Journal of Microscopy, Vol. 00, Issue 0 2015, pp.

18

doi: 10.1111/jmi.12292

Received 2 March 2015; accepted 24 June 2015

HisTOOLogy: an open-source tool for quantitative analysis

of histological sections
C . M A G L I A R O , A . T I R E L L A , , G . M A T T E I , A . P I R O N E & A . A H L U W A L I A ,
Research Center E. Piaggio, Faculty of Engineering, University of Pisa, Pisa, Italy

Institute of Clinical Physiology (IFC), National Research Council of Italy (CNR), Pisa, Italy
Department of Veterinary Science, University of Pisa, Pisa, Italy

Key words. Colour image processing, histochemical staining, image analysis.

Summary
HisTOOLogy is an open-source software for the quantification
of digital colour images of histological sections. The simple
graphical user interface enables both expert and non-expert
users to rapidly extract useful information from stained tissue
sections. The softwares main feature is a generalizable colour
separation algorithm based on k-means clustering which accurately and reproducibly returns the amount of colour per
unit area for any stain, thus allowing the quantification of tissue components. Here we describe HisTOOLogys algorithms
and graphical user interface structure, showing how it can be
used to separate different dye colours in several classical stains.
In addition, to demonstrate how the tool can be employed to
obtain quantitative information on biological tissues, the effect
of different hepatic tissue decellularization protocols on cell removal and matrix preservation was assessed through image
analysis using HisTOOLogy and compared with conventional
DNA and total protein content assays. HisTOOLogys performance was also compared with ImageJs colour deconvolution
plug-in, demonstrating its advantages in terms of ease of use
and speed of colour separation.
Introduction
Differential staining is widely used in histological samples to
investigate tissue architecture because it enables specific features or compounds of interest to be highlighted (Kiernan,
1999; Horobin & Kiernan, 2002; Coleman, 2006). Specific
chemical dyes can be used to stain tissue slices in which each
component is represented by a different colour (e.g. extracellular matrix, proteins, polysaccharides, cell cytoplasm, cell
nuclei). Histological sections can then be easily analysed using
a microscope, recording the size, shape and number of tissue
components.
Correspondence to: Chiara Magliaro, Research Center E. Piaggio, Largo Lucio
Lazzarino, 1, 56122, Pisa, Italy. Tel: +39-0502217061; fax: +39-0502217051;
e-mail: chiara.magliaro@googlemail.com


C 2015 The Authors
C 2015 Royal Microscopical Society
Journal of Microscopy 

As most modern microscopes are equipped with a colour

CCD (charge coupled device) or CMOS (complementary metal
oxide semiconductor) camera, digitized images can also be
evaluated quantitatively using image processing methods.
With this approach several features of a tissue component
such as size, shape and distribution can be quantified. The
integration of staining methods with image processing represents the digital revolution for the investigation of microscopic
tissue anatomy for a variety of applications. For example, the
cellularity in tissues has diagnostic relevance for the detection
of tumours and tumour progression (Rubin et al., 2001). Histological quantification in forensic medicine, for legal investigations and postmortem evaluation has had significant impact
(Chatelain et al., 2012). However, despite its evident advantages, the use of image analysis to identify and quantify biological structures stained with colours/dyes lags behind manual
observation and qualitative evaluation using scores or suchlike (Belsare & Mushrif, 2012). Although there are several
software packages able to detect and quantify colour information, only a few can perform automatic analysis to reduce
inter- and intra-observer variations in a simple manner, compatible with nonexpert users (Gurcan et al., 2009). Most of the
tools are commercial, or semi-automated and require basic
programming knowledge to create user-specific macros, and
the majority are implemented only for specific tasks (Masseroli
et al., 2000; Berger et al., 2006; Joshi et al., 2007; Sharangpani
et al., 2007; Liu et al., 2013). As a consequence, many histologists examine sections visually without resorting to digital
image analysis.
Beside the lack of simple tools, there are some evident issues in computer-aided colour quantification in histological
samples. A colour can be represented with a broad array of
tints, tones and shades; its perception is also influenced by
several factors (e.g. physical and chemical properties, source
light, background). Moreover, colour sensing devices reproduce only a part of the light spectrum, and their spatial resolution is lower than their black and white counterparts.
Fidelity and day-to-day consistency of cameras can also

