The CytoSpec program was developed in the year 2000 by Peter Lasch who worked as a postdoc with Max Diem in the Laboratory for Spectral Diagnosis at Hunter College (CUNY, New York). Starting from the first version in 2000 the software has been continuously improved and updated. Today CytoSpec is known as a specialized software package for vibrational hyperspectral imaging that supports a large number of different tasks for data import, spectral and spatial preprocessing and uni- and multivariate image segmentation. The program is widely used in a scientific environment and was the workhorse in many scientific studies. The CytoSpec software has been continuously developed over the past years and supports now most of the actual hardware platforms, including 64-bit Windows environments, see 64-bit version of CytoSpec for details.
 
How to obtain the program:
 
The full program (commercial version) can be ordered from CytoSpec. Please e-mail your request to CytoSpec (e-mail: order@cytospec.com). A free demo version with limited functionality is also available. Please send an e-mail to service@cytospec.com (link) to obtain a license key for this demo version.
 
To obtain the most recent version of the CytoSpec online help go to the CytoSpec web pages.
 
Introduction to the program/short description:
 
The software package available from CytoSpec is a program designed specifically for the analysis of vibrational hyperspectral (IR and Raman) imaging data sets. The major innovation of the CytoSpec software over competing software packages is its structure that manipulates hyperspectral data cubes, rather than individual spectra. Thus, all operations carried out on the data affect every spectrum, with the number of spectra in the data set limited only by the available memory.
 
The CytoSpec software is a stand-alone, comprehensive package that operates under Windows Vista, Windows 7 and Windows 8/8.1/10. Hyperspectral data sets, in the format defined by the manufacturer of the Raman, or infrared microspectrometer used are imported, converted to and stored in a data matrix format specific to the CytoSpec program.
 
The software permits the standard spectral manipulations customarily found in single spectra analysis software, such as expansion, smoothing, scaling, normalization, etc. Due to the fact that data sets often contain hundreds or thousands of spectra, a number of statistical approaches to the data are built in the software. These can be classified as uni- and multivariate statistical methods. Univariate methods of analysis, included in the software, consist of various mapping displays of hyperspectral data. In these, the user may select band intensities, integrated intensities, frequencies, intensity ratios, etc., to construct pseudo color maps of the spectral data, which may be considered to be slices through the hyperspectral data cube.
 
The multivariate methods of data analysis create spectral correlations and maps by including not just one intensity or frequency point of a spectrum, but by utilizing the entire spectral information. These methods include principal component analysis (PCA), unsupervised methods of cluster analysis and endmember selection methods. The software is configured to allow output of the spectral data in the formats used by other software packages, such as the NeuroDeveloper from Synthon Analytics, an neural network simulator.
 

SYSTEM REQUIREMENTS:
 

Software: Windows Vista or Windows 7/8/8.1/10. Outdated Windows 98, ME, 2k or XP versions are not recommended! CytoSpec 64-bit requires a Windows 64-bit version (Windows Vista or Windows 7/8/8.1/10 64-bit). In addition, CytoSpec 64-bit v.2.00.05 and later requires installation of the Matlab Compiler Runtime (MCR) R2014a 64-bit, see CytoSpec 64-bit for details. CytoSpec 32-bit can be installed on either Windows 32-bit or Windows 64-bit operating system versions. Note that CytoSpec 32-bit cannot use the memory advantage of 64-bit operating systems. Software to display these help files, that is a web browser such as Firefox, Internet Explorer, Opera, Google Chrome, etc.

Hardware: Approx. 100 MB of free disk space for the program plus demo and example files (max) At least 2048 MB of RAM, more RAM is highly recommended Intel Pentium 4 and above, Intel Celeron, Intel Xeon, Intel Core or x86-64 or AMD64-based CPUs (Intel/AMD) At least an 8 bpp 960x720 graphic display. Supported display modes: 960 x 720 (minimum) 1024 x 768 1152 x 864 1280 x 1024 1600 x 1200 and a large variety of wide screen resolutions To use CytoSpec 64-bit a full 64-bit hard- and software environment will be required. Please refer to the CytoSpec 64-bit documentation for details.

