JP2017211285A - Analysis data processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the accuracy of analysis by reducing a phenomenon that when a large peak appears in a detection signal obtained by a TOFMS detector, a baseline sinks immediately after the peak.SOLUTION: An averaging processing unit 172 performs an averaging process using an IIR filer having a small time constant on the basis inputted data and calculates a baseline correction value, and a subtraction unit 173 subtracts the correction value from the original data and thereby obtains data having had subduction corrected. Since a new averaging process is executed only when it is determined by a data selection unit 74 that the inputted data value falls within a prescribed range where the magnitude, etc., of subduction is taken into account for an immediate baseline correction value, it is possible to appropriately correct subduction of a relatively large temporal change without being affected by an intrinsic peak.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an analytical data processing apparatus that processes data acquired by an analytical apparatus, and more specifically, to correct deformation and distortion of a peak waveform appearing in time-series data obtained by a time-of-flight mass spectrometer. The present invention relates to an analysis data processing apparatus that performs the above process.

  In an analyzer such as a mass spectrometer, an analog signal obtained by performing analysis is usually converted into digital data by an analog-to-digital converter (ADC), and then stored in a storage device or data processed. A process of transferring to a computer (PC) is performed.

FIG. 5 is a schematic configuration diagram of circuit blocks after an ion detector in a conventional time-of-flight mass spectrometer (TOFMS).
The ion detector 11 built in the TOFMS main body 1 detects ions separated according to the mass-to-charge ratio by a time-of-flight mass analyzer (not shown), and outputs an analog detection signal having an intensity value according to the number of ions. Output every moment. This detection signal is amplified by an amplifier 12, and then sampled at a predetermined sampling frequency in an analog-digital converter (ADC) 13, and converted into digital data for each sample. The ADC value capturing unit 14 functions as a buffer that sequentially captures and temporarily stores the data values output from the ADC 13.

  The TOFMS main body 1 can obtain time-of-flight spectrum data indicating the relationship between the time of flight and the ion intensity value in the time-of-flight range corresponding to a predetermined mass-to-charge ratio range by one measurement. However, when performing ionization by, for example, matrix-assisted laser desorption ionization (MALDI) method, the amount of ions generated by irradiating a sample with laser light is often small, and the ion amount for each laser light irradiation is small. The variation is also large. Therefore, in order to increase measurement sensitivity, multiple measurements are performed on the same sample, and the time-of-flight spectrum with high peak intensity is obtained by integrating the time-of-flight spectrum data obtained in each measurement over the multiple measurements. I try to get data. The integration processing unit 15 performs integration of such spectrum data. The baseline correction unit 16 corrects the baseline in the signal waveform of the spectrum data after the integration process. The data after the baseline correction is transferred to the data processing PC 3 through the communication path 2 and temporarily stored in, for example, a storage device mounted on the data processing PC 3.

  When ions enter the ion detector 11 at a certain time, a peak as shown in FIG. 6A appears in the analog signal waveform input to the ADC 13. When the number of ions incident on the ion detector 11 almost simultaneously is large, the peak appearing in the analog signal waveform becomes large as shown in FIG. At this time, a phenomenon may occur in which the baseline sinks immediately after the peak. This phenomenon is mainly caused by factors such as stray capacitance of a cable that sends a signal from the output end of the ion detector 11 to the circuit board on which the amplifier 12 is mounted. The baseline once submerged gradually returns to the original level, but when the next peak occurs while the baseline is subducting as shown in FIG. 6C, the peak is smaller than the actual peak. turn into. Therefore, such baseline fluctuations will lead to a decrease in analytical accuracy.

  As shown in FIG. 5, the baseline correction unit 16 is provided in the conventional apparatus, but this baseline correction unit 16 is mainly intended to remove the offset of the signal. For this reason, the baseline correction unit 16 performs an operation of subtracting a known fixed value or a value obtained by averaging signal levels in the vicinity of the baseline over a long period from the original data value as an offset, thereby obtaining a zero value for the baseline. It is corrected so that it is close. Such correction processing by the baseline correction unit 16 cannot appropriately correct the baseline fluctuation that converges in a short time such as the sinking of the baseline immediately after the peak as shown in FIG.

  On the other hand, Patent Document 1 discloses a data processing technique for reducing ringing associated with a peak as peak waveform processing different from the baseline correction as described above. However, the baseline sinking phenomenon immediately after the peak due to charge / discharge of the stray capacitance is different from the ringing caused by reflection from the transmission line end or the like. It is difficult to correct enough.

International Publication No. 2012/039061

  The present invention has been made to solve the above-mentioned problems, and the object of the present invention is to suppress the sinking of the baseline immediately after the peak as described above, so that the peak is continuously generated. Even so, it is an object of the present invention to provide an analytical data processing apparatus capable of maintaining the exact height of the peak and realizing high analytical accuracy.

