Method, computer-accessible medium and system for providing property measurement on a biological tissue sample

A method for determining the presence or absence of at least one property that is at least one of a local property or a global property of a biological tissue sample, comprising:

• obtaining at least one image of the sample;
• searching the at least one image for a presence of at least one particular feature that is contained in a pre-defined set of features;
• assigning, to the at least one particular feature, a corresponding pronunciation index that is a measure of how pronounced the at least one particular feature is in at least one part of the image;
• computing, with a computer processor, at least one discriminant value that is a function of the pronunciation index that is weighted with a particular weight, wherein the weight of each of the pronunciation index is a measure for a relevance of the at least one feature with respect to the at least one property;
• and
• determining whether the at least one property is present or absent in at least one part of the biological tissue sample depending on whether the discriminant value exceeds at least one of a pre-defined threshold or and optimized threshold,
• wherein the image is a multimodal combination of at least two of a coherent anti-stokes Raman scattering (CARS) image, a two-photon excited auto fluorescence (TPEF) image, and a second-harmonic generation (SHG) image, and/or a stimulated Raman spectroscopy (SRS) image, and
• wherein the pre-defined set of features comprises at least one covariance between at least two of the CARS image, the TPEF image, the SHG image or the SRS image.

A method for determining the presence or absence of at least one property that is at least one of a local property or a global property of a biological tissue sample, comprising:

• obtaining at least one image of the sample;
• searching the at least one image for a presence of at least one particular feature that is contained in a pre-defined set of features;
• assigning, to the at least one particular feature, a corresponding pronunciation index that is a measure of how pronounced the at least one particular feature is in at least one part of the image;
• computing, with a computer processor, at least one discriminant value that is a function of the pronunciation index that is weighted with a particular weight, wherein the weight of each of the pronunciation index is a measure for a relevance of the at least one feature with respect to the at least one property;
• and
• determining whether the at least one property is present or absent in at least one part of the biological tissue sample depending on whether the discriminant value exceeds at least one of a pre-defined threshold or and optimized threshold,
• wherein the image is a multimodal combination of at least two of a coherent anti-stokes Raman scattering (CARS) image, a two-photon excited auto fluorescence (TPEF) image, and a second-harmonic generation (SHG) image, and/or a stimulated Raman spectroscopy (SRS) image, and
• wherein at least one of (i) a ranking of the weights associated with the features corresponds to a ranking of Fisher's discriminant ratios (FDR) of the features with respect to a histological index as the at least one property, or (ii) the indices of the corresponding features are derived from the FDR.

A computer-accessible medium which stores a computer program thereon for determining the presence or absence of at least one property that is at least one of a local property or a global property of a biological tissue sample, wherein, when a computer processor executes the computer program, the processor is configured to perform procedures comprising:

• obtaining at least one image of the sample;
• searching the at least one image for a presence of at least one particular feature that is contained in a pre-defined set of features;
• assigning, to the at least one particular feature, a corresponding pronunciation index that is a measure of how pronounced the at least one particular feature is in at least one part of the image;
• computing, with a computer processor, at least one discriminant value that is a function of the pronunciation index that is weighted with a particular weight, wherein the weight of each of the pronunciation index is a measure for a relevance of the at least one feature with respect to the at least one property;
• and
• determining whether the at least one property is present or absent in at least one part of the biological tissue sample depending on whether the discriminant value exceeds at least one of a pre-defined threshold or and optimized threshold, wherein the at least one property includes a presence of a particular structure in a specific place in the biological tissue sample;
• in response to the determining whether the specific structure is present in the specific place, marking the specific structure in the specific place in at least one of (i) the image, or (ii) a processed version of the image;
• and
• computing an appearance of the biological tissue sample that would result from an exposure of the biological tissue sample to at least one staining agent as the processed version of the image,
• wherein the image is a multimodal combination of at least two of a coherent anti-stokes Raman scattering (CARS) image, a two-photon excited auto fluorescence (TPEF) image, and a second-harmonic generation (SHG) image, and/or a stimulated Raman spectroscopy (SRS) image, and
• wherein at least one of a pre-set partial least squares regression model (PLS) or a linear discriminant analysis (LDA) is employed in the computing of the processed version of the image.

A computer-accessible medium which stores a computer program thereon for determining the presence or absence of at least one property that is at least one of a local property or a global property of a biological tissue sample, wherein, when a computer processor executes the computer program, the processor is configured to perform procedures comprising:

• obtaining at least one image of the sample;
• searching the at least one image for a presence of at least one particular feature that is contained in a pre-defined set of features;
• assigning, to the at least one particular feature, a corresponding pronunciation index that is a measure of how pronounced the at least one particular feature is in at least one part of the image;
• computing, with a computer processor, at least one discriminant value that is a function of the pronunciation index that is weighted with a particular weight, wherein the weight of each of the pronunciation index is a measure for a relevance of the at least one feature with respect to the at least one property;
• and
• determining whether the at least one property is present or absent in at least one part of the biological tissue sample depending on whether the discriminant value exceeds at least one of a pre-defined threshold or and optimized threshold,
• wherein the image is a multimodal combination of at least two of a coherent anti-stokes Raman scattering (CARS) image, a two-photon excited auto fluorescence (TPEF) image, and a second-harmonic generation (SHG) image, and/or a stimulated Raman spectroscopy (SRS) image, and
• wherein the pre-defined set of features comprises at least one covariance between at least two of the CARS image, the TPEF image, the SHG image or the SRS image.

A computer-accessible medium which stores a computer program thereon for determining the presence or absence of at least one property that is at least one of a local property or a global property of a biological tissue sample, wherein, when a computer processor executes the computer program, the processor is configured to perform procedures comprising:

• obtaining at least one image of the sample;
• searching the at least one image for a presence of at least one particular feature that is contained in a pre-defined set of features;
• assigning, to the at least one particular feature, a corresponding pronunciation index that is a measure of how pronounced the at least one particular feature is in at least one part of the image;
• computing, with a computer processor, at least one discriminant value that is a function of the pronunciation index that is weighted with a particular weight, wherein the weight of each of the pronunciation index is a measure for a relevance of the at least one feature with respect to the at least one property;
• and
• determining whether the at least one property is present or absent in at least one part of the biological tissue sample depending on whether the discriminant value exceeds at least one of a pre-defined threshold or and optimized threshold,
• wherein the image is a multimodal combination of at least two of a coherent anti-stokes Raman scattering (CARS) image, a two-photon excited auto fluorescence (TPEF) image, and a second-harmonic generation (SHG) image, and/or a stimulated Raman spectroscopy (SRS) image, and
• wherein at least one of (i) a ranking of the weights associated with the features corresponds to a ranking of Fisher's discriminant ratios (FDR) of the features with respect to a histological index as the at least one property, or (ii) the indices of the corresponding features are derived from the FDR.