C. MAGLIARO ET AL.

fluctuate (Ornberg et al., 1999). Another important aspect

to consider is that colours in histological sections are variable because they are influenced by tissue thickness, which
may range from 50 nm to 100 m, and staining procedures,
which are operator and protocol dependent. Thus, stains can
differ greatly according to the amount of dye used as well
as exposure and rinsing procedures. Furthermore, automated
computer-aided algorithms become useless in case of artefacts
or exceptional variations in the sample (e.g. traces of hair in
histological skin samples), which can only be handled with a
priori knowledge.
To overcome the colour and acquisition limitations, we
have developed an efficient pipeline for separating dye colours
in histological sections, based on the k-means method, a
widely used clustering technique that minimizes the average
squared distance between points in the same cluster (Arthur &
Vassilvitskii, 2007). To enable simple and rapid analysis, the
dye separation routines developed have been integrated in
a software application for quantitative histological analysis,
named HisTOOLogy. A simple graphical user interface (GUI)
allows the user to select the number of colours to be detected
and automatically returns useful information such as the area
covered by each dye and the number of nuclei. All the measurements are stored in a standard data format. In addition,
an interactive crop key enables users to analyse selected areas such that obvious artefacts or areas of non-interest can be
eliminated.
The purpose of this paper is to describe the dye separation algorithm and present HisTOOLogy as an open-source software
for the quantitative analysis of histological images. In addition, we describe a case study to show how HisTOOLogy can
be used as a tool for quantitative stain analysis using standard
analytical assays as a benchmark for precision and ImageJ as
a benchmark for efficiency of image analysis.
Theory and image-processing workflow
In histochemical preparations each tissue component is
stained with a specific dye with a characteristic colour. After an image is acquired using a microscope and digital colour
camera, colour information is stored in 24-bit image files, with
each pixel having a tuple of values (split into three or more
channels) according to the colour model selected. For example, in the RGB colour model, colour information is stored in
a triplet of values, representing the intensity of the primary
colours, red, green, and blue. Each channel is composed of
8 bits, giving a range of intensity values from 0 to 255. In general, for a given colour model, different dyes are considered
as recognizable through different ranges of values on each
channel.
Colour separation algorithm
There is no general rule of thumb for the choice of the colour
model and the threshold values on each channel ch for a spe-

cific dye or colour. Therefore, one of HisTOOLogys features

is to establish an automatic method for colour model selection and colour separation. In particular, it uses the k-means
technique (Arthur & Vassilvitskii, 2007), which splits n data
in k mutually exclusive clusters. This method considers the
data as an object having a specific location in space, and finds
a squared Euclidean metric-based criterion, such that objects
within each cluster are as close to each other as possible, and
as far from other objects in other clusters as possible. The
goodness of the clustering can be evaluated through the so
called Silhouette value for each point, which is a measure of
how similar a point is to points in its own cluster compared to
points in other clusters. Silhouette values can range from 1
(very low similarity) to +1 (very high similarity) (Kaufman &
Rousseeuw, 1990; Chen et al., 2002).
In the case of a digital image, the colour information for
each pixel can be interpreted as coordinates in a given colour
model. Although there are several colour models, RGB and
CMYK colour models were chosen as presets because they
both use colour mixing and are widely used in digital image applications. Both models store individual information as
the amount of colour in three channels: (red, blue, green)
for RGB or four channels (cyan, magenta, yellow, black)
for CMYK.
The algorithm implemented can be divided in four steps, as
illustrated in Figure 1.
(1) An image representing a histological stained section is
mapped in the two colour models: hence, it is represented
by a [R (row) C (column) 3] matrix in the RGB model
and by a (R C 4) matrix in the CMYK model. Then,
the 3D matrices are reorganized in 2D arrays of [(R
C) 3] and [(R C) 4], respectively. In this way,
for each row of the 2D array, representing a pixel, the
colour information is mapped on red, green and blue
columns or on cyan, magenta, yellow and black ones.
At this point, the k-means method is applied on both
the datasets obtained, with k = number of colours to
detect. For instance, in the case of haematoxylin & eosin
(H&E) staining, there are three colours to separate, (i)
the blue-violet of Haematoxylin, (ii) the red-pink of the
Eosin, (iii) the areas of tissue which are not stained (e.g.
blood vessel lumen), or the free space area (FSA).
For both colour model datasets the mapping and clustering routine returns a vector (named IDX) of size
[(R C) 1], containing cluster indices (from 1 to k) for
each pixel.
(2) To assess the goodness of the two different clustering
results, and hence choose the best colour model for
colour separation, the algorithm automatically measures the average silhouette value, Sv , for the k clusters.
The colour-model with the highest Sv is automatically
designated as the colour model of choice.