INSTALLATION:
 
Commercial versions: Once you have received the CytoSpec CD, USB stick or downloaded the installation archive from CytoSpec's web server, you are ready to install the program. During the installation procedure CytoSpec's files will be automatically extracted from the cabinet and all necessary steps for starting CytoSpec will be taken.
 
Demo versions: Download the CytoSpec 64-bit installation package from the CytoSpec website.
 
p-code versions: Please see CytoSpec 64-bit for details. Note that with the availability of 64-bit stand alone CytoSpec v. 2.00.05 the CytoSpec p-code version will be discontinued.
 
Installation procedure of CytoSpec's 64-bit, stand alone version 2.00.05 and later

• Login with your Windows user account and and unzip the CytoSpec 64-bit installation package. The zip archive 'CytoSpec.64bit.2.00.xx.setup.zip' contains aside from the CytoSpec setup file also the license key 'cytospec.lic' (commercial users only). This key should be stored in the same folder as the setup file.
• Download the Windows version of the Matlab Compiler Runtime (MCR) Environment R2014a 64-bit from the website of The Mathworks. Install the MCR2014a Windows 64-bit. Do not install other versions than MCR2014a 64-bit!
• Start'CytoSpec.64bit.2.00.xx.setup.exe' and follow the instructions of the installation routine
• Trial version users only: the license file 'cytospec.lic' is not included in the installation package. To register for a free, time-limited demo license please send an e-mail with your name and the name of the institution you are working for to the address service@cytospec.com
  The license key will be provided as an e-mail attachment. Copy the key file sent per e-mail into the DiaryDir folder of the CytoSpec installation (this is usually C:\Users\Public\Documents\Matlab, administrator rights are sometimes required).
• When setup is complete you can find a new program group and a CytoSpec shortcut on the desktop.
• Start the program by double clicking on the CytoSpec icon.
• Trial version users only: Note that the temporary license for CytoSpec demo versions will expire after 90 days (regular demo), or 75 days (demo with full functionality).

 
Installation procedure of 32-bit, stand alone CytoSpec versions

• Login with your Windows user account and and unzip the CytoSpec 32-bit installation package. The zip archive 'CytoSpec.32bit.2.00.xx.setup.zip' contains aside from the CytoSpec setup file also the license key 'cytospec.lic' (commercial users only). This key should be stored in the same folder as the setup file.
• Start 'CytoSpec.32bit.2.00.xx.setup.exe' and follow the instructions of the installation routine
• Trial version users only: the license file 'cytospec.lic' is not included in the installation package. To register for a free, time-limited demo license please send an e-mail with your name and the name of the institution you are working for to the address service@cytospec.com
  The license key will be provided as an e-mail attachment. Copy the key file sent per e-mail into the DiaryDir folder of the CytoSpec installation (this is usually C:\Users\Public\Documents\Matlab, administrator rights are sometimes required).
• When setup is complete you can find a new program group and a CytoSpec shortcut on the desktop.
• Start the program by double clicking on the CytoSpec icon.
• Trial version users only: Note that the temporary license for CytoSpec demo versions will expire after 90 days (regular demo), or 75 days (demo with full functionality).

 

The graphical user interface (gui, aka main window) of CytoSpec comes with three panels where spectra and images are displayed (cf. screenshot of the CytoSpec gui below). The two panels to the right are used to display the pseudo color images obtained from original spectral data (upper right panel) or from processed spectra such as preprocessed data, derivatives, and (de)convolution data (see lower right panel). The latter panel is used also for displaying images obtained by means of various multivariate analysis methods: (i) clustering, (ii) endmember selection, (iii) neural network-based segmentation techniques, or by (iv) principal component analysis.
 
The color scheme utilized to display spectral maps can be changed (see button 'set color' of the main window, cf. also the set colors option from the Tools menu bar). In addition, the CytoSpec program offers a lot of useful tools to adjust parameters such as image contrast and color scaling.
 
Furthermore, the CytoSpec program gives you also easy access to the individual pixel spectra. When clicking into a predefined image produced by univariate imaging methods (left mouse button) the
 

• spectrum with the pixel coordinates of the mouse pointer is displayed in the spectra display panel,
• the pixel coordinates of the active spectrum appear in editable boxes, and
• some information of how the spectral image has been produced is displayed in the area between both image panels.

Many options for displaying spectra can be modified (button 'display options', option 'display options' of the Tools menu bar). For more details please refer to the chapter Working with Spectra - Basic Concepts.
 

