• Ei tuloksia

Automatic analysis of microscopy images

2.2 Hydrogen producing bacteria

2.2.3 Automatic analysis of microscopy images

Despite of the application of microscopy, manual analysis of the microscopy images is user dependent and laborious and moreover the visual quantification of fluorescence intensities is impossible. Possibility to take digital images over bright field or fluorescence microscopy views enables the usage of automatic image analysis for these purposes. The basic flow of automated image analysis is presented in Figure 2.6.

Figure 2.6. Typical stages of image processing.

preprocessing is applied in order to minimiz The next step is segmentation, in which the ob separated objects are measured

In Publication I its analysis resu

measure cell length, width

comparative analysis of images taken from dual stained samples. CellC ha among other

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009) qualify spore harvest purity (Harrold et al. 2011).

Karr et al. (2012) have create bacteria

genotype. In this aspect cell mass, volume and shap

might be small, image processing can aid in enumerating differences that are not observable by naked eye.

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different classes

classification of microbial cells based on morphological features and intensity,

drastic effect on the reactor operation monitoring of H

population structure from microscopy images is a rapid method to obs changes and

classification cou

mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in m pictures to be analyzed or

the images that are analyzed by automated

pictures. The software can not differentiate bacterial cells from similar, thus taking those into the imaging field

should be dilutes

difficult to analyze. Moreover, the cells should locate in a o

6. Typical stages of image processing.

preprocessing is applied in order to minimiz ext step is segmentation, in which the ob separated objects are measured

In Publication I, an easy to used image analysis software its analysis results are validated. C

measure cell length, width

comparative analysis of images taken from dual stained samples. CellC ha among other application

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009) qualify spore harvest purity (Harrold et al. 2011).

Karr et al. (2012) have create bacteria Mycoplasma genitalium genotype. In this aspect

cell mass, volume and shap

might be small, image processing can aid in enumerating differences that are not observable by naked eye.

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

based on

classification of microbial cells based on morphological features and intensity, is presented

drastic effect on the reactor operation

monitoring of H2 producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to obs

changes and allow quick response by operator if needed. Additionally, classification could

mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in m es to be analyzed or

the images that are analyzed by automated

pictures. The software can not differentiate bacterial cells from similar, thus taking those into the imaging field

should be dilutes to proper density before microscopy, since too crowded images are more difficult to analyze. Moreover, the cells should locate in a o

6. Typical stages of image processing.

preprocessing is applied in order to minimiz ext step is segmentation, in which the ob

separated objects are measured to gain the final results (Modified form Lehmussola, 2009).

an easy to used image analysis software lts are validated. C

measure cell length, width and intensity

comparative analysis of images taken from dual stained samples. CellC ha applications to analyze

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009) qualify spore harvest purity (Harrold et al. 2011).

Karr et al. (2012) have create Mycoplasma genitalium

genotype. In this aspect, the phenotype includes, among many other features, the data over cell mass, volume and shape. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not observable by naked eye.

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

cell morphology. In Publicat

classification of microbial cells based on morphological features

is presented. Changes in the community structure can be fast and cause drastic effect on the reactor operation

producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to obs

allow quick response by operator if needed. Additionally,

ld be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in m es to be analyzed or to be

the images that are analyzed by automated

pictures. The software can not differentiate bacterial cells from similar, thus taking those into the imaging field

to proper density before microscopy, since too crowded images are more difficult to analyze. Moreover, the cells should locate in a o

6. Typical stages of image processing. The f

preprocessing is applied in order to minimize errors and artefacts caused by the image acquisition system.

ext step is segmentation, in which the ob

to gain the final results (Modified form Lehmussola, 2009).

an easy to used image analysis software

lts are validated. CellC can be applied to enumeration of and intensity

comparative analysis of images taken from dual stained samples. CellC ha s to analyze cell sizes from soil samples (Elazhairi

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009) qualify spore harvest purity (Harrold et al. 2011).