The present invention made to solve the above problems is an analytical data processing device for processing time-series data obtained by digitizing an intensity signal in which a peak waveform obtained by an analytical device appears, and comprising a peak waveform To suppress baseline fluctuations immediately after the appearance,
a) a data selection unit that selects data whose data value is within a predetermined range near the baseline among sequentially input data;
b) an averaging processing unit that performs an averaging process on a plurality of data in a time-series direction selected by the data selection unit;
c) a subtraction unit that subtracts the value obtained by the averaging process from the input data;
The lower limit of the predetermined range in the data selection unit is defined so as to cover the size of the assumed sinking of the baseline immediately after the appearance of the peak waveform, and the averaging processing period by the averaging processing unit is It is characterized in that it is determined according to the expected sinking time.

  The analysis data processing apparatus according to the present invention can be used, for example, to process data obtained by digitizing a signal reflecting the relationship between the flight time obtained by measurement in TOFMS and an ion intensity signal.

  In the analysis data processing apparatus according to the present invention, the data selection unit determines whether or not the value of sequentially input data is within a predetermined range near the baseline, and averages only the data whose values are within the predetermined range Select as data to be used for processing. The lower limit of the predetermined range is determined so as to cover the assumed subsidence size of the baseline immediately after the peak waveform appears, that is, data indicating the assumed subsidence is used for the averaging process. On the other hand, the upper limit of the predetermined range may be determined so as to cover, for example, when noise is superimposed on the baseline (that is, an averaging process is performed). In other words, since the data corresponding to the obvious peak is excluded from the averaging process, the baseline correction value obtained by the averaging process is not affected by the size of the peak.

  The averaging processing unit performs averaging processing on the plurality of data selected as described above, and the subtraction unit corrects the input data value by subtracting the value obtained by the averaging processing from the input data value. To do. In general, in the baseline correction for the purpose of the signal offset correction as described above, even when the input data values are averaged, the period is set to be quite long so as not to follow the fluctuation of the input data values in a short time. . On the other hand, in the analytical data processing apparatus according to the present invention, although the temporal fluctuation is gentle compared with the peak, the period of the averaging process is conventional so as to follow the sinking of the baseline that fluctuates in a relatively short time. It is set to be considerably shorter than the averaging period in the baseline correction. As a result, the sinking of the baseline that occurs immediately after a large peak can be reduced and the baseline can be brought closer to a flat surface.

  The size and time of the sinking of the baseline depends on the configuration of the apparatus, for example, the type and length of the cable that sends a signal from the output end of the ion detector 11 to the circuit board on which the amplifier 12 is mounted in FIG. To do. Therefore, the predetermined range in the data selection unit and the period of the averaging process in the averaging processing unit can be determined in advance based on, for example, results obtained experimentally by the device manufacturer.

  In the analysis data processing apparatus according to the present invention, preferably, the averaging process in the averaging processing unit is performed using an IIR filter. As is well known, the IIR filter differs from the FIR filter in that the past output value is related to the calculation of the current or latest output value. Therefore, in general, the IIR filter may have a problem of stability, but has an advantage that a steep cutoff characteristic can be obtained with a short order. Therefore, by using the IIR filter, the sinking of the baseline can be sufficiently reduced with a small circuit scale. As described above, in the IIR filter, since the past output value is used for the calculation of the latest output value, the calculation is substantially based on a plurality of input data.

  In the analysis data processing apparatus according to the present invention, more preferably, when the baseline correction value to be subtracted is obtained by averaging processing using an IIR filter, the latest baseline correction value is multiplied by m (where m> 1). It is preferable that a new baseline correction value is obtained by adding the newly input data value and performing the averaging process and then multiplying the value by 1 / m. According to this configuration, information loss during execution of the averaging process is reduced, so that the baseline sinking can be corrected with higher accuracy.

  The analysis data processing apparatus according to the present invention can reduce the sinking of the baseline immediately after a large peak appears. As a result, even if another peak appears immediately after a large peak, it is possible to avoid a decrease in the signal intensity of the rear peak due to the sinking of the baseline, and an analysis value based on the height or area of the peak (For example, quantitative values corresponding to the component amount and component concentration in the sample) can be obtained with high accuracy. Thus, the analysis accuracy of the analyzer can be improved by using the analysis data processing apparatus according to the present invention. In addition, by using an IIR filter for the processing for reducing the sinking of the baseline, the circuit scale can be suppressed, and the analysis accuracy can be improved while suppressing an increase in cost.