 
The section below describes the interactive display mode of the CytoSpec v.2.0 program. In this version the following interactive display functions are available:
 
Mouse modes:

• show: displays the pointer's x- and y-position in the 'spectra' information field
• zoom: can be used to interactively enlarge/reduce a part of the spectrum
• move or 'roll' mode': rolls or moves the spectrum in any direction

Individual pixel spectra from a hyperspectral image can be displayed in the left panel of the gui. Options for displaying the spectra can be selected from the main CytoSpec window (the window 'display options' of earlier CytoSpec versions has been removed). The following functions are available:

• display interactively wavenumber and absorbance values (radio button 'show')
• interactive expand/compress mode (radio button 'zoom')
• interactive shift mode (radio button 'move')
• set curves color (list box 'color of spectra')
• zoom window horizontally/vertically (sliders)
• display multiple spectra (check box 'add mode')
• x-y auto scale for active spectrum (check box 'auto scale')

In the mouse mode 'show' the pointer's x- and y-position will be indicated in the 'spectra' information field of the gui.
 
Using the 'zoom' mode: to zoom in (out) spectra in the display window you have first to activate the appropriate radio button. Then, if you wish to enlarge parts of a spectrum you have to set the first position for the spectrum window by a left mouse click. Hold the button and drag a frame by moving the mouse to the second point. If you release the mouse button, the contents of the frame will be expanded to the full display window. To reduce the size of the spectrum, you can use the right mouse button in the same way.
A left mouse click, that is when no frame is drawn, expands the spectrum in the display window by 200%. A right mouse click will reduce the size to 50%.
 
The 'move' mode is used to roll the spectrum in any direction. The display limits are shifted so that all spectra displayed are shifted by the same amount. Again, you have to activate the appropriate radio button before you can start. Spectra can then be moved if you click (left button) into the display window and move the pointer to the desired end point (the mouse pointer will change to a closed hand symbol). Release the left mouse button to finally roll the spectra into the desired direction.
 
How to select spectra for display? Three methods are available to select spectra for display from a hyperspectral map:
 
Method A: Create an univariate image from the hyperspectral map. Then click into one of the images ('show mode' should be activated). The [x,y] pixel coordinates of the spectrum to be displayed are obtained and the respective spectrum is plotted. Furthermore, the [x,y] pixel coordinates of the active spectrum are displayed in the edit fields of the gui (cf. x/y coordinates of active spectrum. Depending on the type of the image (image produced from original, or processed data) the check box 'show work spectrum' is activated (or deactivated). CytoSpec displays an error message in cases when activated images were re-assembled by multivariate imaging methods (PCA, HCA, etc.).
 
Method B: Alternatively, spectra can be selected by manually editing the edit fields '[x,y] coordinates of actual spectrum' and/or by activating the check box 'processed spec'. If the corresponding image, or data block, does not exist (e.g. there is no image from preprocessed spectra), an error message will be displayed.
 
Method C: Selection of spectra for display can be furthermore done by pressing the buttons [+] (plus) or [-] (minus) located in the 'select spectrum' area of the CytoSpec gui. This will stepwise increase/decrease the y, or y-pixel coordinate of the spectrum to be displayed.
 

 

A number of icons is displayed CytoSpec's main window. The following table will provide a short overview about icons and associated CytoSpec functions.
 
Additionally CytoSpec provides many Keyboard Shortcuts to help speed up image re-assembling and exporting data to other applications. You will find that one can save a lot of time by eliminating the need to move the mouse and to choose respective options from the menus bars.
 
Keyboard Shortcuts
 

Control-A	about	Control-C	set colors
Control-D	display spectra	Control-E	edit parameters
Control-F	frequency images	Control-H	help
Control-I	chemical images	Control-W	FWHM images
Control-L	load data	Control-M	save Matlab
Control-O	customize	Control-P	plot
Control-Q	clear all	Control-S	save data
Control-X	quit the program	Control-1	capture image from original data (bmp)
Control-2	capture image from processed data (bmp)	Control-3	capture spectral plot (bmp)
Control-4	export image data (original data, ASCII)	Control-5	export image data (processed data, ASCII)

