Automatic analyzer and leak detection method

Abstract

The objective of the present invention is to provide an automated analyzing device with which it is possible to determine water leakage even without adding a new component or control pattern for the same. To this end, an automated analysis device according to the present invention comprises: a dispensing nozzle for dispensing a specimen or a reagent; a pressure generation source for generating pressure within the dispensing nozzle; a flow passage providing communication between the pressure generation source and the dispensing nozzle; a pressure sensor for detecting the pressure in the flow passage; a solenoid valve provided in the flow passage to open and close the flow passage; and a control unit for determining the presence or absence of water leakage on the basis of a change in pressure data accompanying suction or ejection of air by the dispensing nozzle.

Description

Technical Field

[0001] The present invention relates to an automatic analyzer and a method for determining water leakage thereof.

Background Art

[0002] The dispensing mechanism of an automatic analyzer that performs quantitative and qualitative analysis of blood, urine, etc. includes a solenoid valve for adjusting the pressure of a dispensing syringe during suction or discharge of a sample or reagent, a pressure sensor for detecting the pressure in the dispensing flow path, and the like. If foreign matter混入 the solenoid valve or the dispensing flow path deteriorates over time, water leakage from the dispensing mechanism may occur due to malfunction of the solenoid valve or poor flow path condition. Therefore, an automatic analyzer equipped with a function for determining the presence or absence of water leakage from the dispensing mechanism has been proposed. For example, Patent Document 1 discloses that a pressure sensor is provided in a flow path, and at least two solenoid valves installed so as to sandwich the pressure sensor, and pressure waveform data is acquired using the pressure sensor with each solenoid valve closed, and the acquired data is compared with data in a normal state to determine water leakage of the flow path.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the technology described in Patent Document 1 has a problem that not only is it necessary to add new components such as solenoid valves, but it is also necessary to add control patterns for these components.

[0005] An object of the present invention is to provide an automatic analyzer capable of determining water leakage without adding new components or their control patterns.

Means for Solving the Problems

[0006] To solve the aforementioned problems, the automated analyzer of the present invention comprises a dispensing nozzle for dispensing a sample or reagent, a pressure source for generating pressure within the dispensing nozzle, a flow path connecting the pressure source and the dispensing nozzle, a pressure sensor for detecting the pressure within the flow path, a solenoid valve provided in the flow path for opening and closing the flow path, and a control unit for determining whether or not there is water leakage based on changes in pressure data associated with the suction or discharge of air by the dispensing nozzle. [Effects of the Invention]

[0007] According to the present invention, it is possible to provide an automated analyzer that can detect water leaks without adding new components or control patterns. [Brief explanation of the drawing]

[0008] [Figure 1] Schematic diagram of an automated analyzer. [Figure 2] A schematic diagram showing the configuration of the dispensing mechanism. [Figure 3] A block diagram of the configuration related to leak detection processing using pressure data detected by a pressure sensor. [Figure 4] A graph of the pressure waveform acquired by the feature extraction unit when the dispensing nozzle draws in and discharges air (solid line is an example of an abnormal condition with water leakage due to a faulty solenoid valve, dotted line is an example of a normal condition without water leakage). [Figure 5] A graph of the vibration frequency distribution acquired by the feature extraction unit when the dispensing nozzle draws in and discharges air (solid line is an example of an abnormal condition with water leakage due to a faulty solenoid valve, dotted line is an example of a normal condition without water leakage). [Figure 6] A flowchart showing the operation of the automated analyzer during leak detection in Example 1. [Figure 7] A graph of the pressure waveform acquired by the feature extraction unit when the dispensing nozzle draws in and discharges air (solid line is an example of an abnormal condition with water leakage due to a flow path defect, dotted line is an example of a normal condition without water leakage). [Figure 8]A graph of the same pressure waveform as in Figure 4 (with an additional predetermined threshold added compared to Figure 4). [Figure 9] A flowchart showing the operation of the automated analyzer during leak detection in Example 2. [Modes for carrying out the invention]

[0009] Hereinafter, an automated analyzer and a water leak detection method according to an embodiment of the present invention will be described with reference to the drawings. [Examples]

[0010] The configuration and operation of the automated analyzer according to Example 1 will be explained using Figures 1 to 6. First, the configuration of the automated analyzer according to Example 1 will be explained based on Figure 1. Figure 1 is a schematic diagram of the automated analyzer.

[0011] As shown in Figure 1, the automated analyzer 100 consists of a transport line 101, a reagent disk 103, a reaction disk 104, a dispensing mechanism 105, a stirring mechanism 106, a spectrometer 107, a control unit 115, an input unit 123, a display unit 124, and the like.

