program.

[0034] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

[0035] Exemplary embodiments of the invention are described below with reference to the following drawings. [Brief description of the drawings]

[0036] [Figure 1] 1 is a flow chart illustrating a method for assisting in identifying at least one feature of a tissue sample in a microscopic image according to some embodiments of the present disclosure. [Figure 2A] FIG. 2 shows a fluorescent microscope image of a region of a rat liver, according to some embodiments of the present disclosure. [Figure 2B] FIG. 13 shows a dark field microscope image of the same region of a rat liver, according to some embodiments of the present disclosure. [Figure 2C] FIG. 13 shows the overlay of a filtered fluorescent microscopy image and a filtered dark-field microscopy image of the same region of a rat liver, according to some embodiments of the present disclosure. [Figure 3A] FIG. 2 shows a fluorescent microscopy image of a region of the human prostate gland, according to some embodiments of the present disclosure. [Figure 3B] FIG. 2 shows a dark field microscope image of the same region of the human prostate, according to some embodiments of the present disclosure. [Figure 3C] FIG. 13 shows the overlay of a filtered fluorescent microscopy image and a filtered dark-field microscopy image of the same region of the human prostate, according to some embodiments of the present disclosure. [Figure 4] FIG. 1 illustrates a data processing apparatus for assisting in identifying at least one feature of a tissue sample in a microscopic image, according to some embodiments of the present disclosure. [Diagram 5] FIG. 1 illustrates a system for assisting in identifying at least one feature of a tissue sample in a microscopic image, according to some embodiments of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] FIG. 1 shows a flow diagram of a method 100 for assisting in identifying at least one feature 14, 16 of a tissue sample in a microscopic image (see FIGS. 2A-2C and 3A-3C) according to some embodiments of the present disclosure. FIGS. 2A-2C show a set of rat liver images to illustrate steps of a method according to an exemplary embodiment of the present disclosure. In particular, FIG. 2A shows a fluorescence microscopy image of a region of a rat liver acquired under 620 nm laser excitation. In this example, SiR DNA nuclear dye (Spirochrome, Switzerland) is used. It is a far-infrared, fluorescent, cell-penetrating and highly specific probe for DNA (not specific for tumor vs. normal cells). FIG. 2B shows a darkfield microscopy image of the same region of a rat liver acquired under 840 nm laser illumination. FIG. 2C shows the overlay of the filtered fluorescence microscopy image and the filtered darkfield microscopy image.

[0038] In step 110, a first image 10 of a first microscopy modality is provided that represents a region of a tissue sample. For example, the first image 10 is a fluorescent microscopy image as shown in FIG. 2A, and the first microscopy modality is fluorescent microscopy imaging.

[0039] In step 120, a second image 12 of a second microscopy modality is provided that represents the region of the tissue sample. For example, the second image 12 is a darkfield microscopy image as shown in FIG. 2B, where the second microscopy modality is darkfield microscopy imaging.

[0040] In step 130, a first high-brightness image (not shown) is generated by applying a first high-brightness filter to the first image 10. Or, a first low-brightness image (not shown) is generated by applying a first low-brightness filter to the first image 10 to obtain first information of at least one feature 14, 16. It is noted that the first high-brightness image or the low-brightness image refers to the first high-brightness image data or the low-brightness image data. Therefore, it is not necessary to display the first high-brightness image or the low-brightness image.

[0041] The selection of the first high intensity filter or the first low intensity filter depends on the feature to be identified. For example, the high intensity region 14 in the fluorescence microscope image, the first image 10 of FIG. 2A, is representative of areas with high sample density, zero or near zero scattering, and many cell nuclei. Thus, a high intensity filter is applied to the fluorescence microscope image of FIG. 2A to obtain cell nuclei information. On the other hand, a low intensity filter is applied to the fluorescence microscope image of FIG. 2A to identify areas with low sample density and high scattering, and indicating areas around tubes or tubules.

[0042] The high brightness filter is defined as follows:

number

[0043] The low brightness filter is defined as follows:

number

[0044] The threshold T is a manual threshold defined by the user, or it is determined from a histogram by an automated method. For global thresholding, the threshold T is the same across all pixels in the entire image. For local thresholding, the threshold T varies across the entire image and is therefore defined as T(x,y).