 
Icons
 

	clear all		load spectral map		save all
	print		customize		export ASCII spectra
	exit		calculate derivatives		quality tests
	batch mode		chemical imaging		chemical movie
	frequency imaging		HCA imaging		Synthon imaging
	autoscale spectra		display spectra		set display limits manually
	grid on/off		set color		capture spectral plot
	capture image (org. data)		capture image (proc. data)		edit
	help		about

 

Internal data organization: In CytoSpec, a data block represents a complete hyperspectral (3D) data cube in which spectral intensities, or absorbance/transmittance data are arranged as functions of two spatial [x,y] and one spectral coordinate. CytoSpec has been designed for manipulating up to four different variations of the actual hyperspectral data cube at the same time. These variations are called data blocks and all different blocks can be stored and loaded in one single file.
 

1. Original spectra: the first block is reserved exclusively for the original spectra, usually absorbance or transmittance spectra, or Raman intensities in case of Raman imaging data. Original data are usually not modified. Exceptions are the functions 'cut' and 'interpolate' of the preprocessing menu and the functions 'swap data blocks', 'rotate', and 'flip' of the 'tools' menu bar.


2. Preprocessed spectra: this data block contains the results of data manipulations on original spectra (e.g. preprocessed data).


3. Derivative spectra: the third data block is reserved for derivative spectra.


4. (De)convolution spectra: block number four contains the results of spatial preprocessing: spatial filtering, 3D-Fourier self deconvolution (FSD).

Spectral and spatial preprocessing, multivariate image analysis, and image re-assembling can be performed by choosing one of the existing data blocks (also called 'source blocks''). The results of data manipulation are stored into so-called 'target data blocks'. Please note that some operations may overwrite existing target data blocks without warning! Target data blocks may vary for distinct functions (for details please refer the the synopses below).
 

I. Preprocessing (except functions 'cut', 'crop', 'interpolate', 'binning',
'derivatives', 'baseline correction', and 'ABS ↔ TR'):

source block		target block
original spectra	»»	preprocessed spectra
preprocessed spectra	»»	preprocessed spectra
derivative spectra	»»	derivative spectra
(de)convolution spectra	»»	(de)convolution spectra

II. 'Derivative calculation':

source block		target block
original spectra	»»	derivative spectra
preprocessed spectra	»»	derivative spectra
derivative spectra	»»	derivative spectra
(de)convolution spectra	»»	derivative spectra

III. 'Cut', 'crop', 'interpolate', 'binning', 'rotate' & 'flip':

source block		target block
original spectra		The functions cut', 'crop', 'interpolate', 'binning', 'rotate ' & 'flip' modify the number of points (pixels), or the point (pixel) spacing in the spectral and/or spatial dimensions, respectively. All available data blocks are modified .
preprocessed spectra
derivative spectra
(de)convolution spectra

IV. Preprocessing: 'baseline correction' & 'PCA based noise reduction':

source block		target block
original spectra	»»	preprocessed spectra
preprocessed spectra	»»	preprocessed spectra
derivative spectra	»»	not possible
(de)convolution spectra	»»	not possible

V. 'Swap data blocks':

source block		target block
original spectra	»»	Data block of original spectra will be overwritten by data block of preprocessed spectra
preprocessed spectra	»»
derivative spectra	»»	not possible
(de)convolution spectra	»»	not possible

VI. 'ABS ↔ TR':

source block		target block
original spectra		Original spectra will be overwritten by converted data. All other data blocks will be deleted!
preprocessed spectra
derivative spectra
(de)convolution spectra

VII. 'Dispersion correction':

source block		target block
original spectra		Dispersion correction can be carried out only with transmittance spectra stored in the data block of original spectra. Spectra are dispersion corrected and converted to absorbance spectra. The target data block is the block of preprocessed spectra. Other data blocks are not modified.
preprocessed spectra
derivative spectra
(de)convolution spectra

VIII. 'Subtraction':

source block		target block
original spectra	»»	preprocessed spectra
preprocessed spectra	»»	preprocessed spectra
derivative spectra	»»	derivative spectra
(de)convolution spectra	»»	not possible

IX. 'Spatial Filtering':

source block		target block
original spectra	»»	(de)convolution spectra
preprocessed spectra	»»	(de)convolution spectra
derivative spectra	»»	derivative spectra
(de)convolution spectra	»»	(de)convolution spectra

 
Starting with CytoSpec version 1.1.04 the presence/absence of data blocks is visualized by four LEDs located in the lower left corner of the main window. A red LED indicates the presence, a black LED the absence of the respective data block.
 