[0012] The transport line 101 transports a sample rack 111, which holds sample containers 110 containing samples, to the sample dispensing position 121. The dispensing mechanism 105 dispenses the sample from the sample containers 110 into the reaction cell 112 (reaction vessel) at the sample dispensing position 121. The transport line 101 is further connected to a rotor 102. By rotating the rotor 102, sample racks 111 are exchanged between other transport lines 101.

[0013] The reagent disk 103 holds the reagent container 113 containing the reagent and rotates and transfers the reagent container 113 to the reagent dispensing position 122. The dispensing mechanism 105 dispenses the reagent from the reagent container 113 into the reaction cell 112 at the reagent dispensing position 122. Only the amount of reagent necessary for colorimetric analysis is dispensed into the reaction cell 112 and reacts with the components in the sample to be analyzed.

[0014] The reaction disk 104 holds the reaction cell 112 and rotates and transfers the reaction cell 112, which is the object of each operation, to positions where an optical spectrometer 107 for performing colorimetric analysis, a stirring mechanism 106, a reaction cell cleaning mechanism 108, etc. each operate. The periphery of the reaction cell 112 is filled with water or the like and maintained at a constant temperature. Thereby, in the reaction solution, which is a mixture of the specimen and the reagent, the chemical reaction between the components in the specimen and the reagent is promoted.

[0015] The dispensing mechanism 105 sucks the specimen to be subjected to colorimetric analysis from the specimen container 110 and discharges it into the reaction cell 112, and also sucks the reagent corresponding to the analysis target from the reagent container 113 and discharges it into the reaction cell 112. The dispensing mechanism 105 includes an arm 118, a dispensing mechanism motor 119, a dispensing nozzle 116, a dispensing flow path 125, a pressure sensor 126, a dispensing syringe 127, a solenoid valve 128, etc. The arm 118 holds the dispensing nozzle 116 and the liquid level sensor 117. The liquid level sensor 117 detects the presence or absence of liquid by changes in capacitance. A shield part 114 is installed near the position where the dispensing mechanism 105 performs the dispensing operation. Also, the dispensing mechanism motor 119, the pressure sensor 126, the dispensing syringe 127, the solenoid valve 128, etc. are electrically connected to the control part 115. The dispensing mechanism motor 119 moves the dispensing mechanism 105 in the vertical direction or the rotational direction. Note that the dispensing nozzle 116, the dispensing flow path 125, the pressure sensor 126, the dispensing syringe 127, and the solenoid valve 128 will be described later using FIG. 2.

[0016] The stirring mechanism 106 stirs the reaction solution in the reaction cell 112 in order to promote the reaction between the component to be analyzed in the specimen dispensed from the specimen container 110 into the reaction cell 112 and the reagent dispensed from the reagent container 113 into the reaction cell 112.

[0017] The LED light source 120 irradiates output light onto the reaction solution that has been stirred by the stirring mechanism 106 and has undergone a chemical reaction. The optical spectrometer 107 spectroscopes the transmitted light that has passed through the reaction solution. Colorimetric analysis by absorbance measurement is performed based on the spectroscoped transmitted light.

[0018] The reaction cell cleaning mechanism 108 sucks the reaction solution from the reaction cell 112 after the colorimetric analysis is completed, and cleans the reaction cell 112 by discharging and sucking a detergent or the like.

[0019] The nozzle cleaning mechanism 109 cleans the tip of the dispensing nozzle 116 of the dispensing mechanism 105 that has dispensed a sample or a reagent. Thereby, the residue adhering to the dispensing nozzle 116 is removed, and it does not affect the next analysis target.

[0020] The input unit 123 is composed of a keyboard, a mouse, a touch panel, etc., and inputs instructions from the user to the control unit 115. The display unit 124 is composed of an LCD (Liquid Crystal Display) or the like, and displays an operation screen or the like.

[0021] The control unit 115 is composed of a processor, a memory, etc., controls the operations of each mechanism in the automatic analyzer 100, and performs arithmetic processing to obtain the concentration of a predetermined component in a sample (liquid) such as blood or urine. Further, as will be described later, the control unit 115 also determines the presence or absence of water leakage in the dispensing mechanism 105 based on the change in pressure data accompanying the suction or discharge of air by the dispensing nozzle 116.

[0022] Note that the configuration of the automatic analyzer 100 described above is merely an example, and a sample disk for holding a sample without providing a transport line 101 and a rotor 102 may be provided separately, or a sample pretreatment system for performing various pretreatments on the sample may be provided separately. Also, in FIG. 1, an apparatus for measuring biochemical items is exemplified as the automatic analyzer 100, but the present invention can also be applied to an automatic analyzer that performs different analyses such as immunological items in addition to biochemical items.

[0023] Next, the configuration of the dispensing mechanism 105 and the process for determining water leakage in the dispensing mechanism 105 will be described based on FIGS. 2 and 3. Note that the dispensing mechanism 105 subject to water leakage determination may be for a sample or for a reagent.

