JPS5975043A - Method and apparatus for measuring tissue by light - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention particularly relates to a tissue photometric method and apparatus for quantitatively determining blood oxygen saturation from photometric values periodically obtained from a single sample using a spectrophotometer equipped with a monochromator.

For oxygen supply within living tissues, the degree of oxygenation of moglohin to oxygen carriers within the tissue capillary network plays a decisive role. Reflectance spectroscopy makes it possible in principle to measure the instantaneous degree of oxygenation in the capillary tubes extending over the surface, based on the difference in the absorption properties of oxygenated and deoxygenated hemoglobin. It is already known that reflection spectroscopy as tissue photometry can be used in vivo. In particular, the use of a flexible 11'L optical fiber makes it possible to inspect small areas. At that time, is it an artifact due to the movement of the tissue surface, or is it possible to reduce the detection time by shortening the detection time?
I have to be able to do it.

A yO fiber photometer has been proposed that can determine the spectrum of a tissue sample as a function of wavelength. The degree of oxygenation of the tissue sample can be determined from the amplitude distribution of the individual spectra.

However, conventional measurements are mostly based on the constant J
・1! It's stuck at the price. The diameter is 1M according to the present invention.
A method and apparatus are provided that enable quantitative evaluation of sample properties. There are two.

This object is achieved according to the invention by the method according to claim 1 and the device according to claim 13. Advantageous embodiments of the method and the device according to the invention are defined in the claims 2 to 12 and 1, respectively.
Listed in Sections 4 to 22.

According to the invention, spectra that occur periodically during tissue photometry can be evaluated in a very simple manner.

In particular, precise quantitative measurement of the hemoglobin oxygenation degree of any sample becomes possible. Thereby, it was shown that this method can be used very easily for clinical testing.

Other advantages and details of the invention will become apparent from the following description of an embodiment according to the drawings.

In FIG. 1 it includes a light source 1 realized by a micro-fiber photometer, or a quinanon ultraviolet lamp. light #i
1, is guided to the tissue sample P via an optical stripe filter system 2 and an optical fiber 3 having a diameter of 70 μm as a light transmission fiber. The reflected light is applied to the wavelength selector through six fibers arranged in a ring around the light transmitting fiber and collectively designated by the reference numeral 4. The wavelength selector consists of an interference fin/disk 5 as a monochromator rotatable by a motor 6, in which, depending on the angle of rotation, various wavelengths between 495 ram and 5], 5 nm can be selected. Light is selected. The interference filter disk 5 passes light of various wavelengths at the measuring position M depending on the rotation angle, with 495,111 wavelengths passing between 0 and 18,000 rotation angles.
In ascending order of wavelength ranges between 11 and 61.5 nm,
Further, at a rotation angle of 180° to 36o00, the light passes in descending order. The intensity of the passing light is measured by a photomultiplier tube 7, evaluated and output in the form of an analog voltage signal. The voltage signal is preferably supplied to a digitally operated evaluation device 10 (described in more detail below).
reach. For each rotation of the monochromator 5 (.zeta.) one tri-force pulse is applied to the evaluation device.

It is possible to use a spectrometer that monochromates the incident light and then periodically scans the spectrum.

In order to control the digitization of the measurement signal as a function of wavelength, specific wavelengths can be marked by deriving separate tri-force pulses.

The configuration of the evaluation device 10 will be explained later with reference to FIG. 6, and first, the method of processing measurement signals will be explained with reference to FIGS. 2 to 5.

FIG. 2 shows that the measurement signal occurs periodically with the rotation of the interference filter disk 50 and that strong noise is superimposed on the measurement signal.

For subsequent signal processing, the signal must be separated into individual periods. In this case, difficulties arise because, on the one hand, the length of the period is an unknown quantity, and on the other hand, the lengths of the circumferences of the seeds, jz, 1υ117) are not the same. The latter is caused, for example, by fluctuations in the rotational speed of the interference filter disc 50.

In the method according to the invention, first a reference spectrum is determined. As a measure of the quality of the spectrum, inter alia the signal-to-noise ratio, the signal height or other signal properties can be used. By comparing the individual spectra using the frequency criterion, a suitable section of the pond for evaluation is found. For example, first the "reference period" is determined and the most similar section is found by correlation. By utilizing the fact that the interference filter disk 5 generates a periodic fundamental function regardless of the measurement signal, the period can also be found from its □□□ grand term or minimum.

Sections that are similar to each other are found from the measured spectrum using the above method J. It is true that the correlation length is determined by the length of the reference spectrum. However, different period lengths can be compensated for by overlapping correlation regions. For the latter method, j stiff filter 1-"
' The trigger pulse of the plate can be irrelevant.

In this way, it is possible to select individual spectra with a constant from among a large number of spectra using the criterion of goodness. However, these individual sounds may also contain noise, such as extra-white sounds.

Noise suppression: lr+J can generally be achieved by weighted averaging.

