Modern microscopy techniques generate an enormous amount of data in the form of images. Manual analysis of these images produces biased results that are often not reproducible. To extract the numerical data from the images, a free and user-friendly software called ImageJ is available at the NIH website. In this interactive tutorial, students will get acquainted with the applications of ImageJ and learn to measure cell area from the images.

This article describes an activity for measuring cell area using ImageJ, an image processing and analysis software developed, maintained, and distributed by Wayne Rasband of the National Institutes of Health (NIH). This freely downloadable and user-friendly program has wide-scale applications in medical imaging, biological sciences, engineering, climatology, and environmental science (Abramoff et al., 2004; Collins, 2007).

Microscopy techniques allow biologists to acquire large amounts of data in the form of images. Manual analysis of these images is time-consuming, tedious, and often gives biased and irreproducible results. For example, a decrease in cell size by 10% or a twofold decrease in the intensity of a fluorescently labeled protein in an image is undetectable by the human eye, but image analysis using ImageJ can detect such subtle but biologically significant changes.

ImageJ is a flexible program in that it can run either as an online applet or as a downloadable application on operating systems such as Windows, Mac OS, Mac OS X, and Linux. It can display, edit, analyze, and process images in many formats, including TIFF, GIF, JPEG, BMP, DICOM, FITS, and "raw" (Rasband, 1997–2008). It measures distances and angles and converts pixel values into user- defined units. Its standard image-processing functions include contrast manipulation, sharpening, smoothing, edge detection, and median filtering. Its applications in biological studies include quantifying and comparing cellular and subcellular components, estimating DNA and protein concentrations, and analyzing processes like chemotaxis and cell migration. The wide range of applications of ImageJ is possible because of the availability of 400 different plug-ins (Collins, 2007). Plug-ins are short add-on programs that provide additional functions to the main program of ImageJ.

Learning Objectives

In this activity, students will (1) learn the applications and use of ImageJ, (2) learn to measure cell area from a bright-field image and compute data to reach conclusions, and (3) explore on their own the use of the cell-counter plug-in for ImageJ.

I have used this activity in an undergraduate-level cell and molecular biology lab. A week before the lab, I send image files and the activity steps as an e-mail attachment to my students. I also provide a link to the NIH website and ask them to download the ImageJ program. To engage the students, I begin the lab by showing bright-field or fluorescent images of treated and untreated cells side by side. I also show images that represent changes in successive stages of a process or phenomenon. After showing images, I pose the following questions to the students:

  • Are there any visual differences between these pairs of images or successive images?

  • How will you quantify, categorize, and compare these differences?

  • Why is it important to quantify these differences?

For measuring cell area, I use bright-field images of Dictyostelium discoideum (amoeba) taken at 20x and 100x magnifications. For calibrating D. discoideum images, I use images of a micrometer slide taken at the same magnifications.

I suggest using microscopic images of any kind of cells or tissues for this activity. You can download images from the website http://rsb.info.nih.gov/ij/images/ or access these images directly in ImageJ by 'File' → 'Open Samples'. Alternatively, bright-field and fluorescent images of a variety of cells can be downloaded from the website of the American Society for Cell Biology (http://cellimagelibrary.org/) or from published articles.

I have successfully implemented this activity in cell and molecular labs. Students find the ImageJ program intuitive, and the activity not only engages them but also enhances their skills and knowledge.

ImageJ Tutorial

  1. 1. Download ImageJ to your computer from the website http://rsbweb.nih.gov/ij/download.html.

  2. 2. ImageJ is not set to any scale, so the first step is to open the program and set the scale using the calibration scale image, which is an image of a micrometer slide taken at the same magnification as the image to be analyzed. 'File' → 'Open'. The scale needs to be set only once for all the images to be analyzed at the same magnification.

  3. 3. Select the straight line icon. Draw a line between two points, 0.1 mm to 0.2 mm (two consecutive big lines) = 100 µ­m (Figure 1).

  4. 4. Go to 'Analyze' → 'Set scale' and enter the values: 'Known distance' = 100. Unit of length = type 'micrometer'. Select 'Global' and click 'OK'.

  5. 5. Go to 'Analyze' → 'Set measurements' → select 'Area', 'Standard deviation', 'Minimum and maximum gray value' and 'mean gray value' and click 'OK'.

  6. 6. Open the image to be analyzed (Figure 2). If the image is fluorescent (colored), convert the image to gray-scale: 'Image' → 'Type' → '8-bit'.

