Menu

Budidng Yeast data can be downloaded here, which includes 120 images. The corresponding project file can be downloaded here.

Step 1: Preprocessing

Action: The images are first loaded into CellTracer and then zoomed in to focus mainly on cell clusters. It is then cropped using the "Crop images to current view" tool in "Edit" menu.
Purpose: Cropping will generally reduce the size of input images thus reduce the image analysis processing time.
Paramenters: None.
Result:
All cropped images can be found here. As an example, the preprocessed result for image 120 is showing bellow.
Preprocessed Image #120

Step 2: Background screening

Action: Applying the Range Filtering algorithm to identify the background region.
Purpose: To identify an initial backgroung region mask.
Paramenters:
  • Maximum Half Cell Width: 10
  • Background Intensity Sppread: 20
  • Structure Element Radius: 10
  • Greedy: 5
  • Fill Holes: 1
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 3: Identifying Border Regions

Action: Applying the Minimum Ranking algorithm to identify the cell border regions.
Purpose: To identify the bright regions that surround cells
Paramenters:
  • Minimum Ranking Threshold: 0.3
  • Minimum Border Volume: 100
  • Upper Ranking Threshold: auto
  • Maximum Cell Half Width: 35
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 4: Identifying Border Regions(again)

Action: Applying the Rank & Count Transform algorithm to further identify the cell border regions.
Paramenters:
  • Minimum Ranking Threshold: 0.75
  • Maximum Cell Half Width: 35
  • Ranking Method: 1
  • Count Transform Threshold: auto
  • Structure Element Radius: 5
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 5: Identifying Cells

Action: Applying the Convex Model algorithm to identify cells.
Paramenters:
  • Minimum Cell Score: 1.0
  • Minimum Cell Volume: 200
  • Struct Element Radius: 5
  • Smoothing Parameter: 0
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 6: Global Alignment

Action: Applying the Global Alignment algorithm to algin images before tracking.
Paramenters:
  • Maximum Shift: 10
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 7: Cell Tracking

Action: Applying the Cell Tracking algorithm.
Paramenters:
  • Neighboorhood Size: 20
  • Minimum Overlapping Score: 0.1
  • Maximum Cell Displacement: 10
  • Neighboorhood Scale Factor: 1
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot
E. coli bacteria data can be downloaded here, which includes 25 images.The corresponding project file can be downloaded here

Step 1: Preprocessing

Action: The images are first loaded into CellTracer and then zoomed in to focus mainly on cell clusters. It is then cropped using the "Crop images to current view" tool in "Edit" menu. A "Double Resolution" operation has also been performed to make bacteria cells larger.
Purpose: Cropping will generally reduce the size of input images thus reduce the image analysis processing time.
Paramenters: None.
Result:
All cropped images can be found here. As an example, the preprocessed result for image 120 is showing bellow.
Preprocessed Image #120

Step 2: Background screening

Action: Applying the Coarse Screening algorithm to identify the background region.
Purpose: To identify an initial backgroung region mask.
Paramenters:
  • Background Spread: 0.3
  • Structure Element Radius: 0
  • Padding Method: 0
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 3: Identifying Border Regions

Action: Applying the Thresholding & Smoothing algorithm to identify the cell border regions.
Purpose: To identify the bright regions that surround cells
Paramenters:
  • Maximum Cell Half Width: 25
  • Lower Ranking Threshold: 0.67
  • Upper Ranking Threshold: 0.90
  • Global Intensity Threshold: 130
  • Structure Element Radius: 5
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 4: Identifying Border Regions(again)

Action: Applying the Robust Voting algorithm to further identify the cell border regions.
Paramenters:
  • Minimum Ranking Threshold: 0.8
  • Minimum Border Volume: 20
  • Ranking Method: 1
  • Count Transform Threshold: 0.25
  • Maximum Cell Half Width: 25
  • Minmum Cell Half Width: 10
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 5: Identifying Cells

Action: Applying the Convex Model algorithm to identify cells.
Paramenters:
  • Minimum Cell Score: 2.0
  • Minimum Cell Volume: 500
  • Struct Element Radius: 3
  • Smoothing Parameter: 0
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 6: Global Alignment

Action: Applying the Global Alignment algorithm to algin images before tracking.
Paramenters:
  • Maximum Shift: 50
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 7: Cell Tracking

Action: Applying the Cell Tracking algorithm.
Paramenters:
  • Neighboorhood Size: 15
  • Minimum Overlapping Score: 0.1
  • Maximum Cell Displacement: 10
  • Neighboorhood Scale Factor: 2
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot
Human cell data can be downloaded here, which includes 249 images.The corresponding project file can be downloaded here

Step 1: Preprocessing

Action: No preprocessing step was performed on this dataset.
Result:
The last image from this dataset is showing bellow.
Human Cell Image #249

Step 2: Background screening

Action: No background Screening was performed on this dataset.

Step 3: Identifying Border Regions

Action: Applying the Thresholding & Smoothing algorithm to identify border regions.
Paramenters:
  • Minimum Cell Half Width: 20
  • Lower Ranking Threshold: 0.9
  • Upper Ranking Threshold: 1.0
  • Global Intensity Threshold: 240
  • Structure Element Radius: 3
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 4: Identifying Border Regions(again)

Action: Applying the Robust Voting algorithm to further identify the cell border regions.
Paramenters:
  • Minimum Ranking Threshold: 0.9
  • Minimum Border Volume: 50
  • Ranking Method: 1
  • Count Transform Threshold: 0.25
  • Maximum Cell Half Width: 20
  • Minmum Cell Half Width: 20
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 5: Identifying Cells

Action: Applying the Convex Model algorithm to identify cells.
Paramenters:
  • Minimum Cell Score: 0.5
  • Minimum Cell Volume: 200
  • Struct Element Radius: 5
  • Smoothing Parameter: 0
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot

Step 6: Global Alignment

Action: No Global Aligment was performed on this example.

Step 7: Cell Tracking

Action: Applying the Cell Tracking algorithm.
Paramenters:
  • Neighboorhood Size: 10
  • Minimum Overlapping Score: 0.1
  • Maximum Cell Displacement: 10
  • Neighboorhood Scale Factor: 1
Result:
The CellTracer snapshot file for this step can be found here. A screen shot for this step is shown below.
Background Screening Screenshot
Three examples are included here to demostrate image segmentation and tracking using CellTracer. All examples share similar workflows although different algorithms are used during cell segmentation. It is possible to have one workflow with different parameter settings to apply to all examples, but at the expense of more computing time.

Running Examples
  • Download the software and unzip it to a local folder.
  • Download the example and unzip it to a local folder
  • Start Matlab and type CellTracer in the Matlab command window. Make sure that the current working directory is set to the folder where CellTracer was installed.
  • Load all the images into CellTracer.
  • Choose your own algorithms and parameters or follow the individual example instructions to run analysis.

Checking results using CellTracer snapshot files
  • Download the software and the project file and unzip them.
  • Start CellTracer as described above and load the project file (ecoli.mat, for example) into CellTracer afterward. You will be prompted to correct the image folder path if necessary.
  • Download the snapshot file of interest, unzip it and then load it into CellTracer using "File-->Restore from snapshot".