It has been shown that good results can be obtained by averaging periods of 5 to 1. For example, FIG. 3 shows two frequency spectra preprocessed in the manner described and averaged over 30 periods.

The first spectrum was taken from the oxygenated sample and the second spectrum was taken from the deoxygenated sample.

The preprocessed signal is periodic and...! Since the i′j domain is limited, it can be transformed into frequency space by Fourier transform. Through this conversion, the wavelength function is expressed as a frequency diagram, and it is expressed as
As shown in the figure.

The interference filter disk 5 is particularly effective at high rotational speeds.
As can be seen, a nearly symmetrical signal distribution over the wavelength range results. Signal processing speed is increased by measuring only one half cycle and then obtaining a mirror image. High rotation speed makes sense! Logical information loss can be ignored.

Changes in hemoglobin oxygenation result in pronounced changes in the corresponding amplitude-frequency diagram, with oxygenation causing an increase in amplitude.

Although small changes in oxygenation do result in small changes with little or no discernible change in the f-word course, they do make a clear difference in frequency space, and this river front is exploited for subsequent evaluation. .

Figure 5 shows various trials starting from the hydrogenated sample, passing through intermediate steps, and ending with the deoxidized sample.
] Shown are a series of spectra obtained. As a baramere] 00 seconds time is entered and J5, oxygenation decreases with time. Such a spectrum is obtained, for example, during the tack of a pre-oxygenated sample (1-glycan) and this transformation process is typical.

The following parameters are determined for the spectrum 4).

q) m(1): Sample with unknown oxygenation (equivalent to H).

1]) II (t): Most jj! This corresponds to sample I, which has a spectrum containing 8 g of 17g acid.

c) t(1)'i moi; insect <[1 brother oxygen "Hisare"
This corresponds to a sample with 2 nispect.

The proportion of It(t) involved in In(t) becomes clear based on the following one-eye relationship.

m (t) = α・h (t) 1β・t(t)
(1) The coefficient α is a measure of the proportion of the spectrum that is deoxygenated and β is a measure of the proportion that is deoxygenated.

At the same time, the coefficient C (is the case where equation (1) is completely oxygenated!
Spectra 1 and 2 are of completely deoxygenated samples.
・It can be found to be correlated with Tor.

That (deyo・tsute, tissue oxygen treasure from photometric values; l
), target 1t [find the possibility given.

The measurement results showed that the proportion of high frequency υ in the measured signal changes in the form of a characteristic 1', especially with oxygen f. That is, by preparing a calibration curve in advance, the degree of oxygenation of an unknown sample can be immediately determined from this characteristic.

In FIG. 6, A means an analog signal from the photomultiplier tube 7. This signal is band-limited using filter]1, and then A-
It is converted into a digital signal at a sampling frequency of 1 by the D converter 12 and then provided to the memory 15 for temporary storage purposes. Via the trigger signal generator 1:3, pulses can be derived from the rotation of the mechanical disk 5, which indicate the spectral timing nozzle and serve as starting and reset pulses. Teiji6ru version is quiminge f
Advantageously, it is regulated in accordance with the wavelength (JJ obtained in each case from the interference filter disk 5 of FIG. 1 by t14).

As mentioned above, the standard of goodness given in advance is OK. -! Select a suitable one from among the recorded spectra. In this case, Tricagusus plays the role of selecting the reference spectrum for U, Umbi 1z]. The reference period °° is compared with the spectrum stored in the memory 15 at -11, +f.

Thereby, a number of -'j spectra are determined by comparison from a large number of stored measurements and these are fed together to the device 20 for layer roy suppression.

Selective addition and one'
An F-equalization is performed, which is equivalent to spectral filtering, so that the signal is almost noise-free so that it can be applied to the device 25 for Fourier transformation.

It is advantageous to perform distortion correction of the spectrum before Fourier transformation. Equipment provided for this purpose ti 2
2 also plays the role of forming the above-mentioned signal reading image.

From the Fourier-transformed spectrum, a signal selector 26 as a maximum single-direction extractor selects a frequency or a frequency range ]H] having a Q′jj characteristic indicating the degree of oxygenation.

Such a feature is the high frequency proportion as mentioned above. For quantitative detection, the amplitude at the maximum point can be used directly, or for more precise l'1IIJ determination, the sum of the amplitudes of the frequencies above and the amplitudes of the frequencies below a given minimum value can be used. The ratio to the sum of amplitudes may be determined.1. In FIG. 11, the area of each ini is shown by hatching.

A calibration device 27 is attached to the signal selector 26 for obtaining the signal characteristics. In the analyzer 28, the degree of oxygenation of the unknown sample is determined by comparing the measured values with a given characteristic curve. , on the indicating device 29' in digital or anatomical form.Additionally, the take i, lf, is stored in a format suitable for processing by the machine (8a).