[0024] Figure 2 is a schematic diagram showing the configuration of the dispensing mechanism. As shown in Figure 2, the dispensing mechanism 105 includes a dispensing nozzle 116, a dispensing channel 125, a dispensing syringe 127, a pump 132, a solenoid valve 128, a pressure sensor 126, etc. The dispensing nozzle 116 dispenses the sample or reagent. The dispensing channel 125 connects the dispensing nozzle 116, the pressure sensor 126, the dispensing syringe 127, the solenoid valve 128, etc. The dispensing syringe 127 (pressure source) generates pressure in the dispensing nozzle 116 for aspirating or dispensing the sample or reagent by being driven vertically with the solenoid valve 128 closed. The pump 132 generates pressure to supply cleaning water for internal rinsing to clean the inside of the dispensing nozzle 116. The solenoid valve 128 is provided in the dispensing channel 125 between the pump 132 and the dispensing syringe 127 and opens and closes the dispensing channel 125. The solenoid valve 128 is open only when supplying rinsing water for internal washing, and remains closed during other times such as when aspirating or dispensing samples or reagents, or during standby. The pressure sensor 126 detects the pressure in the dispensing channel 125. The opening and closing of the solenoid valve 128 and the operation of the dispensing syringe 127 are controlled by the control unit 115 (omitted in Figure 2). Since the pump 132 is basically always running when the automatic analyzer 100 is in operation, the control unit 115 can adjust the fluid and pressure state in the dispensing channel 125 by controlling the opening and closing of the solenoid valve 128.

[0025] Figure 3 is a block diagram of the configuration related to leak detection processing using pressure data detected by a pressure sensor. As shown in Figure 3, the leak detection system consists of a pressure sensor 126, an amplifier 129, an A / D converter 130, a feature extraction unit 131, a control unit 115, a display unit 124, etc. The amplifier 129 amplifies the pressure waveform data acquired by the pressure sensor 126 and outputs it to the A / D converter 130. The A / D converter 130 converts the pressure data amplified by the amplifier 129 into a digital signal and outputs it to the feature extraction unit 131.

[0026] The feature extraction unit 131 receives pressure data converted into a digital signal from the A / D converter 130, extracts features (judgment values) for leak detection, and sends the extracted features to the control unit 115. Here, feature extraction means calculating features such as the average pressure value over a predetermined period (a period corresponding to the peak or trough of the pressure waveform) (hereinafter simply referred to as "average pressure value"), the vibration time width of the pressure waveform, and the vibration frequency of the pressure waveform from the acquired pressure data. Details of the feature extraction method for each feature in the feature extraction unit 131 will be described later.

[0027] As shown in Figure 3, the control unit 115 includes a storage unit 115a, a feature comparison unit 115b, and an anomaly determination unit 115c. The storage unit 115a stores pressure data converted into digital signals by the A / D converter 130, feature quantities extracted by the feature extraction unit 131, and various threshold values ​​set for each solenoid valve 128 or dispensing mechanism 105. The threshold values ​​are used to determine whether or not there is a water leak and are predetermined based on pressure data acquired when the dispensing mechanism 105 is in a normal state without water leaks. The feature comparison unit 115b compares the feature quantities extracted by the feature extraction unit 131 with the threshold values ​​stored in the storage unit 115a and sends the comparison result to the anomaly determination unit. The anomaly determination unit 115c uses the comparison result from the feature comparison unit 115b to determine whether or not there is a water leak in the dispensing mechanism 105.

[0028] When the abnormality detection unit 115c determines that there is a water leak (abnormality), an alarm display signal is output to the display unit 124, and a warning screen is displayed on the display unit 124. This prompts the operator to take action, such as replacing the seal piece. In addition, the control unit 115 stops the analysis operation of the sample. The displayed warning screen can be in various forms, such as displaying text such as "Water leak detected" or displaying a water leak warning light. On the other hand, when the abnormality detection unit 115c determines that there is no water leak (normal), the pressure data converted into a digital signal by the A / D converter 130 and the features extracted by the feature extraction unit 131 are stored in the storage unit 115a, and the analysis operation starts or continues.

[0029] Note that the configurations shown in Figures 2 and 3 are merely examples. For example, the feature extraction unit 131 may be located within the control unit 115, or it may be considered part of the control unit in a broader sense. Furthermore, the control unit that controls the operation of each mechanism and determines the component concentration, and the control unit for the water leakage detection system may be provided separately.

[0030] Here, the method for extracting each feature in the feature extraction unit 131 and the comparison method in the feature comparison unit 115b will be explained based on Figures 4 and 5. Figure 4 is a graph of the pressure waveform acquired by the feature extraction unit when the dispensing nozzle sucks in and discharges air. The solid line shows an example of an abnormal state with water leakage due to a faulty solenoid valve, and the dotted line shows an example of a normal state without water leakage. The reason for sucking in and discharging air instead of liquid is that the difference between normal and abnormal states is small when sucking in and discharging liquid, making it difficult to make a highly accurate judgment.