Karr et al. (2012) have created a whole

Mycoplasma genitalium. The goal of the model is to predict the phenotype from the phenotype includes, among many other features, the data over e. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

cell morphology. In Publicat

classification of microbial cells based on morphological features

Changes in the community structure can be fast and cause drastic effect on the reactor operation. The framework could be used

producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to obs

allow quick response by operator if needed. Additionally,

be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in m to be used for training a classifier.

the images that are analyzed by automated

pictures. The software can not differentiate bacterial cells from similar, thus taking those into the imaging field

to proper density before microscopy, since too crowded images are more difficult to analyze. Moreover, the cells should locate in a o

The first step is image acquisition with

errors and artefacts caused by the image acquisition system.

ext step is segmentation, in which the objects of interest are separated from the background.

to gain the final results (Modified form Lehmussola, 2009).

an easy to used image analysis software

ellC can be applied to enumeration of

and intensity from digital images. It can also be used for comparative analysis of images taken from dual stained samples. CellC ha

cell sizes from soil samples (Elazhairi

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009) qualify spore harvest purity (Harrold et al. 2011).

d a whole-cell computational model for small genome . The goal of the model is to predict the phenotype from the phenotype includes, among many other features, the data over e. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

cell morphology. In Publication II, a framework classification of microbial cells based on morphological features

Changes in the community structure can be fast and cause The framework could be used

producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to obs

allow quick response by operator if needed. Additionally,

be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in m used for training a classifier.

the images that are analyzed by automated method, such as CellC, pictures. The software can not differentiate bacterial cells from similar, thus taking those into the imaging field should be avoided

to proper density before microscopy, since too crowded images are more difficult to analyze. Moreover, the cells should locate in a o

irst step is image acquisition with

errors and artefacts caused by the image acquisition system.

jects of interest are separated from the background.

to gain the final results (Modified form Lehmussola, 2009).

an easy to used image analysis software, called

ellC can be applied to enumeration of

from digital images. It can also be used for comparative analysis of images taken from dual stained samples. CellC ha

cell sizes from soil samples (Elazhairi

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009)

cell computational model for small genome . The goal of the model is to predict the phenotype from the phenotype includes, among many other features, the data over e. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

ion II, a framework classification of microbial cells based on morphological features

Changes in the community structure can be fast and cause The framework could be used

producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to obs

allow quick response by operator if needed. Additionally,

be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in m

used for training a classifier. The most important factor of method, such as CellC,

pictures. The software can not differentiate bacterial cells from should be avoided

to proper density before microscopy, since too crowded images are more difficult to analyze. Moreover, the cells should locate in a o

irst step is image acquisition with a digital camera. Then errors and artefacts caused by the image acquisition system.

jects of interest are separated from the background.

to gain the final results (Modified form Lehmussola, 2009).

, called CellC, is introduced and ellC can be applied to enumeration of

from digital images. It can also be used for comparative analysis of images taken from dual stained samples. CellC ha

cell sizes from soil samples (Elazhairi

to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009)

cell computational model for small genome . The goal of the model is to predict the phenotype from the phenotype includes, among many other features, the data over e. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

ion II, a framework

classification of microbial cells based on morphological features, such as cell size

Changes in the community structure can be fast and cause The framework could be used, e.g.

producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to obs

allow quick response by operator if needed. Additionally,

be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

Automatic image analysis has many difficulties that should be kept in mind

The most important factor of method, such as CellC, is the

pictures. The software can not differentiate bacterial cells from other particles

should be avoided. Additionally, samples to proper density before microscopy, since too crowded images are more difficult to analyze. Moreover, the cells should locate in a one focus level. Therefore,

digital camera. Then errors and artefacts caused by the image acquisition system.

jects of interest are separated from the background.

to gain the final results (Modified form Lehmussola, 2009).

is introduced and ellC can be applied to enumeration of bacteria and

from digital images. It can also be used for comparative analysis of images taken from dual stained samples. CellC has been used cell sizes from soil samples (Elazhairi-Ali et al. 2013), to enumerate bacterial cell for enrichment cultures of seawater (Gray et al. 2009), and to

cell computational model for small genome . The goal of the model is to predict the phenotype from the phenotype includes, among many other features, the data over e. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not