Memory management modes: CytoSpec offers two distinct memory management modes. This functionality was introduced to allow highly memory consuming operations also with limited RAM resources. With version 2.00.01 and the availability of the 64-bit version of CytoSpec the memory mode 'compress' is obsolete and has been removed.
 

1. 'Speed' the fastest mode, but highly memory consuming. Recommended for HCA with hyperspectral data sets containing up to 128 x 128 pixel spectra (32-bit version, 4 GB of RAM required). In this mode all spectral data are hold in memory with 8 byte precision as float64 values. The option is recommended also for HCA of larger hyperspectral data sets when using the 64-bit version of CytoSpec (Matlab toolbox)

 

2. 'Intermediate' - relatively fast, but less memory consuming. Try this option when datasets are large and/or the amount of installed RAM is reduced. Spectral data are stored with 4 byte precision as float32 values on disk. Only the required data block will be loaded and is held during the calculations in memory. When performing HCA, the distance matrix will be held in RAM.

 

3. 'Compression' - this memory option has been removed with version 2.00.01 (see 64-bit version of CytoSpec for details).

 

 

With CytoSpec version 2.00.01 (build version 338) the window 'display options' has been removed. The functionality is now an integral part of the main window of CytoSpec.
 
How to select spectra for display? Three methods are available to select spectra for display from a hyperspectral map:
 
Method A: Create an univariate image from the hyperspectral map. Then click into one of the images ('show mode' should be activated). The [x,y] pixel coordinates of the spectrum to be displayed are obtained and the respective spectrum is plotted. Furthermore, the [x,y] pixel coordinates of the active spectrum are displayed in the edit fields of the gui (cf. x/y coordinates of active spectrum. Depending on the type of the image (image produced from original, or processed data) the check box 'show work spectrum' is activated (or deactivated). CytoSpec displays an error message in cases when activated images were re-assembled by multivariate imaging methods (PCA, HCA, etc.).
 
Method B: Alternatively, spectra can be selected by manually editing the edit fields '[x,y] coordinates of actual spectrum ' and/or by activating the check box 'processed spec'. If the corresponding image, or data block, does not exist (e.g. there is no image from preprocessed spectra), an error message will be displayed.
 
Method C: Selection of spectra for display can be furthermore done by pressing the buttons [+] (plus) or [-] (minus) located in the 'select spectrum' area of the CytoSpec gui. This will stepwise increase/decrease the y, or y-pixel coordinate of the spectrum to be displayed.
 

The basic idea of hyperspectral imaging (or mapping) is to derive for each spectrum of a given set of spatially resolved spectra one single spectral parameter. This parameter is then color scaled and plotted as a function of the spatial [x,y] coordinates. There are a number of techniques, which can be used for hyperspectral imaging. The most frequently used technique is the so-called chemical mapping technique. In chemical imaging (also referred to as functional group mapping) the absorbance, transmittance, Raman intensity, half-width, or the frequency of a given vibrational band is color encoded and plotted against the spatial [x,y] coordinates. This univariate imaging techniques permits to visualize the spatial distribution of functional groups, or specific chemical substances and is extensively applied in biomedical science, remote sensing and other research areas.
 
The CytoSpec program allows creating chemical images by a number of different methods using the absorbance or transmittance, integral absorptions, Raman intensities, ratios etc. These methods are outlined in the chapter Chemical imaging
 
Furthermore, the CytoSpec software offers the opportunity to re-assemble so-called Frequency Imaging. In this method, the maxima of specific bands are obtained and images are re-assembled by plotting the frequency values as a function of the spatial coordinates. A similar strategy allows producing uni-variate pseudo color images from half-widths of vibrational bands, see function FWHM Imaging for details.
 