The theory is explained below! The ILL method and apparatus are used for general tissue side-lighting, or to determine the blood oxygen saturation in the tissue to the intestine.
'Fj-like detection. The method according to the invention can also be applied when the hemoglobin concentration is variable.
f-! '). Same as hemoglobin test...
Due to the force θ, other absorptions and fluorescent dyes are also
The relationship between the frequency and the frequency is detected by the tissue side light. In this way, >jlI, i・1・valence, which was previously impossible, becomes possible.

[Brief explanation of drawings]

Figure 1 is a schematic configuration diagram of a photometer with spectrometer A1, Figure 2 is the original signal curve for tissue photometry measured by spectrometer η1, and Figure 3 is a diagram of the sample exposed to oxygen and a deoxidized sample. Figure 11 shows the results of Figure 3 in frequency space; Figure 5 shows the results obtained at various stages of the deoxygenation process in frequency space; Figure shown, No. 6
The figure is a block circuit diagram of a second apparatus for carrying out the method illustrated in FIGS. 2-5. 1...Light source, 2...Filter system, :3. d...Optical fiber,...Interference filter disc, 6...Motor, 7
...Photomultiplier tube, 10...Evaluation device,]]...
Filter, 12... A-D converter, 13... Tri-force signal generator, 14... Timing device, 15...
Memory, 10... Comparison device, 20... Noise suppression postfix, 22... Distortion correction device, 25... Fourier transform device t, 2 (i... Belief selector, 27... Calibration Device, 28... Analyzer, 20... Indicator, A...
・Photomultiplier tube analog signal, O]〈...standard of quality,
P...Tissue sample.

Claims (1)

[Scope of Claims] 1) In a tissue photometry method for quantitatively determining blood oxygen saturation from photometry values periodically obtained from one sample using □ spectroscopy δ1 using a monochromator. ;+) a first spectrum is extracted from the periodic signal sequence; 1)) a given number of further spectra are selected based on a selectable criterion of goodness; and C) a tissue characterized in that a Fourier transform of the spectrum is carried out, and d) characteristic values of the sample, in particular blood oxygen saturation, are determined by comparison with a given calibration curve on the basis of signal features in the Fourier transform. Photometry method. 2) The method according to claim 1, wherein one first spectrum is produced in the step a). 3) The method according to claim 1, wherein the quality criterion in step b) includes a comparison of spectral shapes. 4) A method according to claim 3, characterized in that the shape comparison is performed after digitization of the spectra. 5) Claim 4, characterized in that digitization is suppressed in relation to the wavelength by a monochromator.
The method described in section. 6) The method according to item 1 of the scope of Patent No. 8 Seigyu, characterized in that the measurement signal is subjected to noise suppression. 7) A method according to claim 6, characterized in that the noise suppression is carried out by selective addition and subsequent averaging. 8) Claim 9) characterized in that between 15 and 60 spectra, preferably 30 spectra, are selected for the selective force calculation 9) In the process of d) identified as signal features 2. A method as claimed in claim 1, characterized in that the amplitude of a frequency or frequency range is extracted. 10) The method according to claim 9, characterized in that the maximum amplitude is extracted. 1]) The ratio of the sum of amplitudes of frequencies above a given signal minimum value to the sum of amplitudes of frequencies below a given signal minimum value is used as a signal feature. the method of. 12) A method according to claim 1, characterized in that: f positive curves are determined by correlation with an oxygenated sample and a deoxygenated sample. 13) A tissue photometry method for quantitatively determining blood oxygen saturation from a single photometry series obtained periodically from a single sample using a spectrophotometer equipped with a monochromator, which comprises a) the spectrum of J is extracted; b) a given number of further spectra are picked based on a selectable goodness criterion; C) a Fourier transform of the spectrum is performed into frequency space; and d) within the Fourier spectrum. In an apparatus for carrying out tissue photometric methods in which the specific characteristics of the sample, in particular the blood oxygen saturation, are determined by comparison with a given calibration curve on the basis of signal characteristics of
a selection device (11-16), a device for Fourier transformation (25) and a signal selection eW (26) for comparing the selected features with a given calibration curve. Tissue photometry device. 4) Device according to claim 13, characterized in that the selection device (11-16) has means for selecting the measured spectrum on the basis of a given quality criterion. 15) Device according to claim 13, characterized in that a device (2C,) for noise suppression is additionally provided. 16) Device according to claim 15, characterized in that the device (20) for noise suppression is an average value forming device. 17) The bale apparatus according to claim 13, characterized in that a temporary storage memory (15) is provided for later processing of the obtained spectrum. 18) The device according to claim 13, characterized in that the device (27) for determining the calibration curve by calculation for the belief selection g"i (26) (26) belongs to the month. 19 14.) A device (22) for mirroring and/or distortion correction of the measured spectrum is inserted before the device (25) for Fourier transformation. Apparatus. 20) The apparatus according to Claims@Claim 13, characterized in that a digital computer is provided for evaluation. 21) The apparatus is characterized in that a microprocessor with a memory of its own is used as the digital computer. The apparatus according to claim 20.