The Histogram Widget is the convenient location of all image contrast and display information. It is composed of the image histogram itself and various image controls.
Adjust Image Curve
The most intuitive way to interact with image contrast and display is through the Histogram. An Image Histogram graphs the tonal distribution of an image. The x axis represents pixel intensity, with 0 is the lowest possible value and 65535 being the highest possible value in a 16 bit image. The y axis represents the number of pixels at each value. The scale of the x axis can be set to linear, logarithmic or logicle. (Logicle provides the best visibility for low intensity datasets.)
While the histogram itself models the concrete tonal distribution in the original data, the histogram's Curve indicates tonal distribution in the display. The curve is drawn between the minimum value in the display and the maximum value. Adjusting these values is by far the most common way to manipulate image contrast. When the minimum value is raised, every pixel below the minimum value is displayed as 0. Likewise, when the maximum value is lowered, all pixels at or above the new maximum are displayed at full intensity. The curve describes the relationship between pixel distribution and display intensity.
You can adjust the minimum and maximum values displayed in your image by sliding the nodes on the end of the curve or by changing the "Min" and "Max" value at the bottom of the histogram.
The relationship between pixel distribution and display intensity does not have to be linear. The degree of non-linearity is measured as "gamma". To adjust gamma, grab one of the nodes on the curve and lift it up or down. Here is the effect of arcing a curve downwards:
And here is the effect of moving a curve upwards:
The Min value at the bottom of the histogram can function as a Threshold, but it is intended to adjust image contrast, not threshold images. If you are interested in seeing the results of hard thresholds, use tools at the top of the histogram, which are for the Data Range. Set a range by moving the sliders or by inputting Min and Max values. The "Highlight Range" option colors all of the pixels included in your data range red.
(If you feel the data range is not helpful, you can uncheck the "Data Range" on the upper right corner of the histogram widget to make it go away.)
Export Display Settings
Sometimes it is important to document the adjustments you have made to image contrast. To see all of your display settings, including the Min and Max of the curve, the Gamma, and the Data Range, go to the "Report" dropdown menu and select "Display Settings". The results will appear in the report.
To export the data ranges you set, go to File -> Export -> Export Image Threshold Settings. The result will appear as a text file named "ImageDisplaySettings" in the export folder. (To open your export folder, File -> Show in Finder/Explorer -> Export)
A list of channels for the image is shown when you open an image. Each is represented by a color displayed on the image and next to the channel name. A set of icons at the top of the list provides the options to apply display or color setting changes, grouping of channels, and quality analyses to the entire list of markers.
The “eye” icon indicates visibility of a channel. Turn on channels in the list by enabling the eyeball. This allows for any number of combinations of channels to be on simultaneously. It is important to have an apt number (~ 3-5) of channels displayed on the image at once, so as not to hinder your ability to distinguish between them. Too many colors overlapping on the same location will wash the display out to white. The eyeball icon has three states: “off”, “on”, and “effect” (i.e. blinking, 3D, fade in, fade out).
The color for a channel can be changed by clicking on the colored circle next to the channel name. This will open a color palette users can customize or select from a list. Simply choose the color and click OK.
Selecting the color button on the Channel List provides the user with options to color each channel using a select predetermined color scheme.
The user can choose between a 6, 12, 18, or 24 color palette, known biomarker preference colors, or an optimized color palette customized by QiTissue. Preset nuclei and cycle colorization is also available. Checking the checkbox at the top only applies the function to the channels that are toggled on for display.
The "heart" icon located at the top right indicates a set of favorites. QiTissue utilizes a favorites system so the user can quickly recall channel combinations, feature plots, sequencer timelines, and more.
In the channels list, this means you can create a custom panel of biomarkers to display (also known as perspectives), with any contrast or display settings saved to it. To favorite a current perspective, click the heart icon and select "Add favorite". All customized perspectives will appear in the favorites dropdown list, along with QiTissue's own preset perspectives.
Registration refers to the relative alignment of biomarker channels. You can find the widget in the view menu or with the icon on the top bar. Both automatic and manual registration are possible.
Select Auto Register for offset registration, or Register (fine adjustment) to include rotation or skew. Checking "Transform and create new dataset" will register your images and prompt you to select a location in your directory. QiTissue will generate ome.tif files for each channel in the new location, creating an entirely new dataset that is registered.
Coloring the nuclei channels with a rainbow scheme in the Channel List can be helpful when viewing auto-registration results.
This alignment can also be manually adjusted in Channel tab of the Properties widget. The default X and Y values are 0. Type in new values or drag the arrows in the box up and down to shift the active channel.
Do not change the registration of your first round. You do not need to register every channel in a round; just register channel one and then select "Copy manual nuclei registration to other channels" from the Cyclic IF dropdown menu (Expert mode only).
In 3D view, the tissue is interpreted as landscape. Heightmap Channel, Qi re-interprets pixel intensity as "height" in a 3D model. This provides a novel experience of a tissue and a distinctive view of a key marker. The View Navigator lets you explore tissue architecture as a whole and zoom in to cell level detail anywhere as needed.
A small overview of the entire image is indicated by a white rectangle in the current “Field Of View” (FOV). You can adjust the FOV in several ways: using a zoom slider, scroll bars, and keyboard shortcuts or gestures on a trackpad. In the zoom navigator window, you can always see a reference to what is happening compared to the full image. The rectangle in the zoom navigator display is also moveable to directly scroll to a region of interest.
Clicking "Fit Selected" will zoom into a segmented cell (only when selected) and fill the entire field of view. This is useful to precisely see the boundaries applied by the segmentation.
While in 3D mode, you will need additional controls to navigate (i.e. rotating the sample). On the top right of an image , the icons for zooming, scrolling and rotating will appear. Click on the icon, hold down the mouse button, and move to see the control effect. Clicking the “Reset view” button, or the camera icon will transition back to the full view.