[0031] Before the dispensing nozzle 116 begins to draw in air, the pressure is near 0 kPa (atmospheric pressure), regardless of whether there is water leakage. During the period of air aspiration from the start to the end (first period), the pressure first swings in the negative direction (negative pressure) before returning to near atmospheric pressure. In this first period, there is no significant difference between the pressure waveform when the solenoid valve 128 is malfunctioning, such as when foreign matter is mixed into the solenoid valve 128, and the pressure waveform when it is functioning normally. This is thought to be because the drive of the dispensing syringe 127 has a greater effect on pressure fluctuations than the effect of the state of the solenoid valve 128 on pressure fluctuations.

[0032] Next, during the period from the end of air intake to the start of air discharge (second period), the pressure repeatedly fluctuates in the positive (positive pressure) direction and the negative (negative pressure) direction, gradually approaching atmospheric pressure. In particular, in the case of a solenoid valve malfunction, a pressure loss occurs inside the dispensing channel 125 due to water leakage from the tip of the dispensing nozzle 116, so the pressure attenuation is greater and the amplitude of the pressure waveform becomes smaller compared to the normal case.

[0033] Subsequently, during the discharge period (third period) from the start to the end of air discharge, the pressure fluctuates in the positive direction before returning to near atmospheric pressure. In this third period, there is no significant difference between the pressure waveform when the solenoid valve is malfunctioning and when it is functioning normally. This is thought to be because the drive of the dispensing syringe 127 has a greater influence on the pressure fluctuations than the influence of the state of the solenoid valve 128 on the pressure fluctuations.

[0034] Next, during the period from the end of air discharge until the dispensing mechanism 105 moves (the fourth period), the pressure repeatedly fluctuates in the negative (negative pressure) direction and the positive (positive pressure) direction, gradually approaching atmospheric pressure. In particular, in the case of a solenoid valve malfunction, a pressure loss occurs inside the dispensing channel 125 due to water leakage from the tip of the dispensing nozzle 116, so the pressure decay is greater and the amplitude of the pressure waveform becomes smaller compared to the normal case. In other words, in the case of a solenoid valve malfunction, the oscillation time width, which is the time from the end of air discharge until the pressure decays to near atmospheric pressure and stabilizes, becomes shorter compared to the normal case.

[0035] In this specification, for example, the oscillation time width of the pressure waveform after the end of air discharge is defined as "the time taken from the start point to the end point, when the pressure fluctuation from the reference pressure value remains within ±0.5 kPa for a predetermined period of time or longer, starting from the end of air discharge." The reference pressure value may be 0 kPa or the pressure value obtained at the start of air intake. The predetermined time can be set to any time, such as 30 ms. For example, in Figure 4, the oscillation time width in the normal case is shown as Δt0, and the oscillation time width in the case of solenoid valve malfunction is shown as Δt1. Furthermore, the period for extracting the oscillation time width is not limited to the fourth period; if the pressure fluctuation converges sufficiently in the second period as well, the oscillation time width may be extracted from the pressure waveform of the second period.

[0036] As described above, the presence or absence of water leakage causes differences in the pressure waveform. Therefore, the feature extraction unit 131 extracts the average pressure value, which is the average pressure of the peaks or troughs in the pressure waveform of any of the first to fourth periods, from the pressure waveform as one of the features for determining the presence or absence of water leakage. The peaks or troughs in each period (the time periods in which the peaks or troughs appear) may be identified based on the drive pattern of the dispensing syringe 127 stored in the memory unit 115a beforehand, or they may be identified based on the pressure waveform in the normal case. Furthermore, as mentioned above, in the first and third periods, there is no significant difference in the pressure waveform between the case of solenoid valve malfunction and the case of normal operation, so the feature extraction unit 131 may extract the average pressure value corresponding to the peaks or troughs only in the second and fourth periods.

[0037] On the other hand, the pressure average threshold, which serves as the criterion for determining whether or not there is a water leak, is stored in the memory unit 115a. In Figure 4, P2 represents the pressure average threshold corresponding to the first peak in the pressure waveform of the second period, and P4 represents the pressure average threshold corresponding to the first trough in the pressure waveform of the fourth period. For example, a value whose amplitude is a predetermined amount smaller than the peak of a normal peak or trough is set as the pressure average threshold. The feature comparison unit 115b then compares the pressure average, which is a feature extracted by the feature extraction unit 131, with the pressure average threshold stored in the memory unit 115a.

[0038] Furthermore, the feature extraction unit 131 may extract the aforementioned vibration time width as another feature for determining the presence or absence of water leakage. In this case, the feature comparison unit 115b compares the vibration time width extracted by the feature extraction unit 131 with the vibration time width threshold stored in the memory unit 115a to determine the presence or absence of water leakage. For example, a value that is a predetermined time shorter than Δt0, which is the vibration time width in the normal case, is set as the vibration time width threshold.

