rus
Issue #7-8/2023
A.I.Akhmetova, I.V.Yaminsky, T.O.Sovetnikov
FEMTOSCAN ONLINE: 3D VISUALIZATION AND PROCESSING OF BIONANOSCOPY DATA
FEMTOSCAN ONLINE: 3D VISUALIZATION AND PROCESSING OF BIONANOSCOPY DATA
DOI: https://doi.org/10.22184/1993-8578.2023.16.7-8.450.455
Atomic force microscopy is a unique tool for obtaining the 3D morphology of biological objects and measuring their properties. To apply the method and interpret the obtained data, an important role is played by software that allows you to correctly process the resulting images, remove scanning artifacts and collect valuable information about objects in the image [1, 2]. FemtoScan Online software implements several functions that greatly facilitate image processing and data collection about objects of interest.
Atomic force microscopy is a unique tool for obtaining the 3D morphology of biological objects and measuring their properties. To apply the method and interpret the obtained data, an important role is played by software that allows you to correctly process the resulting images, remove scanning artifacts and collect valuable information about objects in the image [1, 2]. FemtoScan Online software implements several functions that greatly facilitate image processing and data collection about objects of interest.
Теги: atomic force microscopy bionanoscopy data processing scanning probe microscopy substrate alignment атомно-силовая микроскопия бионаноскопия выравнивание подложки обработка данных сканирующая зондовая микроскопия
INTRODUCTION
Atomic force microscopy allows not only to recreate 3D relief of the samples under study, but also to obtain numerous data on adsorption of particles on various substrates, adhesion, elasticity of the measured objects and other biomechanical properties. AFM can be used to form a morphometric portrait of blood cells [3], assess the effect of cytotoxic substances on tumour cells [4], and study the nature of particle adsorption on sensor substrates [5].
With the abundance of AFM capabilities, it is necessary to pay attention to the methodology of data processing. Scanning in atomic force microscopy is carried out line by line, and at transition to a new line, steps appear on the image (Fig.1a), which, of course, do not exist in reality. This artefact becomes especially evident when the amplitude of height difference changes sharply from row to row, when single objects are observed in AFM, between which there are rows that include only the relief of the smooth surface of the substrate (Fig.1a).
To get rid of this stepped relief, FemtoScan Online software uses the Row Averaging function, which brings the average height in each image row to a single value. In most cases, line averaging together with the Remove Average Slope function allows to obtain a good picture, but images with single objects are not always well processed. Due to the presence on the image of rows of the underlay alone, the average values over the whole frame will be too different, after applying the function a noticeable "Shadow" appears near the largest objects (Fig.1b).
The darkened areas that appear can be removed by selecting large objects with the Area Selection tool and using the Line alignment excluding the selected area function.
For better alignment of the background, you can use the aligning algorithm for underlay using the Image Subtraction function. To do this, we will make three images: the first is the original image, which we will process, the second is the background without objects, and the third is our final image, which at the first stage should be duplicated from the first one. First, we select all the objects of interest in the original image (Fig.2a) and delete them. Next, duplicate the image without objects (Fig.2b) and use the Align to lines function to get the second image with aligned background. Now subtract the image with aligned background from the image without objects (Fig.2b). The difference of the two images is shown in Fig.2c. The obtained result is subtracted from the third image. The final frame can be compared with the original one in Fig.3.
After applying the algorithm, the knocked-out rows finally disappear (Fig.3).
Using the Select Objects function we can also prepare a detailed analysis of particle morphology, for example, for CaP particles we can obtain data on the geometry of each particle (Fig.4): measure perimeter, area, volume, form factor, roughness, average and maximum height.
A particle on the substrate is automatically selected with a green contour, the threshold parameters of selection can be changed. On the basis of selection the software generates a table with data (Table 1):
P – perimeter, the length of the object boundary in the XY plane;
S – area occupied by the object in projection to the XY plane;
RMS – dispersion of the object height (roughness);
Form Factor 1 (FF1) is the ratio of the radius of the circle of equivalent area to the radius of the circle of equivalent perimeter. For a circular object, this Form Factor is equal to one. The more rugged the perimeter of the object, the closer its value is to zero;
Form Factor 2 (FF2) is the ratio of the doubled length of the object’s skeleton to its perimeter. For a thin thread this ratio is equal to one, for a circle it is zero;
H is the maximum height of the object;
H_av – average height of the object.
Having collected such statistics on a sufficient sample of particles, it is possible to form a morphometric portrait of the sample, which will further allow to identify objects when measuring particles of different sizes.
CONCLUSIONS
Software plays an essential role in the probe microscopy data interpretation. With the help of AFM it is possible to evaluate the purity of samples, characteristic geometric dimensions of particles, and propensity of particles to fracture depending on the substrate used. Understanding the algorithms of the software, as well as the use of various functions, maximises the amount of data that can be obtained using this method.