C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy 

H IST O O L OGY

Fig. 1. The k-means clustering colour separation algorithm for histochemical stains. In the first step, 24-bit colour images are represented in the RGB
and CYMK colour models. The algorithm transforms the image matrix into an array of size (number of pixels number of colours). For a given input
value of k (number of colours to separate), k-means clustering is performed to partition the array into clusters, returning the IDX vector containing
cluster indices (i.e. k value) for each row. In the second step, the average Silhouette value Sv is calculated for each colour model, and the model with the
highest average Sv is selected for subsequent steps. The third step involves pixel colour assignment, whereby each pixel is allocated a specific colour (from
1 to k) according to whether it falls within a predetermined selection (see text for selection criterion). An optional multiband thresholding procedure is
also available for fine tuning of pixel colour assignment. Finally, the fourth step involves image display and pixel counting. Simplified images showing all
k colours are displayed and the algorithm returns the total area occupied by each colour.

(3) For the colour model assigned, the algorithm returns a

matrix, called matrix of colours, of size (k 9) for RGB or
(k 12) for CMYK, whose elements contain, for each
channel ch, the minimum mch , the maximum Mch and
the mean m
ch values of pixel intensity for all k colours.
A pixel is attributed to a given colour i = 1: k (and
so assigned a logical value of 1 for that colour), if its
intensity value I satisfies the range (mch, < Ich < Mch )

C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy 

for all the channels ch. For example, in the RGB colour
model, a pixel with I = (IR , IG , IB ) is assigned to a colour
i = 1: k if m R < I R < MR mG < I G < MG m B <
I B < MB . For each colour, subsequent counting of the
pixels satisfying this criterion gives the amount of area
covered by that colour.
For low contrast images or images with a low signalto-noise ratio or with artefacts, the user may feel it

C. MAGLIARO ET AL.

necessary to manually adjust the results of the colour

separation. In such cases, the algorithm also allows an
optional step, implementing a multiband thresholding
routine. This can be obtained changing the ranges of
pixel intensity [mch : Mch ] for a given channel ch. To facilitate multiband thresholding the algorithm calculates
the average silhouette values Sv for all the different channel combinations (i.e. for RGB colour-model, SvRG, SvGB
and SvRB ). The pair of channels with highest Sv (where
the colours are best clustered) are flagged as optimal for
colour separation.
(4) When the colour separation for all k colours is complete,
the algorithm automatically returns:

r
r
r

The Simplified Colour Images: k images of the same size as

the original in which each pixel covered by colour i = 1: k is
attributed the m
ch value stored in the matrix of colours, with
all other pixels black. These images enable visual evaluation
of the algorithms performance;
The Simplified Merged Image: an image of the same size as
the original which merges all the Simplified Colour Images
obtained. It can be useful to compare the results of the colour
separation with the original image;
The total area covered by each colour;
The mean value of pixel intensity for each channel.

Fig. 2. HisTOOLogy index GUI showing the principal functions for loading, sizing, and cropping an image. The image loaded is H&E stained
porcine liver, with a pixel size of 0.22 m.

HisTOOLogy: an outline
The colour separation algorithm described above has been
implemented as an open-source software named HisR
TOOLogy developed in Matlab
(The MathWorks-sTM, Inc.,
Natick, Massachussets, USA). These and other standard image analysis algorithms are built into a series of GUIs designed
for quantification of histological stains. Here we report a detailed description of all the functions and GUIs implemented in
HisTOOLogy.
Load an image and set pixel size.
Users can load 24-bit colour
images saved in the most common formats (i.e. *.tiff, *.gif and
*.bmp). The image is visualized in the HisTOOLogy index GUI
(Fig. 2) and a structure (the datamatrix) which stores all the
information related to the image, such as pathname and filename is created. The datamatrix is given the same name as
the image file and stored in the same folder. A pop-up menu
enables users to assign the pixel size and the magnification
used.
Region of interest selection.
An image may need to be cropped
to remove regions of noninterest. A tool to select a region of
interest (ROI) enables the user to choose specific areas or to
eliminate others [e.g. to remove vignetting artefacts (Inou`e &
Spring, 1997)]. The tool opens a new window, in which the
user can select the region to crop using a mouse cursor.