[0039] Furthermore, the feature extraction unit 131 may extract the vibration frequency of the pressure waveform as yet another feature for determining the presence or absence of water leakage. Figure 5 is a graph of the vibration frequency distribution acquired by the feature extraction unit when the dispensing nozzle sucks in and discharges air, with the solid line showing an example of an abnormal state where water leakage is caused by a faulty solenoid valve, and the dotted line showing an example of a normal state where there is no water leakage. The vibration frequency distribution shown in Figure 5 is calculated by the feature extraction unit 131 applying operations such as Fourier transform to the pressure data acquired from the A / D converter 130. The pressure data used in the calculation may be normalized to atmospheric pressure or to the first data point.

[0040] As shown in Figure 5, under normal conditions, a peak is observed at a predetermined low-frequency component. However, in the case of a solenoid valve malfunction, the pressure attenuation is large and the amplitude of the pressure waveform is small, so a peak is observed at 0 Hz (DC component). Therefore, a value that is a predetermined amount smaller than the frequency at which the peak occurs under normal conditions may be stored in the memory unit 115a as a vibration frequency threshold, and the presence or absence of water leakage may be determined by whether or not a peak exists in the high-frequency range above this threshold. Alternatively, if the vibration frequency peak extracted by the feature extraction unit 131 is 0 Hz, it may be determined that there is water leakage without comparing it with the threshold.

[0041] As described above, the feature extraction unit 131 extracts features such as the average pressure, the vibration time width of the pressure waveform, and the vibration frequency of the pressure waveform. The feature comparison unit 115b then compares these features, but the comparison method is not limited to comparison with the absolute value of a threshold. For example, the feature comparison unit 115b may compare whether the difference between the features in a normal state and the extracted features is within a predetermined range (threshold). Alternatively, the feature comparison unit 115b may calculate only the difference, and the abnormality determination unit 115c may determine whether or not there is a water leak based on whether or not the difference is within a predetermined range.

[0042] Furthermore, the abnormality detection unit 115c may periodically monitor time-series changes (e.g., change per unit time) of feature quantities, etc., based on the feature quantities and pressure waveform data stored in the memory unit 115a, and diagnose a water leak if there is a change that exceeds a predetermined tolerance value. If such a predictive diagnosis reveals a water leak, the abnormality detection unit 115c outputs a preventive alarm to the display unit 124. In addition, the abnormality detection unit 115c may predict the timing of the water leak by obtaining an approximation curve from the time-series changes and output a recommended maintenance timing to the display unit 124.

[0043] Furthermore, the results of the abnormality detection unit 115c may be periodically transmitted to a server at the service center, which is connected to the control unit 115 via a network or the like. This allows the service center to check recommended maintenance times and enables timely maintenance. In addition, by aggregating data from multiple automated analyzers on the service center server and performing machine learning, it is possible to improve the accuracy of the threshold used for leak detection.

[0044] As described above, pressure data acquisition and leak detection are performed at at least one of the following times: during analysis preparation, during analysis, or during standby. For example, these can be set to occur at any time, such as once an hour, in the morning, at noon, or at night. However, to prevent waste of samples and reagents, it is desirable to perform leak detection during analysis preparation. Note that "analysis preparation" refers to the period during which various startup processes are performed after the automatic analyzer 100 is activated, "analysis" refers to the period during which the automatic analyzer 100 is performing sample analysis, and "standby" refers to the idle period during which the automatic analyzer 100 is not performing sample analysis.

[0045] In particular, during the analysis operation of the automated analyzer 100, segmented air is aspirated to separate the pressure transmission medium (system water) filling the dispensing channel 125 from the sample or reagent to be dispensed, in order to prevent changes in concentration due to mixing. Therefore, when acquiring pressure data and detecting leaks during analysis, it is desirable to utilize the timing of aspirating this segmented air. Since segmented air is aspirated each time a sample or reagent is dispensed, there is an advantage in being able to detect leaks at a high frequency. As a result, even if a leak occurs during analysis, it is possible to immediately detect it.

[0046] Next, the processing flow of the automatic analyzer 100 during leak detection will be explained based on Figure 6. Figure 6 is a flowchart showing the operation of the automatic analyzer during leak detection in Example 1.

[0047] First, the control unit 115 opens the solenoid valve 128 and supplies flushing water to the dispensing channel 125 to clean the inside of the dispensing nozzle 116 (step S601). By performing this internal cleaning before drawing air into the dispensing nozzle 116 for leak detection, the conditions of the dispensing nozzle 116 and the dispensing channel 125 can be set to the same conditions each time, even if leak detection is performed multiple times, thereby improving the accuracy of leak detection.

