|
A concept of the new type of biological morphometry for the cellular and tissue levels of organization has been developed, which makes it possible to morphometrize not an optically detectable form (morphology, habitus) or other related tissue parameters, but the patterns of its response to the external stimuli / field recorded by a chip on a CMOS/CCD array, or the patterns of its own non-optical characteristics (for example, the distribution of magnetic properties and isotopy in radioautography on a chip, or a radioisotope label), forming a compartment or compartmentalized (in vesicular or liposomal compartments) pool of a substance/label, characterized by a physical property: a descriptor, the contours of which are morphometrized by the chip due to the multilevel signal conversion technique. Using multilevel signal conversion techniques in sandwich chips with multiple converters, one can achieve not only mapping of the individual characteristics, the sample descriptors, but also establish colocalization of these descriptors/characteristics over the sample area or in the microfluidic channels, that is, to obtain a correlation map of the sample properties on a chip. Using the film converters of various physical signals and sample properties into an optical signal/response in different spectral ranges on the chips with three spectrozonal channels (either with Bayer mosaic filters or their analogs with other spectrocolorimetric characteristics, or with the unified pixels capable of capturing information from all the three channels in the case of APC-S sensors, or with multilayer implementation of the spectrozonal color separation into the additive channels as in the FOVEON X3 arrays), it is possible to colocalize at least three sample parameters on a chip if each of these parameters is converted into an optical signal that strictly corresponds to the parameters of the chromatic coordinates of a particular filter (or a photodiode layer of different depths in the case of X3). In the latter case, one-to-one colocalization/mapping can be established, in which all the parameters are fixed at each point, and pixels or “sensels” of different descriptors are superimposed, which is unattainable in the case of Bayer mosaic filters and their analogues. Being not just an image, but a resolved distribution map of the certain sample properties converted into an optical signal, the sample registration file itself (a bitmap without a discrete cosine transforms simultaneously a file containing information about the descriptors characterizing the sample with position sensitivity. For each pixel of the image, a descriptor can be determined from the codes (image parameters) for a given conversion of a non-optical signal into an optical one, characterizing a compartmentalized sample.
|