Fig. 3. Once the image in Figure 2 is analysed using the k-means colour
separation algorithm (for k = 3), the RGB GUI appears showing the original and Simplified Merged Image and the channel sliders for multiband
thresholding.

Start analysis.
Clicking on the Start Analysis button, a
window appears prompting the user to insert the number of
colours to separate. Assuming that in a specific image the reference ranges of the intensity values on each channels are
the same, the algorithm speed is improved by analysing a
representative crop in which all the colours to detect are represented. For example, for H&E staining, the colours to detect
are the blue-violet of the haematoxylin, the red-pink of the
Eosin and, if needed, the white of FSA. The first and the second
steps of the colour separation algorithm described above are
automatically performed on the image crop in order to choose
the optimal colour model.
Colour model dedicated GUIs. At this point a GUI specific for
the colour channel assigned according to the Sv analysis automatically appears and the colour separation algorithm is
applied to the whole original image (shown in the first panel
on the left in Fig. 3). The Simplified Merged Image representing

C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy 

H IST O O L OGY

Table 1. Datamatrix structure

1

Filename Pathname Pixel

size

Image
area

ROI
Colour Colour
Mean
area
#1
covered
value
area
Ch1

10

11

Mean
value
Chn

Object
counting

12

13

Colour
#2

14

15

Colour
#k

Column 1 to 5: staining-independent parameters; Columns 6 to the end: staining dependent parameters. The number of columns after #5 in the datamatrix
are variable.

the results obtained with the k-means method is automatically

displayed below the original (bottom left panel of Fig. 3). To
visualize the Simplified Colour Images the user has to type a
colour (1 to k) in the dedicated edit. The corresponding Simplified Colour Image is displayed in the top right of Figure 3
with sliders for each channel set on the mch and Mch values,
respectively.
The mch and Mch values can be changed moving the sliders,
whereby the multiband thresholding algorithm is invoked.
The Simplified Colour Image and the Simplified Merged Image in
the panels are automatically refreshed.
For convenience, the Rename button can be used to
rename the colour of interest. For the example shown in
Figure 3, we have used FSA instead of Colour#1 and eosin
in place of Colour#2.
Once the colour separation is satisfactory, the Ok button
is clicked and the data (i.e. the Simplified Colour Images, the
Simplified Merged Image and the amount of covered area for
each colour 1 to k) are automatically saved in the same folder
as the original image.
Object counting. An Object count button has been included
to identify and count objects, which may be useful for nuclear
stains because it allows estimating the number of cells in the
image.
Finally, by clicking on the Save button, all the data and
the simplified images for all the colours are stored and saved
in the datamatrix.

different tissue sections prepared with classical stains. Twenty

slides were randomly chosen from a slide bank at the Department of Veterinary Science, University of Pisa. The slides were
of different tissues belonging to different animals, all stained
by laboratory technicians. The images were taken using a
microscope (Leitz Diaplan, Pisa, Italy) with a 25 objective,
interfaced with a RGB digital camera (Nikon Digital Sight DSU1, Nikon, Italy) and integrated with Nikon NIS-Elements
BR-4.13.00 software. The pixel size was 0.22 m.
The performance of HisTOOLogy in colour separation and
ROI quantification was also compared with ImageJ, a wellknown image processing software tool used in the scientific
community (Abr`amoff et al., 2004). Images were analysed
using a plug-in which implements stain separation using the
colour deconvolution method (Ruifrock & Johnston, 2001).
Four users (two experts and two non-experts) were asked to
use both procedures, to analyse an image representing fresh
frozen porcine liver (prepared as discussed above). Note that
the ImageJ plug-in calculates only the deconvolved image
for each colour to detect, and does not return a binary image. In fact, to extract quantitative parameters, it is necessary to create a macro to threshold each deconvolved image and then compute the areas covered by each dye. These
operations may be difficult for non-expert users and likely explain why histological analyses of stained tissue are frequently
qualitative.

Comparative quantification of samples

Datamatrix structure. The datamatrix structure automatically stores all the extracted parameters in a specific location,
as shown in Table 1. Although the first fields of the structure
are the same for all the images analysed (Table 1, column 1
to 5), the stain-dependent information (representing the areas
in m2 , the mean values of pixel intensity for each channel
and the object count for each colour) are stored in successive
fields.

