Automated analysis device and analysis method

Abstract

The present invention provides an automated analysis device and an analysis method capable of predicting, with a high degree of accuracy, the time when an LED light source will need to be replaced, thereby enabling a reduction in equipment costs and workload associated with replacement of the LED light source. This automated analysis device comprises a container that accommodates a liquid, a light source that illuminates the liquid with light, a plurality of light detectors that detect the light in each of mutually different wavelength bands, a storage unit that records, over time, measurement data indicating measurement results of light intensity or photoelectric current, and an analyzing unit that diagnoses the state of the light source, wherein: the light source includes an LED light source; and the storage unit records, in association with each other, measurement data in a steady state of the LED light source, measured for light in one wavelength band detected by one of the plurality of light detectors, and energization time data indicating the energization time until the time of light detection. The analyzing unit and the analysis method obtain a correlation between the light intensity or the photoelectric current and the energization time on the basis of the measurement results and the energization time, and predict the replacement timing of the LED light source on the basis of the correlation.

Description

Automatic Analyzer and Analysis Method

[0001] The present invention relates to an automatic analyzer and an analysis method for analyzing a liquid having a function of diagnosing the state of an LED light source for optical analysis.

[0002] An automatic analyzer for clinical examination automatically quantifies each component contained in a specimen such as blood or urine by an optical method. A general procedure is to dispense a specimen and a reagent into a container such as a cell to prepare a reaction solution, and measure the absorbance or luminescence of the reaction solution to quantify each component. For example, in the case of absorbance analysis, light emitted from a light source is irradiated onto the reaction solution, and the transmitted light that has passed through the reaction solution is detected by a photodetector. The absorbance is calculated based on the intensity of the incident light and the intensity of the detected light, and the amount of each component is determined from the relationship between the absorbance and the concentration.

[0003] As a light source for optical analysis, in order to analyze a plurality of analysis items collectively, it is desirable to have a wide emission spectrum. Also, in order to measure the absorbance or luminescence by each component with high accuracy, it is desirable to be able to stably obtain a light amount of a certain level or more. Conventionally, xenon lamps, halogen lamps, etc. have been used as light sources for optical analysis. However, while these light sources are likely to obtain a light amount of a certain level or more, they have high energy consumption and limited lifetimes. For example, in the case of a halogen lamp, it needs to be replaced after lighting for about 1000 hours.

[0004] In recent years, light emitting diodes (LEDs) have begun to be used as light sources for optical analysis. The lifetime until an LED stops lighting is generally said to be tens of thousands of hours. In many cases, it is recommended to replace the LED when the light amount emitted by the LED has decreased to 70% of the initial value. According to LEDs, a lifetime more than 10 times that of conventional lamp light sources can be obtained. Therefore, it is expected that the frequency of maintenance and replacement can be significantly reduced.

[0005] Considering biochemical parameters among clinical laboratory tests as an example, the reagents to be used and the measurement wavelengths to be measured differ depending on the component being analyzed. The range of measurement wavelengths can be wide, such as from the ultraviolet region to the near-infrared region. When the measurement wavelength range is wide, the wavelength of the light emitted from the light source also ranges wide. With current technology, it is difficult to cover the entire range of measurement wavelengths with a single LED. As a method of emitting light of a wide range of wavelengths, methods using multiple LEDs or phosphor-type light sources are practical.

[0006] In phosphor-based light sources, a combination of an excitation light source that emits excitation light and a phosphor that emits fluorescence when irradiated with excitation light is used. As the excitation light source, a blue LED or similar device that emits short-wavelength light is used. As the phosphor, a combination is used that covers a wide range of wavelengths through direct excitation by the excitation light source or cascade excitation via other phosphors. Typical phosphors include blue phosphors such as BAM:Eu, Sr(PO)Cl:Eu, ZnS:Ag, (Ca,Ba,Mg)(PO)Cl:Eu, and (Y,Gd) ３ (Al, Ga) ５ O １２ : Ce ３＋ , (Ba, Sr, Ca) ２ SiO ４ :Eu ２＋ These are examples of yellow phosphors.

[0007] Even in LED light sources, which are expected to have a long lifespan, it is known that components such as LED chips, phosphors, and encapsulants undergo changes over time. The total radiant flux emitted by an LED light source gradually decreases as these components change over time. If it is possible to predict with high accuracy when LED light sources used in optical analysis need to be replaced, it is expected that appropriate condition-based maintenance (CBM) will become possible. This will also lead to a reduction in costs such as the cost of parts required each time a replacement is needed, as well as a reduction in the workload for operators of automated analyzers.