Other imaging techniques are based on multivariate methods of spectral data analysis. The CytoSpec program (v.2.00.06) offers nine different methods of multivariate spectral imaging:

 
Principal Component Analysis (PCA) imaging
Imaging based on hierarchical cluster analysis: HCA imaging
k-means cluster imaging
Fuzzy C-means cluster imaging
Imaging based on ANN (Artificial Neural Network) analysis (SNNS imaging)
ANN Imaging based on ANN models developed by the help of Synthon's NeuroDeveloper software
Vertex Component Analysis (VCA) imaging
n-findr imaging
Imaging with distances values

These methods are extensively described in the respective chapters of the CytoSpec online help.
 

When starting the CytoSpec program the Main Window and a report window (see screenshot below) will appear. While the graphics user interface (gui) of CytoSpec allows to interact with the program, the report window shows input parameters and displays non-standard errors. These messages are stored in a log-file ('history.log'), which can be found at the following locations:
 

CytoSpec standalone versions (64-bit, 32-bit, v.2.00.05 and later), Windows operating systems (Vista/W7/W8/W8.1/W10): DiaryDir of CytoSpec, usually C:\Users\Public\Documents\Matlab\history.log
 
CytoSpec pcode toolbox under Matlab (64-bit, pre-2.00.05 versions), Windows operating systems (Vista/W7/W8/W8.1/W10): Matlab user directory, usually C:\users\Username\Documents\Matlab\history.log with 'Username' being the Windows user name
 
CytoSpec standalone version (32-bit, pre-2.00.05 versions), Windows operating systems (2k/XP/Vista/W7): RootDir of the CytoSpec program, usually C:\program files\CytoSpec\CytoSpec\history.log or C:\program files (x86)\CytoSpec\CytoSpec\history.log
 

When reporting problems and software failures (bug reports) please send the file 'history.log' to the following e-mail address: service@cytospec.com
 

Example of the CytoSpec report window
 

How to obtain CytoSpec?

A. Ordering the full (commercial) version

B. Downloading an evaluation (demo) version

Installation, System Requirements

Most actual version: CytoSpec stand alone 2.00.06 64-bit (build 360) - please report software bugs

Download CytoSpec's stand alone 64-bit package, requires license key 'cytospec.lic'' and MCR2014a version 2.00.06, build 360, Aug 2019 size: 18000035 bytes (packed) CRC checksum for data: 69AA2BF5

CytoSpec stand alone version 2.00.06 64-bit (build 358)

Download CytoSpec's stand alone 64-bit package, requires license key 'cytospec.lic'' and MCR2014a version 2.00.06, build 358, Nov 2018 size: 17986702 bytes (packed) CRC checksum for data: F376ABAB

CytoSpec stand alone versions 2.00.05 (32- and 64-bit)

Download CytoSpec's stand alone 64-bit package, requires license key 'genkey.gen'' and MCR2014a version 2.00.05, build 353, Feb 2018 size: 10605365 bytes (packed) CRC checksum for data: 2FE9B595

Download CytoSpec's stand alone 32-bit package, requires license key 'genkey.gen' version 2.00.05, build 353, Feb 2018 size: 15594207 bytes (packed) CRC checksum for data: 6CF5EDAB

Test data (binary example files, ASCII data)

Download binary and ASCII test data size: 86876843 byte (packed) CRC checksum for data: 05B03B47

CytoSpec 64-bit Matlab pcode, v. 2.00.04, a Matlab license for R2014a or newer will be required

Download CytoSpec's 64-bit p-code toolbox, requires license key 'genkey.gen' version 2.00.04, build 349, July 2017 size: 824098 bytes (packed) CRC checksum for data: C0F9FF9F

Earlier 32-bit demo versions of CytoSpec:

Download CytoSpec's standalone demo, requires license key 'genkey.gen' version 2.00.04, 32-bit, build 349, July 2017 size: 15643993 byte (packed) CRC checksum for data: 2F3FBCB6

Download CytoSpec's standalone demo, requires license key 'genkey.gen' version 2.00.03, 32-bit, build 348, April 2016 size: 15384132 byte (packed) CRC checksum for data: 40C33B2B

Please note that you accept with downloading the following license conditions:

You are using the program at your own risk! CytoSpec does not take any responsibility for damages, problems etc. resulting from use of this program. CytoSpec also does not give any warranty for bug-free operation, fitness for a particular purpose or the appropriate behavior of the program.
The software is provided 'AS IS'. For your personal use you can make copies and run as many instances as required, but it is not allowed to further distribute this software.

View the complete license agreement