[0045] In step 140, a second high-brightness image (not shown) is generated by applying a second high-brightness filter to the second image 12. Or, a second low-brightness image (not shown) is generated by applying a second low-brightness filter to the second image to obtain second information of the at least one feature 14, 16. It is also noted that the second high-brightness image or low-brightness image refers to the second high-brightness image data or low-brightness image data. It is not necessary to display the second high-brightness image or low-brightness image.

[0046] The selection of the second high-brightness filter or the second low-brightness filter depends on the feature to be identified. For example, the high-brightness region 16 in the dark-field microscope image, the second image 12 of FIG. 2B, is representative of an area with high sample density and zero or near-zero scattering, indicating a site around a tube or tubule. Thus, a high-brightness filter is applied to the dark-field microscope image of FIG. 2B to obtain information about the tube or tubule. On the other hand, a low-brightness filter is applied to the dark-field microscope image of FIG. 2B to identify an area with high sample density, zero or near-zero scattering, indicating a site with many cell nuclei.

[0047] Correlation of the image pair is calculated in step 150. The image pair includes one of a first high brightness image and a first low brightness image and one of a second high brightness image and a second low brightness image to correlate first and second information of at least one feature.

[0048] In one example, at least one feature identified includes cell nuclei, and a correlation is calculated between the high-intensity fluorescent microscopy image and the low-intensity dark-field microscopy image, as the images are representative of areas of high sample density, with zero or near zero scattering, indicating areas with many cell nuclei.

[0049] In another example, at least one feature identified includes tubes and tubules, and a correlation between a low-intensity fluorescent microscopy image and a high-intensity dark-field microscopy image is calculated because the images are representative of areas of low sample density and high scattering indicating sites containing tubes and tubules.

[0050] The two previous examples are also combined to generate a superposition so that both features, cell nuclei and tubes or tubules, can be identified in the superposition image 18 as shown in FIG. 2C. In this example, the correlation is calculated based on Boolean operations. Binary images are used, where each pixel can have only one of two values ​​to indicate whether it is part of at least one feature to be identified or not. The correlation is calculated by multiplexing the binary images of the image pair, although more advanced correlation methods are also used. In the superposition image 18, it is easy to see the regions 14 with high sample density and zero or close to zero scattering, indicating sites with many cell nuclei, and the regions 16 with low sample density and high scattering, indicating sites with tubes or tubules.

[0051] Instead of Boolean operations, correlations can also be calculated using more complex methods such as

number

number

number

[0052] In step 160, the calculated correlation is output to assist in identifying at least one characteristic of the tissue sample.

[0053] 3A-3C show a set of human prostate images to illustrate steps of a method according to an exemplary embodiment of the present disclosure. In FIG. 3A, a fluorescent microscopy image 10 of a region of a human prostate is shown acquired under 620 nm laser excitation. SiR DNA nuclear dye (Spirochrome, Switzerland) was used, which is a far-infrared, fluorescent, cell-penetrating and highly specific probe for DNA (not specific for tumor vs. normal cells). In FIG. 3B, a darkfield microscopy image 12 of the same region of a rat liver is shown acquired under 530 nm laser illumination. A filtered fluorescent and filtered darkfield microscopy image overlay 18 is shown in FIG. 3C, and a similar analysis to that of FIG. 2A-2C was performed.

[0054] Correlating feature information obtained from two different microscopy modalities overcomes the problem of image artifacts or simple signal intensity variations across the sample, thereby making the feature identification process reliable, which is advantageous for machine-based methods to provide robust and rapid tissue analysis.

[0055] Optionally, in step 170, at least one feature is identified based on at least one of the first and second images and the calculated correlation.

[0056] In one example, this identification step is performed manually by a user.

[0057] In another example, the identification step is performed automatically based on image segmentation and feature extraction methods. The calculated correlation, resulting for example from a cross-correlation matrix, can serve as a pixel-wise weighting function, which can be multiplied on the first image or the second image to identify a single feature, such as a tube, tubule or cell nucleus, or on the sum of the two images to identify both features. Image segmentation and feature extraction are performed on at least one of the first weighted image and the second weighted image.

