Digital staining for histopathology multispectral images by the combined application of spectral enhancement and spectral transformation

Abstract

In this paper we introduced a digital staining method for histopathology images captured with an n-band multispectral camera. The method consisted of two major processes: enhancement of the original spectral transmittance and the transformation of the enhanced transmittance to its target spectral configuration. Enhancement is accomplished by shifting the original transmittance with the scaled difference between the original transmittance and the transmittance estimated with m dominant principal component (PC) vectors;the m-PC vectors were determined from the transmittance samples of the background image. Transformation of the enhanced transmittance to the target spectral configuration was done using an nxn transformation matrix, which was derived by applying a least square method to the enhanced and target spectral training data samples of the different tissue components. Experimental results on the digital conversion of a hematoxylin and eosin (H&E) stained multispectral image to its Masson's trichrome stained (MT) equivalent shows the viability of the method.

Pathology Imaging Informatics for Clinical Practice and Investigative and Translational Research

“…For example, it has been shown that multispectral microscopy is able to detect fibrosis in tissue, which can then be digitally stained with results comparable if not superior to those obtained using Trichrome stained slides. 50 Remarkably, multispectral imaging has also been shown to correctly distinguish between cytology cells with high morphological similarity, as shown by one group that analyzed parathyroid adenoma versus thyroid follicular adenoma. 51 In practice, differentiation of these two entities typically requires immunohistochemical staining of the cell block.…”

“…This will also improve accuracy in subsequent image analysis. 50,53 Multispectral fluorescence imaging allows for accurate quantitative evaluation of certain markers that have clinical meaning, such as the estrogen receptor (ER), progesterone receptor (PR) and the proliferation marker Ki67. In addition to quantitative evaluation, this type of imaging can be explored to reveal the type, distribution, intrinsic characteristics and biomarker state of each cell.…”

Digital simulation of staining in histopathology multispectral images: enhancement and linear transformation of spectral transmittance

“…In this paper, we extended the discussion of the digital staining method presented in Ref. 15. This paper is organized as follows.…”

“…To circumvent the limitations of the previous digital staining method, the application of spectral enhancement combined with spectral transformation was introduced. 15 Application of the proposed method to H&E stained tissue images resulted in the vivid discrimination between tissue structures that share similar colorimetric attributes, particularly the discrimination between smooth muscle and collagen fiber. While spectral enhancement increased the spectral color difference between the object of interest and background, application of spectral transformation converted the H&E stained spectral color of the tissue structures to the colors specific to their reaction to the specified stain, i.e., Masson's trichrome.…”

“…[1][2][3][4][5][6][7] In pathology, particular interest is focused on the possible application of this imaging modality to improve tissue detection, classification and visualization. [8][9][10][11][12][13][14][15] In contrast to the conventional RGB-color image acquisitions systems, which use wideband filters, multispectral imaging uses N narrowband filters to capture images across wide spectral range. The utilization of narrowband filters enables a multispectral image acquisition device to capture spectral features that are not normally captured with RGB-color image acquisition devices.…”