Basic terms and common measurement parameters

(1) Pixel and gray scale

1. Pixels are the most basic unit that constitutes the image of a display, and are arranged in rectangular columns in rows and columns (X and Y). Any moment column corresponds to a point, and these points are called pixels. The more pixels on the screen per unit area, the clearer the image, that is, the higher the resolution, the finer and truer the displayed image. The more pixels in the digital image, the more grayscale quantization levels, the more accurate the results analyzed by the image analyzer. but. When the number of pixels reaches a certain amount, the storage and calculation of the image analyzer computer increase geometrically, which slows down the processing speed of the computer.

2. Gray scale refers to the degree of lightness (darkness) of each pixel of the image, that is, the quantization level of the pixel's image from black to white. The grayscale quantization levels of the microscopic image analysis system are all 2n, and the more common one is 28, that is, 256 quantization levels. The more quantized levels of gray, the higher the gray resolution of the image analysis system, and the richer the light and dark levels of the displayed image. The closer to the light and dark (dark and light) levels of the true color of the captured image. The size of the gray value is inversely proportional to the content of the substance. The smaller the gray value, the deeper the dyeing and the more the amount of the substance (0 represents black, 2n-1 represents white). Generally, the gray level that the human eye can distinguish is only 15-25, and the computer is stronger than the human eye in terms of gray resolution. Therefore, when there is no obvious difference when the human eye is observed with a microscope, the image analysis system may reflect Significant difference between the two.

(2) Color

According to the color of the image, it can be divided into black and white images and color images, and the latter can be further divided into pseudo color images and true color images. Black-and-white images have no color information. Pseudo-color images are only artificially set to different colors based on grayscale values ​​for further analysis. The true color image is decomposed into three primary color signals of red, green and blue by the image acquisition device, and each primary color signal has its gray level, and finally re-encoded by the computer to be restored to a color image. Therefore, the true color image contains more information than the first two.

(3) Optical density

Optical density, also known as absorbance, refers to the logarithm of the ratio of light intensity (Io and Ib) before and after the light passes through a substance, that is, log (Io / Ib). The optical density value is directly proportional to the substance content. The smaller the optical density value, the smaller the degree of light absorption, and the lower the substance content.

(4) Length

For linear tissue structures with irregular shapes, it is difficult to calculate the length by general methods, such as capillaries and nerve fibers in tissues, various membranous structures in cell ultrastructure, etc. In the past, words such as sparse or dense arrangement were often used to describe, The image analyzer can measure the length of various image perimeter lines. For example: on an electron micrograph of cervical epithelial cells in biopsy tissue, the length of the cell membrane in a certain range is measured to be 8128 pixels, and the length of the desmosome is 1229 pixels. The length of the desmosome is 15% of the total length of the cell membrane. It is observed that the desmosome value of cancer cells is much smaller than that of normal cells. Therefore, it can be suggested that one of the reasons for cancer cell metastasis may be related to the reduction of the cell connection between cancer cells.

(5) Area

In the observation of light microscope or electron microscope immunohistochemical specimens, they all involve the problem of changes in certain structural areas in the specimens. The image analyzer can measure the number of pixels per square micrometer under a certain magnification. The method is to draw the outline of the structure to be measured. How many pixels are included in this outline range can be displayed immediately, so that the area value can be calculated quickly . For example, in the ferritin method and colloidal gold immunoelectron microscopy technique, the above method can be used to count the reaction product particles per unit area, thereby performing comparison under various experimental conditions.