[0048] Subsequently, the control unit 115 drives the pressure sensor 126 to begin acquiring pressure data in the dispensing channel 125 (step S602). Furthermore, with the solenoid valve 128 closed, the control unit 115 drives the dispensing syringe 127 to draw air into the dispensing nozzle 116 (step S603). Subsequently, with the solenoid valve 128 closed, the control unit 115 drives the dispensing syringe 127 to discharge air from the dispensing nozzle 116 (step S604). Once the air discharge is complete, the control unit 115 terminates the acquisition of pressure data by the pressure sensor 126 (step S605).

[0049] Next, the feature extraction unit 131 extracts at least one of the following from the acquired pressure data as a feature: the average pressure, the vibration duration of the pressure waveform, and the vibration frequency of the pressure waveform (step S606). Then, the feature comparison unit 115b reads a threshold value that has been pre-stored in the memory unit 115a and compares the feature value extracted in step S606 with the read threshold value (step S607).

[0050] If the comparison result in step S607 is within the threshold range (if the feature quantity is above the threshold), the abnormality determination unit 115c determines that there is no water leakage (normal) and stores data such as the feature quantity and pressure waveform in the storage unit 115a (step S608). If the water leakage is detected during the analysis preparation operation, the analysis is started; if the water leakage is detected during the analysis operation, the analysis is continued or resumed; and if the water leakage is detected during standby, the system returns to standby (step S609).

[0051] On the other hand, if the comparison result in step S607 is outside the threshold range (if the feature quantity is less than the threshold), the abnormality detection unit 115c determines that there is a water leak (abnormality) and outputs a water leak alarm to the display unit 124 (step S610). Also, if a water leak is detected during the analysis operation, the analysis operation is stopped (step S611). [Examples]

[0052] Example 2 will be explained with reference to Figures 7 to 9. The automated analyzer in Example 2 is basically the same as the automated analyzer in Example 1 in terms of configuration, but unlike Example 1, it identifies not only the presence or absence of water leakage, but also the location of the water leakage cause.

[0053] According to Example 1, if a water leak is detected, an alarm is issued and the automatic analyzer 100 is stopped, thus avoiding the waste of samples, reagents, and time. However, during subsequent maintenance, it is necessary to investigate the source of the water leak. This is because the cause of the water leak from the dispensing mechanism 105 may not only be a malfunction of the solenoid valve due to foreign matter contamination, but also a flow failure due to damage to the dispensing channel 125 or loosening of the channel connection. Therefore, in Example 2, a threshold was added to Example 1 to allow for a more detailed comparison of feature quantities.

[0054] Next, the method for determining the presence or absence of water leakage and the location of the cause in Example 2 will be explained using Figures 7 and 8. Figure 7 is a graph of the pressure waveform acquired by the feature extraction unit when the dispensing nozzle sucks in and discharges air. The solid line shows an example of an abnormal state with water leakage due to a flow path defect, and the dotted line shows an example of a normal state without water leakage. Figure 8 is a graph of the same pressure waveform as Figure 4, that is, the solid line shows an example of an abnormal state with water leakage due to a solenoid valve defect, but with the addition of predetermined thresholds (P3, P4', ΔTa, ΔTb) compared to Figure 4.

[0055] Comparing Figure 7 and Figure 8, it can be seen that in the case of water leakage caused by a flow path defect, the amplitude of the pressure waveform is generally smaller compared to the case of water leakage caused by a solenoid valve defect. This is because, when there is a flow path abnormality, in addition to the pressure loss due to water leakage from the tip of the dispensing nozzle 116, there is also a pressure loss due to water leakage from the abnormal point in the dispensing flow path 125. Furthermore, in the case of a flow path abnormality, the vibration time width, which is the time it takes for the pressure to decay and stabilize to near atmospheric pressure after the air suction or discharge has ended, is even shorter than in the case of a solenoid valve defect. Three typical methods for determining the cause, based on these trends, will be explained below with examples.

[0056] The first method of determination involves extracting the average pressure values ​​of the peaks or troughs in the pressure waveform from the pressure data in the first and third periods, where the difference from the normal case was small in the case of a solenoid valve malfunction, and comparing them with the average pressure threshold values ​​for these periods. In the examples in Figures 7 and 8, P3 is added as the average pressure threshold value corresponding to the peak portion in the third period. In the case of a flow path malfunction shown in Figure 7, the average pressure value of the peak portion in the third period is less than the threshold P3 and can be considered abnormal, while in the case of a solenoid valve malfunction shown in Figure 8, the average pressure value of the peak portion in the third period exceeds the threshold P3 and cannot be considered abnormal. It should be noted that for the first period as well, by setting an average pressure threshold value corresponding to the trough portion, it is possible to determine whether or not there is a flow path malfunction, similar to the third period.

