The Human Eye
Ever wonder why we have so much difficulty seeing landscape details in white-out and foggy conditions? It’s because the human vision system is designed to see color much more effectively than it is to see grayscale. Humans are reportedly able to detect up to 10 million unique colors … but we can distinguish only 720 shades of gray.
Our perception of grayscale becomes even less reliable with the addition or subtraction of background contrast. The interactive image below can be used to illustrate that principle. The horizontal gray bar below appears to be the same shade from left to right when the scroll bar is moved all the way to the right. Try this yourself and move the cursor all the way to the right, then move it back to the center and left and see how your perception of the gray bar changes.
While the gray bar appears to continuously change its shade, or density, as you move the panel back and forth, in actuality, all you are doing is continuously changing the shade of the background, thereby changing the level of contrast between the foreground and the background. The gray bar never changes shade, but we interpret it differently based upon the background.
These two limitations, along with others associated with display monitors add complexity into interperting what is and is not seen accurately in grayscale images such as X-rays, magnetic resonance imaging (MRI), computed tomography (CT), ultrasound, and other digital imaging modalities.
Eliminating the issues to see beyond the capabilities of the human vision system allows for more accurate diagnosis of images.
(drag the cursor to the right and left in the image below)
Imago’s TN-1 will allow us to
See Beyond the Limitations of the Human Eye
What TN-1 Does Differently
Imago’s TN-1 technology features a unique methodology named Local Micro-Contrast Convergence (LMCC). This digital imaging processing technique employs a sequence of transformations that cause relationships among neighboring pixel groups in an image to aggregate into predictable color and luminosity patterns that are consistent with the structure of whatever material is the subject of the image (human tissue, other animal tissue, plant tissue, or any other substance within an image), as well as consistent with the imaging modality that generated the image. As a result, each material is uniquely characterized and can be visually differentiated, using color and grayscale levels that are easily within the range of the human vision system.
Different algorithmic sequences may be created and applied to optimize the characterization of distinct material properties through multiple visualizations of a single image, each visualization highlighting a different characteristic, such as object boundaries (margins), textures, fine structures, core composition, symmetry, and changes within objects. In breast imaging, for example, such clinically or diagnostically important “objects” would include benign and malignant tissue abnormalities, along with their boundaries, morphology, and textures.
TN-1 currently consists of a suite of 9 different algorithms, all which provide new visualizations with the intention of being used for different diagnostic purposes. Some of the TN-1 algorithms are shown below.
Making the Invisible – Visible
(move mouse over pictures for larger views)
Imago’s TN-1 can be directly integrated as a series of algorithms through a functional application program interface (API) directly into radiological or other imaging systems. Alternatively, TN-1 has been developed as a standalone product and can batch-process images in sub-second time per visualizations fully integrated into the radiology workflow RISC standards environment. A cloud-based version is available now for product demonstrations and ongoing Company research and development.