Materials and methods

Colour separation for classic stains and performance
To demonstrate HisTOOLogys ability to separate the dye
colour components of different stains, we analysed a series of

C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy 

Besides the quantification of colour components in stains,

HisTOOLogy can also be used for the comparative analysis of
histological sections. To evaluate the softwares performance
in comparative quantification of different samples the tool
was used to assess the efficiency of a series of different liver
decellularization protocols, comparing H&E stained decellularized tissue samples with fresh hepatic tissue. In fact hepatic
tissue can be decellularized to produce an extracellular matrix
which preserves the microarchitecture of liver, but the degree
to which the native biochemical and structural composition is
conserved depends very much on the nature of the treatment
(Mattei et al., 2014). Thus our aim was to determine if
HisTOOLogy can be used to correctly identify a decellularization method which removes cells while conserving the
extracellular matrix. Laboratory bimolecular assays for

C. MAGLIARO ET AL.

protein and DNA content are usually used for evaluating

decellularized tissues, and thus were used as benchmarks.
Standardization of sample preparation and acquisition.
Because
tissue preparation and histological staining procedures may
vary from laboratory to laboratory and from user to user,
reliable comparative analysis requires standardised sample
preparation and a preset optical configuration for image acquisition. Specifically, it is essential to ensure that all samples
have the same slice thickness and that the staining treatment
is identical. In addition, when acquiring images, the microscope, objective, colour camera and lamp intensity should be
the same for all samples of interest.
Preparation of decellularized liver sections.
Here we used the
same scheme and methods as in (Mattei et al., 2014), to decellularize 1-cm-diameter 5-mm-thick porcine liver discs (i.e.
DF: a detergent-free protocol; NI, a nonionic decellularization
protocol and I, an ionic decellularization protocol), comparing histological sections with those of freshly frozen untreated
tissue (FF). Details of the decellularization procedures as well
as representative H&E micrographs of fresh and decellularized
liver samples are reported in the supplementary materials.
Immediately after decellularization porcine livers were fixed
with 4% paraformaldehyde solution for 24 h, dehydrated with
ethanol and then embedded in paraffin and sliced in 5 m sections onto glass slides for histochemical staining with H&E,
using the standard laboratory protocol provided in the Supplementary Materials. They were subsequently mounted with
coverslips and observed with a light microscope using a 25
objective. All digital images were acquired using the same
camera settings. Specifically, for each porcine liver treatment,
five images each from three samples were acquired for each of
the n = 4 tissue conditions analysed.
The total DNA content (DNAc) and Total Protein Content
(TPC) were also determined for FF and all decellularized liver
samples to quantify cellularity and matrix preservation using traditional analytical methods Briefly, after solubilizing
the tissue samples in 1 M NaOH, DNAc was measured using the
Nanodrop (Thermo Scientific) method and TCP through the
Bradford assay (Mattei et al., 2014). Details of the assays are
reported in the Supplementary materials.
Image processing.
HisTOOLogy was used with H&E stained
samples to quantify the eosin covered area (ECA), the FSA
and the number of nuclei (NC). The pixel size was set to a
value of 0.22 m, corresponding to the optical configuration used. To assess the goodness of HisTOOLogy measurements, experimental DNAc and TPC (Mattei et al., 2014) data
were compared with NC and ECA, respectively. In fact, haematoxylin reacts with and stains the DNA, while the eosino-philic
structures are generally intracellular or extracellular proteins
(Horobin & Kiernan, 2002).

Fig. 4. Detection of eosin, haematoxylin, and tissue free areas for porcine
liver (the same image as in Figs. 1 and 2) using HisTOOLogy and the
ImageJ deconvolution plugin. Note that the ImageJ plugin does not return
the simplified image representing the free space regions.

Fig. 5. The final Simplified Merged Image of H&E stained porcine liver as
returned by HisTOOLogy after processing.