[0057] Therefore, even if the extracted average pressure is below the threshold and can be considered abnormal in the second and fourth periods, if the extracted average pressure is above the threshold and cannot be considered abnormal in the first and third periods, it can be determined that the water leak is due to a faulty solenoid valve. On the other hand, if the extracted average pressure is below the threshold and can be considered abnormal in any period, it can be determined that the water leak is due to a faulty flow path.

[0058] Next, the second determination method will be explained. The second determination method involves setting two thresholds, one larger and one smaller, within the second or fourth period, and comparing the average pressure value extracted during that period with these thresholds. In the examples in Figures 7 and 8, in addition to P4, P4' is added as the average pressure value threshold corresponding to the valley portion in the fourth period. In the case of a flow path abnormality shown in Figure 7, the average pressure value in the valley portion of the fourth period is less than both threshold P4 and threshold P4', while in the case of a solenoid valve abnormality shown in Figure 8, the average pressure value in the valley portion of the fourth period is less than threshold P4 but exceeds threshold P4'. Therefore, if the average pressure value extracted in the fourth period falls between the two thresholds, it can be determined to be a solenoid valve abnormality, and if it is smaller than both thresholds, it can be determined to be a flow path abnormality. Note that for the second period as well, by setting a threshold with a smaller amplitude than threshold P2, it is possible to distinguish between a flow path abnormality and a solenoid valve abnormality, similar to the fourth period.

[0059] Next, the third determination method will be explained. The third determination method involves setting two thresholds, one large and one small, for the vibration time width, rather than the average pressure value, and comparing the extracted vibration time width with these thresholds. In the examples in Figures 7 and 8, the smaller threshold ΔTa and the larger threshold ΔTb are added as vibration time width thresholds in the fourth period. In the case of a flow path abnormality shown in Figure 7, the vibration time width Δt2 in the fourth period is less than both the threshold ΔTb and the threshold ΔTa, while in the case of a solenoid valve abnormality shown in Figure 8, the vibration time width Δt1 in the fourth period is less than the threshold Δb but exceeds the threshold Ta. Therefore, if the vibration time width extracted in the fourth period is between the two thresholds, it can be determined to be a solenoid valve abnormality, and if it is smaller than both thresholds, it can be determined to be a flow path abnormality. Depending on the pressure waveform, it is also possible to distinguish between a flow path abnormality and a solenoid valve abnormality in the second period by setting two thresholds, one large and one small, similar to the fourth period.

[0060] Next, the processing flow of the automatic analyzer 100 during leak detection will be explained based on Figure 9. Figure 9 is a flowchart showing the operation of the automatic analyzer during leak detection in Example 2. Here, we will explain using the first detection method, one of the three detection methods mentioned above, as an example.

[0061] First, steps S901 to S909 in Figure 9 are the same as steps S601 to S609 in Figure 6 described above. However, in this embodiment, if the determination result in step S907 is outside (less than) the threshold range of the second or fourth period, the feature comparison unit 115b reads the threshold for the first or third period that is pre-stored in the storage unit 115a, and compares the features extracted during that period with the read threshold (step S910).

[0062] If the comparison result in step S910 is within the threshold range (i.e., the feature quantity is above the threshold), that is, if only the second and fourth periods out of the first to fourth periods are outside the threshold range, the abnormality determination unit 115c determines that there is a solenoid valve abnormality and outputs a solenoid valve malfunction alarm to the display unit 124 (step S911). Also, if a water leak is detected during the analysis operation, the analysis operation is stopped (step S912).

[0063] On the other hand, if the judgment result in step S910 is outside the threshold range (i.e., the feature quantity is less than the threshold), that is, if all periods from the first to the fourth are outside the threshold range, the abnormality determination unit 115c determines that there is a flow path abnormality and outputs a flow path malfunction alarm to the display unit 124 (step S913). Also, if a water leak is detected during the analysis operation, the analysis operation is stopped (step S914).

[0064] In the case of the second and third determination methods, which involve setting multiple thresholds within the same period, the comparison method in step S910 will differ from the steps described above, but the other steps will be the same as in the first determination method.

[0065] The present invention is not limited to the embodiments described above, and various modifications are included. For example, in addition to a syringe, a diaphragm, a micropump, etc., may be used as the negative pressure source. Furthermore, the determination method using the threshold described above is merely an example, and other methods can be used as appropriate depending on the structure of the dispensing mechanism, etc. [Explanation of Symbols]

[0066] 101...Transport line, 102...Rotor, 103...Reagent disc, 104...Reaction disc, 105...Dispensing mechanism, 106...Agitation mechanism, 107...Spectrometer, 108...Reaction cell washing mechanism, 109...Nozzle washing mechanism, 110...Sample container, 111...Sample rack, 112...Reaction cell, 113...Reagent container, 114...Shielding unit, 115...Control unit, 115a...Storage unit, 115b...Feature comparison unit, 115c ...abnormality detection unit, 116...dispensing nozzle, 117...liquid level sensor, 118...arm, 119...motor for dispensing mechanism, 120...LED light source, 121...sample dispensing position, 122...reagent dispensing position, 123...input unit, 124...display unit, 125...dispensing channel, 126...pressure sensor, 127...dispensing syringe, 128...solenoid valve, 129...amplifier, 130...A / D converter, 131...feature extraction unit, 132...pump.