Besides enumeration and quantification of the cell properties, image analysis can be u for further downstream analysis. One example is the division of the objects to different

ion II, a framework for automated , such as cell size, shape Changes in the community structure can be fast and cause e.g., for continuous producing bioreactor with mixed culture. The automatic analysis of the population structure from microscopy images is a rapid method to observe upcoming allow quick response by operator if needed. Additionally, the automatic be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

ind while taking The most important factor of is the quality of the other particles if both look . Additionally, samples to proper density before microscopy, since too crowded images are more ne focus level. Therefore,

digital camera. Then errors and artefacts caused by the image acquisition system.

jects of interest are separated from the background. The

is introduced and bacteria and to from digital images. It can also be used for been used Ali et al. 2013), and to

cell computational model for small genome . The goal of the model is to predict the phenotype from the phenotype includes, among many other features, the data over e. Since variation caused by a single mutation to these aspects might be small, image processing can aid in enumerating differences that are not

Besides enumeration and quantification of the cell properties, image analysis can be used for further downstream analysis. One example is the division of the objects to different for automated , shape Changes in the community structure can be fast and cause for continuous producing bioreactor with mixed culture. The automatic analysis of the erve upcoming the automatic be used to analyze dynamic changes and interspecies interactions in mixed bacterial cultures or cocultures to gain more understanding over bacterial behavior.

while taking The most important factor of quality of the if both look . Additionally, samples to proper density before microscopy, since too crowded images are more ne focus level. Therefore,

stained cells are often attached to filters to enable analysis at one focus level and to create higher threshold between object and background.

setting should be properly adjusted to avoid bac

training a classifier, it is important to take vast amount of pictures from each

2.3

The selection of cultivation method is a trade

amount of data. Experimentation with unknown cultures or new mutants is often with short batch experiments and sometimes

continuous cultures. In Figure presented.

Figure

provide large amount of information but less informative.

In order to screen large amount of bacterial production

exposure

chemochromic membranes The basic principle of the use of are grown

then chemochromic membrane production

membranes has been used in protein engi enhance

deletion mutants of

the color change is transient and solution to the

Schrader et al

stained cells are often attached to filters to enable analysis at one focus level and to create higher threshold between object and background.

setting should be properly adjusted to avoid bac

training a classifier, it is important to take vast amount of pictures from each

Cultivation method or reactor type

The selection of cultivation method is a trade

amount of data. Experimentation with unknown cultures or new mutants is often with short batch experiments and sometimes

continuous cultures. In Figure presented.

Figure 2.7. Alternative

provide large amount of information but less informative.

In order to screen large amount of bacterial production capabilities

exposure to random mutagenesis.

chemochromic membranes The basic principle of the use of are grown in well plates or then chemochromic membrane production can be

membranes has been used in protein engi hance the H2

deletion mutants of

color change is transient and solution to these

Schrader et al. (2008).

stained cells are often attached to filters to enable analysis at one focus level and to create higher threshold between object and background.

setting should be properly adjusted to avoid bac

training a classifier, it is important to take vast amount of pictures from each

Cultivation method or reactor type

The selection of cultivation method is a trade

amount of data. Experimentation with unknown cultures or new mutants is often with short batch experiments and sometimes

continuous cultures. In Figure

Alternative experiment provide large amount of information but less informative.

In order to screen large amount of bacterial capabilities are

to random mutagenesis.

chemochromic membranes.

The basic principle of the use of well plates or then chemochromic membrane

can be seen as bluish spots on the film (Seibert membranes has been used in protein engi

production of deletion mutants ofE. coli

color change is transient and se problems another high

(2008). The

stained cells are often attached to filters to enable analysis at one focus level and to create higher threshold between object and background.

setting should be properly adjusted to avoid bac

training a classifier, it is important to take vast amount of pictures from each

Cultivation method or reactor type

The selection of cultivation method is a trade

amount of data. Experimentation with unknown cultures or new mutants is often with short batch experiments and sometimes

continuous cultures. In Figure 2.7, some

experimental setups. Continuous cocultures are time consuming experimentations, but

experimental setups. Continuous cocultures are time consuming experimentations, but