The schematic block diagram of the circuit block after the ion detector in TOFMS using the analytical data processing apparatus by one Example of this invention. FIG. 2 is a schematic block configuration diagram of a baseline subtraction correction unit in FIG. 1. FIG. 2 is a detailed block configuration diagram of a baseline subtraction correction unit in FIG. 1. The wave form diagram which shows the effect of the baseline sinking correction in the analysis data processing apparatus of a present Example. The schematic block diagram of the circuit block after the ion detector in general TOFMS. Explanatory drawing of the baseline sinking phenomenon immediately after the peak.

Hereinafter, an embodiment of an analysis data processing apparatus according to the present invention will be described with reference to the accompanying drawings.
1 is a schematic configuration diagram of a circuit block after an ion detector in TOFMS using the analytical data processing apparatus according to the present embodiment, FIG. 2 is a schematic configuration diagram of a baseline subsidence correction unit in FIG. 1, and FIG. 1 is a detailed block configuration diagram of a baseline subtraction correction unit in FIG. In FIG. 1, the same components as those described in FIG.

  In the analysis data processing apparatus of this embodiment, the configuration from the output stage of the ion detector 11 to the baseline correction unit 16 is exactly the same as that in FIG. 5 is different from FIG.

  The baseline subtraction correction unit 17 is sequentially inputted with the input data corrected by the offset of the original detection signal and the long-time drift in the baseline correction unit 16 (according to the time-series order in which the data was obtained). The The baseline subtraction correction unit 17 performs an averaging process based on the input data to generate baseline correction data that reflects a relatively short-time baseline change such as a baseline subduction immediately after the peak. An averaging processing unit 172 to be generated, a subtraction unit 173 that outputs corrected data in which the sinking of the baseline is corrected by subtracting the baseline correction data from the input data, and a predetermined value near the baseline A data selection unit 174 that determines whether or not the signal value falls within the range, and selects only data that falls within that range as data that can be used for the averaging process, and the input data is required for processing by the averaging processing unit 172 A delay processing unit 171 that delays by time.

  The averaging processing unit 172 follows a relatively sudden change in the baseline such as the sinking of the baseline immediately after the peak, and reduces the time constant so that high frequency noise can be sufficiently removed (that is, the calculation cycle is reduced). Keep it short. As described above, the main cause of the sinking of the baseline is charging / discharging of the stray capacitance between the output stage of the ion detector 11 and the input stage of the amplifier 12, and therefore the output stage of the ion detector 11 and the amplifier 12. The size and length (time) of subsidence differ depending on the length and type (structure, material, etc.) of signal lines (cables, etc.) connecting the input stage. If the configuration of the device is determined, the length and type of such signal lines are almost determined, and for example, the manufacturer of this device can determine an appropriate time constant (or calculation cycle) based on the experimental results of the actual device. .

  On the other hand, if the time constant in the averaging processing unit 172 is reduced, the signal fluctuation due to the original peak may follow, but in order to avoid this, the data corresponding to the signal fluctuation due to the original peak is averaged. The baseline correction data, which is the output of the averaging processing unit 172, is kept constant during a period in which such data is input. In order to exclude the data corresponding to the signal fluctuation due to the original peak, the data selection unit 174 determines whether or not the value of the input data is within a predetermined range with respect to the baseline correction data at that time. However, the data corresponding to the subsidence and the high-frequency noise superimposed on the baseline need to be within a predetermined range. Therefore, the lower limit of the predetermined range is determined in consideration of the magnitude of the sinking of the baseline, while the upper limit of the predetermined range is determined in consideration of the level of high frequency noise.

As an example, the signal range of data selected by the data selection unit 174 as data used for the averaging process can be determined as follows.
(1) The upper limit value U is set to a value 4LSB × √ (number of integrations) larger than the baseline correction data value corresponding to the latest baseline.
(2) A value lower than the baseline correction data value corresponding to the latest baseline by 8 LSB × √ (number of integrations) is set as the lower limit L.

  Here, LSB is the least significant bit of the digital value converted by the ADC 13, and the number of integrations is the original spectrum data for obtaining one mass spectrum data in the integration processing unit 15 (predetermined obtained by one measurement in TOFMS). This is the number of times of integration of the time-of-flight range, that is, data over a predetermined mass-to-charge ratio range. Therefore, for example, when the number of integration is 16, the upper limit value U is 16 LSB and the lower limit value L is 32 LSB. Here, the allowable range of 4 LSB for determining the upper limit value U is a value mainly considering the noise level of the ADC 13, and the allowable range of 8 LSB for determining the lower limit value L is a value mainly considering the magnitude of the sinking of the baseline immediately after the peak. The numerical values such as 4LSB, 8LSB, and the number of integration can be easily changed by setting the registers.