Claims

1. A dispensing nozzle for dispensing samples or reagents, A pressure source that generates pressure within the dispensing nozzle, A flow path connecting the pressure source and the dispensing nozzle, A pressure sensor for detecting the pressure in the aforementioned flow path, A solenoid valve provided in the aforementioned flow path for opening and closing the aforementioned flow path, A control unit that determines whether or not there is water leakage based on the change in pressure data during the fourth period after the discharge of air sucked in by the dispensing nozzle has finished, An automated analyzer equipped with the following features.

2. In the automated analyzer described in claim 1, The control unit, From the aforementioned pressure data, the average pressure value of the peaks or troughs of the pressure waveform is extracted. An automated analyzer characterized by determining the presence or absence of water leakage by comparing the extracted average pressure value with a predetermined average pressure threshold.

3. In the automated analyzer described in claim 2, The control unit is characterized by extracting the average pressure value of the peaks or troughs of the pressure waveform from the pressure data during the second period from the end of air intake to the start of air discharge.

4. In the automated analyzer described in claim 3, The control unit, The pressure average value of the peaks or troughs of the pressure waveform is extracted from the pressure data during the first period of air suction, or from the pressure data during the third period of air discharge. If the extracted average pressure values ​​in the second and fourth periods are smaller than the respective average pressure threshold values, and if the extracted average pressure values ​​in the first and third periods are larger than the respective average pressure threshold values, then it is determined that there is an abnormality in the solenoid valve. An automated analyzer characterized in that, in any of the first, second, third, and fourth periods, if the extracted average pressure value is smaller than the respective average pressure value threshold, it is determined that there is an abnormality in the flow path.

5. In the automated analyzer described in claim 1, The control unit, The oscillation time width of the pressure waveform is extracted from the pressure data during the fourth period after the discharge of the air has ended. An automated analyzer characterized by determining the presence or absence of water leakage by comparing the extracted vibration time interval with a predetermined vibration time interval threshold.

6. In the automated analyzer according to claim 3 or 5, The pressure average value threshold or vibration time-width threshold has two thresholds, one large and one small, within a single period. The control unit, If the extracted average pressure or vibration duration falls between two thresholds, it is determined that there is an abnormality in the solenoid valve. An automated analyzer characterized by determining an abnormality in the flow path if the extracted average pressure or vibration time width is smaller than either threshold.

7. In the automated analyzer described in claim 1, The control unit, The vibration frequency of the pressure waveform is extracted from the aforementioned pressure data. An automated analyzer characterized by determining the presence or absence of water leakage by comparing the extracted vibration frequency with a predetermined vibration frequency threshold.

8. In the automated analyzer described in claim 1, The control unit is characterized by cleaning the inside of the dispensing nozzle before drawing in the air with the dispensing nozzle.

9. In the automated analyzer described in claim 1, An automated analyzer characterized in that, when the acquisition of pressure data and leakage detection are performed during the analysis operation, the air is segmented air that is drawn in before the sample or reagent is aspirated, separating the system water in the dispensing nozzle from the sample or reagent.

10. A method for detecting water leakage in an automatic analyzer comprising: a dispensing nozzle for dispensing a sample or reagent; a pressure source for generating pressure within the dispensing nozzle; a flow path connecting the pressure source and the dispensing nozzle; a pressure sensor for detecting the pressure within the flow path; and a control unit for determining the presence or absence of water leakage based on pressure data obtained from the pressure sensor, wherein The dispensing nozzle performs the steps of drawing in or discharging air, The steps include: the pressure sensor detecting a change in pressure data within the flow path due to the suction or discharge of air; The control unit extracts at least one of the following from the pressure data: the average pressure value of the peak or trough of the pressure waveform, the vibration time width of the pressure waveform, and the vibration frequency of the pressure waveform, and compares it with a predetermined threshold. It has, A method for detecting water leakage in an automated analyzer, characterized in that the control unit cleans the flow path and the inside of the dispensing nozzle before the step in which the dispensing nozzle draws in or discharges air.

11. (delete)

12. In the method for determining water leakage in an automatic analyzer according to claim 10, A method for determining water leakage in an automated analyzer, characterized in that, when the acquisition of pressure data and water leakage determination are performed during the analysis operation, the air is segmented air that is drawn in before the sample or reagent is aspirated, separating the system water in the dispensing nozzle from the sample or reagent.

Images