[0058] 4 illustrates a data processing apparatus 200 for assisting in identifying at least one feature of a tissue sample in a microscopic image according to some embodiments of the present disclosure. The data processing apparatus 200 comprises an input unit 210, an information extraction unit 220, a correlation unit 230, and an output unit 240 that are part of or include an ASIC, electronic circuitry, a processor (dedicated, shared, or group), and / or a memory device (dedicated, shared, or group) executing one or more software or firmware programs, a combinatorial logic circuit, and / or other suitable components that provide the described functionality.

[0059] The input unit 210 is configured to receive a first image of a first microscopic modality representative of a region of a tissue sample and a second image of a second microscopic modality representative of said region of the tissue sample.

[0060] The information extraction unit 220 is configured to generate a first high-brightness image by applying a first high-brightness filter to the first image or generate a first low-brightness image by applying a first low-brightness filter to the first image to obtain a first information of at least one feature, and generate a second high-brightness image by applying a second high-brightness filter to the second image or generate a second low-brightness image by applying a second low-brightness filter to the second image to obtain a second information of at least one feature. For example, the first microscopy modality and the second microscopy modality are different modalities selected from at least one of fluorescent microscopy imaging, dark-field microscopy imaging, and bright-field microscopy imaging. In one example, the first microscopy modality is fluorescent microscopy imaging, and the second microscopy modality is dark-field microscopy imaging.

[0061] The correlation unit 230 is configured to calculate a correlation of an image pair including one of a first high brightness image and a first low brightness image and one of a second high brightness image and a second low brightness image to correlate the first information and the second information of at least one feature. For example, the image pair includes a first high brightness image and a second low brightness image, or a first low brightness image and a second high brightness image. For example, the correlation unit 206 is configured to calculate the correlation based on a Boolean operation.

[0062] The output unit 240 is configured to output the calculated correlation to assist in identifying at least one characteristic of the tissue sample.

[0063] Optionally, the data processing device further comprises a feature identification unit 250. The feature identification unit 250 is configured to identify at least one feature based on at least one of the first and second images and the calculated correlation.

[0064] 5 illustrates a system 300 for assisting in identifying at least one feature of a tissue sample in a microscopic image, according to some embodiments of the present disclosure. The system 300 comprises a data processing device 200 as described above and below, and a display 310. The display 310 is configured to display at least one of the first image and the second image, and a calculated correlation between the first extracted information and the second extracted information of the at least one feature output from the data processing device.

[0065] In another exemplary embodiment of the invention, a computer program or a computer program element is provided, characterized in that it is adapted to execute, on a suitable system, the steps of the method according to one of the previous embodiments.

[0066] The computer program elements are therefore stored in a computer unit, which is also part of an embodiment of the present invention. The computer unit is adapted to execute or cause the execution of the steps of the method described above. Moreover, the computer unit is adapted to operate the components of the aforementioned apparatus. The computer unit may be adapted to operate automatically and / or to execute user commands. The working memory of a data processor has the computer program loaded therein. The data processor is thus equipped to execute the method of the present invention.

[0067] This exemplary embodiment of the present invention covers both computer programs that use the present invention from the very beginning and computer programs that have been converted to use the present invention by updating existing programs.

[0068] Moreover, the computer program element may provide all the steps required for carrying out the procedures of the exemplary embodiments of the methods as described above.

[0069] According to a further exemplary embodiment of the present invention, a computer readable medium, such as a CD-ROM, is provided having stored thereon the computer program elements described in the previous paragraphs.

[0070] The computer program may be stored and / or distributed on a suitable medium, such as an optical storage medium or a solid-state medium provided together with or as part of other hardware, as well as distributed in other forms, such as over the Internet or other wired or wireless communication systems.

[0071] However, the computer program may also be provided in a network, such as the World Wide Web, and downloaded from such a network into the working memory of a data processor. According to a further exemplary embodiment of the invention, a medium is provided for enabling the downloading of a computer program element configured to implement a method according to one of the previously described embodiments of the invention.