An example of specific processing in the averaging processing unit 172 and the subtraction unit 173 will be described with reference to FIG.
As shown in FIG. 3, the delay processing unit 171 in FIG. 2 is a register here. The averaging processing unit 172 includes a subtraction unit 1721, an addition unit 1722, a register 1723, a 1 / m multiplication unit 1724, and an m multiplication unit 1725. In the averaging processing unit 172, the output value (here, the previous value of 1 / m times) is fed back to the input side and used for the averaging process, and this averaging process is realized by an IIR filter. I understand that.
Now, the time sequence of the sampling clock in the ADC 13 is n, the value of the input data is D (n), and the baseline (baseline correction data) value calculated by the averaging process is B (n). If the corrected data value is F (n),
F (n) = D (n) -B (n) (1)
It is.

When the input data value D (n) is within the range of 4LSB × √ (number of integrations) to −8LLSB × √ (number of integrations) described above with respect to the baseline value B (n), the registers 171 and 1723 are stored. In both cases, data is transferred according to the sampling clock, and the baseline value B (n) is updated according to the timing of the next sampling clock, that is, the (n + 1) th sampling clock, as in the following equation. On the other hand, since the register 1723 does not operate when the input data value D (n) is not within the signal range with respect to the baseline value B (n), the baseline value B (n) is not updated.
B (n + 1) = {(m−1) / m} × B (n) + (1 / m) × D (n) = B (n) + (1 / m) × {D (n) −B ( n)}… (2)
However, if the calculation is simply multiplied by 1 / m as in the above equation (2), information loss occurs. Therefore, in the example shown in FIG. 3, in order to avoid this, the m-times multiplication unit 1725 performs an operation by multiplying the baseline value B (n) by m-times, and then the final stage 1 / m multiplication unit 1724 uses m. Is excluded. That is, the baseline value B (n) is updated as shown in Equation (3).
B (n + 1) = (1 / m) × [m × B (n) + {D (n) −B (n)}] (3)

  FIG. 4 shows the result when the above-described baseline subtraction correction processing is performed on the peak-like signal output from the actual TOFMS with m = 2. 4A is a comparison of waveforms before and after correction for a signal having a single peak, and FIG. 4B is a comparison of waveforms before and after correction for a signal having two consecutive peaks of the same magnitude. is there. It can be confirmed that the subsidence of the baseline is sufficiently corrected by the correction, and that the drop in the height of the second peak is sufficiently suppressed even when the peaks are continuous.

Table 1 shows the calculation result of the area value of each peak in FIG. It can be confirmed that the calculation accuracy of the peak area value is improved by performing the baseline subtraction correction in this way.

  In the configuration illustrated in FIG. 1, the baseline sink correction unit 17 is provided at the subsequent stage of the baseline correction unit 16, but the order may be changed.

Moreover, the said Example is an example of this invention, and it is clear that even if it changes suitably, amends, and is added within the meaning of this invention, it is included by the claim of this application.
For example, in the above embodiment, the present invention is used for processing data obtained by TOFMS, but the present invention can naturally be applied to processing of data obtained by various analyzers other than TOFMS. is there.

DESCRIPTION OF SYMBOLS 1 ... TOFMS main body 11 ... Ion detector 12 ... Amplifier 13 ... ADC
DESCRIPTION OF SYMBOLS 14 ... ADC value acquisition part 15 ... Integration processing part 16 ... Baseline correction | amendment part 17 ... Baseline sinking correction | amendment part 171 ... Delay processing part (register)
172 ... Averaging processing unit 1721 ... Subtraction unit 1722 ... Addition unit 1723 ... Register 1724 ... 1 / m multiplication unit 1725 ... m multiplication unit 173 ... Subtraction unit 174 ... Data selection unit

Claims (3)

An analytical data processing device that processes time-series data obtained by digitizing an intensity signal in which a peak waveform obtained by an analytical device appears, in order to suppress baseline fluctuations immediately after the peak waveform appears,
a) a data selection unit that selects data whose data value is within a predetermined range near the baseline among sequentially input data;
b) an averaging processing unit that performs an averaging process on a plurality of data in a time-series direction selected by the data selection unit;
c) a subtraction unit that subtracts the value obtained by the averaging process from the input data;
The lower limit of the predetermined range in the data selection unit is defined so as to cover the size of the assumed sinking of the baseline immediately after the appearance of the peak waveform, and the averaging processing period by the averaging processing unit is An analytical data processing apparatus characterized in that it is determined in accordance with an assumed sinking time. The analysis data processing device according to claim 1,
An analysis data processing apparatus, wherein the averaging process is performed by an IIR filter. The analysis data processing device according to claim 2,
The averaging processing unit adds the newly input data value after multiplying the most recent baseline correction value by m (where m> 1) when obtaining the baseline correction value to be subtracted by averaging processing. An analytical data processing apparatus characterized by obtaining a new baseline correction value by multiplying the value by 1 / m after averaging processing.

Images