Results and discussion

HisTOOLogy as a generalized colour separation software for stains
Figure 3 shows the sequence of original and Simplified Merged
Images of fresh frozen porcine liver stained with H&E within
the GUI. In addition, a selection of the 20 images from the slide
bank and their corresponding Simplified Merged Images are
reported in the Supplementary Materials. The images were all
analysed rapidly and the automatic colour separation method


C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy 

H IST O O L OGY

Table 2. Results of dye quantification with HisTOOLogy for porcine liver slices treated with different decellularization protocols

Control (fresh hepatic tissue)

DF (detergent-free decellularization protocol)
NI (nonionic decellularization protocol)
I (ionic decellularization protocol)

ECA (mm2 )

FSA (mm2 )

NC

(2.232 0.108) 102

(2.016 0.275) 102
(1.359 0.203) 102
(1.260 0.069) 102

(2.441 0.098) 103

(5.417 2.833) 103
(1.374 0.290) 102
(1.400 0.069) 102

481.2 20.03
379.6 95.66
2.2 1.3
1.4 1.6

Fig. 6. Comparison of HisTOOLogy results and standard biochemical assays for pig liver subject to different decellularization treatments: (A) ECA vs. TPC
and (B) NC vs. DNAc.

based on k-means clustering was deemed visibly faithful to the

original by all users.

pathologist and technicians, and help turn the historical

techniques of histological staining into a more modern
science, supported by digital methods(Coleman,2006).

HisTOOLogy versus ImageJ

Colour thresholding, the most critical step in histochemical
analysis, was assessed comparing results obtained with HisTOOLogy and ImageJ. According to the four users employed to
test the tool against ImageJ, HisTOOLogy was considered faster
and more intuitive. In fact, the non-expert users were able to
quantify NC and ECA in less than 5 min using HisTOOLogy,
whereas it took more than 10 min with ImageJ. In particular,
the number of operations necessary to determine regions covered by different dyes with ImageJ is difficult for researchers or
technicians who are not familiar with the software. Furthermore, as shown in Figure 4, with the colour deconvolution
plugin there is direct no separation between the TFA, ECA and
NC.
In addition, as reported in Figure 5, HisTOOLogy automatically merges the segmentation results in a simplified image,
which shows the areas corresponding to haematoxylin, eosin
and tissue-free regions.
The rapid and efficient manner in which this open-source
software returns quantitative information on stained tissue
samples should facilitate its widespread use by biologists,

C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy 

HisTOOLogy versus complementary techniques

The analysis of the images with HisTOOLogy enables the identification of the differences between the porcine liver samples
treated with different decellularization methods and the control. The results are summarized in Table 2.
To better investigate the relationship between the analytical
DNAc and TCP data and the corresponding quantification of
NC and ECA, we used a scatter plot (Fig. 6). Because both of
the data are subject to uncertainties, each point has two error
bars. The Pearson productmoment correlation coefficients
(Pearsons r = 0.993 for NC vs. DNA and 0.988 for EAC vs.
TPC) show a positive and linear correlation between the two
pairs of variables. The high value of Pearsons r indicates that
DNA quantification is related to the cell count determined by
HisTOOLogy and the total protein content is related to the
areas stained by eosin.
These results show that the combination of dye specificity
with rapid and precise quantification of digital colour images
allows the analysis and interpretation of stained tissue components in an efficient and meaningful manner.

C. MAGLIARO ET AL.

Conclusion
Histochemical staining procedures use coloured dyes to
highlight a compound. To exploit stain based histological
techniques for quantification of tissue components, we developed HisTOOLogy, an open-source software tool with a
simple, user-friendly interface. As colour quantification requires multiple parameters, we first implemented a method
based on k-means clustering which automatically chooses the
best colour-model for colour separation and clusters image
pixels covered by the same colour. The algorithm also indicates the best channel pair for multiband thresholding should
users wish to improve colour selection using sliders in the
GUI.
Using this method, as well as classical image processing
methods such as cropping and object counting, HisTOOLogy extracts quantitative parameters (e.g. cell density, collagen content) and stores all the measurements in a structured
format (i.e. datamatrix). The data thus stored can be further
used with statistical tools.
In this work, HisTOOLogy was applied to analyse the effects
of different decellularization protocols, comparing the results
with complementary techniques (e.g. DNA and TPC). The results show positive and linear correlation between the two
pairs of variables, and attest to the validity of image based
quantification of stains for assessing tissue composition.
HisTOOLogy can be downloaded at http://www.centro
piaggio.unipi.it/content/histoology-download.
Acknowledgement
This work was partially funded by the European Union Seventh Framework Program (FP7/20072013) under the grant
agreement no. 304961, ReLiver. The A. Tirellas current address is: annalisa.tirella@manchester.ac.uk.
Disclosures
The authors have nothing to disclose.
References
Abr`amoff, M.D., Magalhaes, P.J. & Ram, S. J. (2004) Image processing
with ImageJ. Biophoto. Intl. 11, 3643.
Arthur, David, & Sergei, Vassilvitskii (2007) k-means++: The advantages of careful seeding. Proceedings of the eighteenth annual ACMSIAM symposium on Discrete algorithms. Soc. Ind. Appl. Math.,
10271035.
Belsare, A. & Mushrif, M. (2012) Histopathological image analysis using image processing techniques: an overview. Signal Image Process. 3,
2336.