[0072] It should be noted that the embodiments of the present invention are described with reference to different subject matters. In particular, some embodiments are described with reference to method type claims, and other embodiments are described with reference to device type claims. However, a person skilled in the art should gather from the above and below description, in addition to any combination of features belonging to one type of subject matter, unless otherwise notified, and any combination between features relating to different subject matters is also considered to be disclosed together with this application. However, all features can be combined to provide synergistic effects beyond the simple addition of features.

[0073] While the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustration and description are to be considered as illustrative or exemplary and not restrictive. The present invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art, from a study of the drawings, the disclosure, and the dependent claims, in practicing the claimed invention.

[0074] The word "comprising" in the claims does not exclude other elements or steps, and the singular does not exclude a plurality. A single processor or other unit fulfills the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be interpreted as limiting the scope.

Claims

1. 1. A method for operating a data processing apparatus for identifying at least one feature of a tissue sample in a microscopic image, the data processing apparatus comprising an input unit, an information extraction unit, a correlation unit and an output unit, receiving, by the input unit, a first image of a fluorescence microscopy imaging modality representative of the region of the tissue sample; receiving, by the input unit, a second image of a dark field microscopic imaging modality representative of the region of the tissue sample; the information extraction unit generating a first high brightness image by applying a first high brightness filter to the first image to obtain first information of the at least one feature; the information extraction unit generating a second low-brightness image by applying a second low-brightness filter to the second image to obtain second information of the at least one feature; the correlation unit calculating a correlation between the first high brightness image and the second low brightness image to correlate the first information and the second information of the at least one feature; and said output unit outputting said calculated correlation to assist in identifying at least one characteristic of said tissue sample; or the information extraction unit generating a first low-brightness image by applying a first low-brightness filter to the first image to obtain the first information of the at least one feature; the information extraction unit generating a second high brightness image by applying a second high brightness filter to the second image to obtain the second information of the at least one feature; the correlation unit calculating a correlation between the first low-luminance image and the second high-luminance image to correlate the first information and the second information of the at least one feature; and and outputting the calculated correlation to assist in identifying the at least one characteristic of the tissue sample. and 2. The method of claim 1, wherein the correlation unit calculates the correlation based on a Boolean operation.

3. 1. A data processing apparatus for identifying at least one feature of a tissue sample in a microscopic image, the data processing apparatus comprising an input unit, an information extraction unit, a correlation unit and an output unit, the data processing apparatus comprising: the input unit receiving a first image of a fluorescence microscopy imaging modality representative of a region of the tissue sample and a second image of a dark field microscopy imaging modality representative of the region of the tissue sample; the information extraction unit applies a first high-brightness filter to the first image to generate a first high-brightness image to obtain first information of the at least one feature, and applies a second low-brightness filter to the second image to generate a second low-brightness image to obtain second information of the at least one feature; the correlation unit calculating a correlation between the first high brightness image and the second low brightness image to correlate the first information and the second information of the at least one feature; and and wherein the output unit outputs the calculated correlation to assist in identifying the at least one characteristic of the tissue sample; Or, the information extraction unit applies a first low-brightness filter to the first image to generate a first low-brightness image to obtain first information of the at least one feature, and applies a second high-brightness filter to the second image to generate a second high-brightness image to obtain second information of the at least one feature; the correlation unit calculating a correlation between the first low-luminance image and the second high-luminance image to correlate the first information and the second information of the at least one feature; and and outputting the calculated correlation to assist in identifying the at least one characteristic of the tissue sample. A data processing device that performs the above.

4. 4. A data processing apparatus as claimed in claim 3, wherein said correlation unit calculates said correlations based on Boolean operations.

5. A system for assisting in identifying at least one feature of a tissue sample in a microscopic image, comprising a data processing device according to claim 3 or 4 and a display, The system, wherein the display displays at least one of the first image and the second image, and the calculated correlation between the first information and the second information of the at least one feature output from the data processing device.

6. A computer program for instructing a data processing apparatus according to claim 3 or 4, comprising: A computer program which, when executed by a processing unit, performs the steps of the method according to claim 1 or 2.

7. A computer readable medium storing a computer program according to claim 6.

Images