  7. 7. Go to 'Process' → 'Binary' → 'Make binary'. This step is known as "automatic thresholding" or "image segmentation" (Figure 3). It creates a clear distinction between the objects of interest (cells) and the background by setting pixels for objects at a certain value so that they appear white and the background appears black.

  8. (Alternatively, you can do manual thresholding by 'Image' → 'Adjust' → 'Threshold'. Objects in the picture will appear red and the 'Threshold' window pops up; click 'Auto' then click 'Apply'. Objects will appear white. Close the window.)

  9. 8. Click the rectangle tool icon. With the rectangular selection tool, select the text (example, scale bar) and the cells (cells that are touching each other) that you do not want to be measured. Go to 'Edit' → 'Clear'.

  10. 9. Go to 'Analyze' → 'Analyze particles', Size µ­m2 = '50-infinity'. Set the size limit to an appropriate range so as to avoid measuring small dots and other things in the image that you know for sure are not the objects of interest. Leave blank for 'Circularity'. Show 'Outlines', Select Display → 'Results', 'Exclude on edges', 'Include holes', 'Summarize', 'Records starts', Add to 'ROI manager'. Click 'OK'.

  11. 10. 'Results' window pops up with the areas of cells measured. In 'Results' window, go to 'Results' → 'Options' and I/O option box pops up. In the I/O option box, type '.xlsx' in 'file extension for tables'. Select all four 'Results Table Options' and click 'OK'. Go to 'File' → 'Save as', you will be required to name the Excel file and save it in your folder.

  12. 11. If you want to know which cells were measured by ImageJ, merge the original image with the outline image ('Drawing of image...'). For this, invert the outline image by 'Edit' → 'Invert' (Figure 4).

  13. 12. Open the original image again and use the menu command 'Plugins' → 'RGB gray merge'. You can download the plug-in from http://rsb.info.nih.gov/ij/plugins/rgb-gray-merge.html. 'Gray- RGB Stack Merge' dialog box opens. 'Gray stack': select outline image ('Drawing of image...'), 'Red stack': select original image. Green stack: 'None', Blue stack: 'None'. Check 'Keep source stacks together'. Click 'OK'. A new window pops up with the original image in red. The outlined cells in this window are the cells measured by ImageJ (Figure 5).

Figure 1.

Micrometer slide at 100x magnification. The yellow line indicates 100 µ­m distance.

Figure 1.

Micrometer slide at 100x magnification. The yellow line indicates 100 µ­m distance.

Figure 2.

Dictyostelium discoideum at 100x magnification (DMRB, Leica Inc.).

Figure 2.

Dictyostelium discoideum at 100x magnification (DMRB, Leica Inc.).

Figure 3.

Automatic thresholding (step 7).

Figure 3.

Automatic thresholding (step 7).

Figure 4.

The cells measured by ImageJ (step 11) are outlined.

Figure 4.

The cells measured by ImageJ (step 11) are outlined.

Figure 5.

Cells that were measured by ImageJ (outlined) and the cells that were excluded from the measurement. The picture is obtained by merging the outlined image (Figure 4) and the original image (Figure 2) using a plug-in, RGB Gray merge (step 12).

Figure 5.

Cells that were measured by ImageJ (outlined) and the cells that were excluded from the measurement. The picture is obtained by merging the outlined image (Figure 4) and the original image (Figure 2) using a plug-in, RGB Gray merge (step 12).

Exercise

After completing the activity, the students should be able to answer the following questions:

  • What is the average area of the cells measured using ImageJ?

  • What is the range of cell size? If the range is very large, categorize the cells into different size groups and draw a histogram showing the number of cells in each group (Figure 6).

Figure 6.

Distribution of cell area for Dictyostelium cells at 100x.

Figure 6.

Distribution of cell area for Dictyostelium cells at 100x.

Explore Activity

In a given image, there are three different types of cells. Count the number of cells of each type using the cell-counter plug-in for ImageJ.

Acknowledgment

I thank Dr. Curt Anderson, Department of Biological Sciences, Idaho State University, for introducing me to ImageJ.

References

References
Abramoff
M.D.
Magelhaes
P.J.
Ram
S.J.
(
2004
).
Image processing with ImageJ
.
Biophotonics International
,
11
,
36
42
.
Collins
T.J.
(
2007
).
ImageJ for microscopy
.
BioTechniques
,
43
,
S25
S30
.
Rasband
W.S.
(
1997–2008
).
ImageJ. Bethesda, MD: U.S. National Institutes of Health. Available online at http://rsb.info.nih.gov/ij/
.