Berger, C.E., deKoeijer, J.A., Glas, W. & Madhuizen, H.T. (2006)

Color separation in forensic image processing. J. Forensic Sci. 51,
100102.
Chatelain, D., Hebert, A., Trouillet, N. et al. (2012) Interet de lanalyse
anatomopathologique dans une serie de 400 autopsies medicolegales.
Annales de pathologie. 34(1), 413, Elsevie Masson.
Chen, G., Jaradat, S.A., Banerjee, N., Tanaka, T.S., Ko, M.S.H. & Zhang,
M.Q. (2002) Evaluation and comparison of clustering algorithms
in analysing ES cell gene expression data. Stat. Sinica 12, 241
262.
Coleman, R. (2006) The long-term contribution of dyes and stains to
histology and histopathology. Acta Histochem. 108, 8183.
Gurcan, M.N., Boucheron, L.E., Can, A., Madabhushi, A., Rajpoot, N.M.
& Yener, B. (2009) Histopathological image analysis: a review. IEEE
Rev. Biomed. Eng. 2, 147171.
Horobin, R.W. & Kiernan, J.A. (2002) Conns Biological Stains: A Handbook of Dyes, Stains and Fluorochromes for Use in Biology and Medicine,
Biological Stain Commission. Taylor & Francis, Abingdon.
Inoue, S. & Spring, K.R. (1997) Video Microscopy: The Fundamentals.
Springer, New York.
Joshi, A.S., Sharangpani, G.M., Porter, K., et al. (2007) Semi-automated
imaging system to quantitate Her-2/neu membrane receptor immunoreactivity in human breast cancer. Cytometry A. 71, 273
285.
Kaufman, L. & Rousseeuw, P.J. (1990) Finding Groups in Data: An Introduction to Cluster Analysis. John Wiley & Sons, New York.
Kiernan, J.A. (1999) Histological and histochemical methods: theory and
practice. Butterworth-Heinenmann, Oxford.
Liu, F., Mackey, A.L., Srikuea, R., Esser, K.A. & Yang, L. (2013) Automated
image segmentation of haematoxylin and eosin stained skeletal muscle
cross-sections. J. Microsc. 252, 275285.
Masseroli, M., Caballero, T., OValle, F., Del Moral, R.M., Perez-Milena,
A. & Moral, R.G.D. (2000) Automatic quantification of liver fibrosis: design and validation of a new image analysis method: comparison with semiquantitative indexes of fibrosis. J. Hepatol. 32, 453
464.
Mattei, G., Di Patria, V., Tirella, A., Alaimo, A., Elia, G., Corti, A., Paolicchi,
A. & Ahluwalia, A. (2014) Mechanostructure and composition of
highly reproducible decellularized liver matrices. Acta Biomater. 10,
875882.
Ornberg, R.L., Woerner, B.M. & Edwards, D.A. (1999) Analysis of stained
objects in histological sections by spectral imaging and differential absorption. J. Histochem. Cytochem. 47, 13071313.
Rubin, R., Strayer, D.S. & Rubin, E. (2011) Rubins Pathology: Clinicopathologic Foundations of Medicine. Lippincott Williams & Wilkins,
Philadelphia.
Ruifrok, A.C. & Johnston, D.A. (2001) Quantification of histochemical
staining by colour deconvolution. Anal. Quant. Cytol. Histol. 23, 291
299.
Sharangpani, G., Joshi, A., Porter, K., Deshpande, A., Keyhani, S., Naik,
G., Gholap, A. & Barsky, S. (2007) Semiautomated imaging system to
quantitate estrogen and progesterone receptor immune reactivity in
human breast cancer. J. Microsc. 226, 244255.


C 2015 The Authors
C 2015 Royal Microscopical Society, 00, 18
Journal of Microscopy