ACKNOWLEDGMENTS
This work was performed under the state order with the financial support of the Physical Department of Lomonosov Moscow State University (Registration subject 122091200048-7).
FemtoScan Online software is provided by Advanced Technologies Center, www.nanoscopy.ru
The authors are grateful to the Innovation Promotion Foundation and the Moscow City Government for all-round support of the activities of YIC "Nanotechnologies".
PEER REVIEW INFO
Editorial board thanks the anonymous reviewer(s) for their contribution to the peer review of this work. It is also grateful for their consent to publish papers on the journal’s website and SEL eLibrary eLIBRARY.RU.
Declaration of Competing Interest. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Atomic force microscopy allows not only to recreate 3D relief of the samples under study, but also to obtain numerous data on adsorption of particles on various substrates, adhesion, elasticity of the measured objects and other biomechanical properties. AFM can be used to form a morphometric portrait of blood cells [3], assess the effect of cytotoxic substances on tumour cells [4], and study the nature of particle adsorption on sensor substrates [5].
With the abundance of AFM capabilities, it is necessary to pay attention to the methodology of data processing. Scanning in atomic force microscopy is carried out line by line, and at transition to a new line, steps appear on the image (Fig.1a), which, of course, do not exist in reality. This artefact becomes especially evident when the amplitude of height difference changes sharply from row to row, when single objects are observed in AFM, between which there are rows that include only the relief of the smooth surface of the substrate (Fig.1a).
To get rid of this stepped relief, FemtoScan Online software uses the Row Averaging function, which brings the average height in each image row to a single value. In most cases, line averaging together with the Remove Average Slope function allows to obtain a good picture, but images with single objects are not always well processed. Due to the presence on the image of rows of the underlay alone, the average values over the whole frame will be too different, after applying the function a noticeable "Shadow" appears near the largest objects (Fig.1b).
The darkened areas that appear can be removed by selecting large objects with the Area Selection tool and using the Line alignment excluding the selected area function.
For better alignment of the background, you can use the aligning algorithm for underlay using the Image Subtraction function. To do this, we will make three images: the first is the original image, which we will process, the second is the background without objects, and the third is our final image, which at the first stage should be duplicated from the first one. First, we select all the objects of interest in the original image (Fig.2a) and delete them. Next, duplicate the image without objects (Fig.2b) and use the Align to lines function to get the second image with aligned background. Now subtract the image with aligned background from the image without objects (Fig.2b). The difference of the two images is shown in Fig.2c. The obtained result is subtracted from the third image. The final frame can be compared with the original one in Fig.3.
After applying the algorithm, the knocked-out rows finally disappear (Fig.3).
Using the Select Objects function we can also prepare a detailed analysis of particle morphology, for example, for CaP particles we can obtain data on the geometry of each particle (Fig.4): measure perimeter, area, volume, form factor, roughness, average and maximum height.
A particle on the substrate is automatically selected with a green contour, the threshold parameters of selection can be changed. On the basis of selection the software generates a table with data (Table 1):
P – perimeter, the length of the object boundary in the XY plane;
S – area occupied by the object in projection to the XY plane;
RMS – dispersion of the object height (roughness);
Form Factor 1 (FF1) is the ratio of the radius of the circle of equivalent area to the radius of the circle of equivalent perimeter. For a circular object, this Form Factor is equal to one. The more rugged the perimeter of the object, the closer its value is to zero;
Form Factor 2 (FF2) is the ratio of the doubled length of the object’s skeleton to its perimeter. For a thin thread this ratio is equal to one, for a circle it is zero;
H is the maximum height of the object;
H_av – average height of the object.
Having collected such statistics on a sufficient sample of particles, it is possible to form a morphometric portrait of the sample, which will further allow to identify objects when measuring particles of different sizes.
CONCLUSIONS
Software plays an essential role in the probe microscopy data interpretation. With the help of AFM it is possible to evaluate the purity of samples, characteristic geometric dimensions of particles, and propensity of particles to fracture depending on the substrate used. Understanding the algorithms of the software, as well as the use of various functions, maximises the amount of data that can be obtained using this method.
ACKNOWLEDGMENTS
This work was performed under the state order with the financial support of the Physical Department of Lomonosov Moscow State University (Registration subject 122091200048-7).
FemtoScan Online software is provided by Advanced Technologies Center, www.nanoscopy.ru
The authors are grateful to the Innovation Promotion Foundation and the Moscow City Government for all-round support of the activities of YIC "Nanotechnologies".
PEER REVIEW INFO
Editorial board thanks the anonymous reviewer(s) for their contribution to the peer review of this work. It is also grateful for their consent to publish papers on the journal’s website and SEL eLibrary eLIBRARY.RU.
Declaration of Competing Interest. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Readers feedback