[0008] Conventionally, in various fields such as automated analytical instruments, various methods have been proposed to diagnose the condition of LED light sources, for purposes such as predicting when the LED light source needs to be replaced.

[0009] For example, Patent Document 1 discloses a technique for predicting the lifespan of an LED by utilizing the change in light intensity over time. In Patent Document 1, the lifespan of an LED is predicted in a short period of time by sequential measurement of the luminous flux of an LED under adverse conditions of high temperature and high humidity and by using an approximation formula based on the Arrhenius law.

[0010] Furthermore, Patent Document 2 discloses a technology that assumes a case where LEDs are used in lighting fixtures and lifespan prediction by accumulating lighting time is not effective. In Patent Document 2, the degree of degradation of the LED light source is determined based on the forward voltage-forward current characteristics of the LED. This method is characterized by the fact that the degradation state of the LED light source can be determined based on electrical characteristics without measuring the light intensity.

[0011] Furthermore, Patent Document 3 discloses an analytical apparatus equipped with an LED light source for irradiating a target for analysis, a determination method for determining the state of the LED light source, and a prediction method for predicting when the LED light source should be replaced. In Patent Document 3, the determination and prediction are made using the time required for the amount of light from the LED light source to stabilize after it has been turned on.

[0012] Japanese Patent Publication No. 2013-011462, Japanese Patent Publication No. 6150390, Japanese Patent Publication No. 2024-138743

[0013] Conventionally, methods for diagnosing the state of an LED light source have been known, such as those described in Patent Documents 1 to 3, which are based on the time-dependent change in the amount of light detected by a photodetector, or methods based on the time-dependent change in the electrical characteristics of the LED light source. However, no method has been found that can be easily implemented in an automated analyzer, nor a method that can accurately predict the future state of an LED light source, such as the timing of future LED light source replacement.

[0014] From the perspective of implementing the method for diagnosing the condition of an LED light source in an automated analyzer, it is desirable to have a method that allows data to be collected while the automated analyzer is in use. When collecting data for diagnosis, it is preferable to have a method that does not require adjustment of the ambient temperature and humidity, and does not require accelerated degradation testing. In the method described in Patent Document 1, the test is conducted in a harsh environment of high temperature and high humidity, making it difficult to collect data for diagnosis while the automated analyzer is in use.

[0015] Furthermore, as a method for diagnosing the state of an LED light source, a method is desired that can accurately predict the future state of the LED light source and perform optical analysis using the LED light source with high precision. In the method described in Patent Document 2, the state of the LED light source is diagnosed based on the forward voltage-forward current characteristics of the LED. Therefore, when the range of measurement wavelengths is wide, for example, when using a phosphor-type LED light source, the differences in changes over time for each wavelength band are not easily reflected in the diagnostic results.

[0016] When the range of wavelengths to be measured optically is wide, there is a problem in that measurement results obtained using a photodetector are prone to errors across different wavelength bands. For example, when using a phosphor-type LED light source as a light source for optical analysis, the excitation light source and the phosphor, among other light-emitting materials, tend to undergo different changes over time. Because the degree of change over time differs for each light-emitting material, the level of the measurement results varies greatly across different wavelength bands, which becomes a problem.

[0017] If the measurement levels vary across wavelength bands, the data used to diagnose the condition of an LED light source will result in variations in the baseline for each wavelength band and errors in the measured values ​​relative to the baseline. If the measurement levels vary significantly across wavelength bands, the diagnostic results will change drastically depending on which wavelength band's data is used for the diagnosis. As a result, it becomes impossible to accurately diagnose the condition of the LED light source.

[0018] For example, when using a phosphor-based LED light source in an automated analyzer, the excitation light emitted by the excitation light source is often irradiated onto multiple types of phosphors, and the excitation light emitted from the excitation light source and the fluorescence of multiple wavelength bands emitted from the multiple types of phosphors are irradiated onto the analyte, and the light from the analyte is detected by multiple photodetectors for each wavelength band. In this case, due to differences in the changes over time for each emitter, a situation may occur where the amount of light in one wavelength band is sufficient for analysis, while the amount of light in other wavelength bands is greatly attenuated.

[0019] In such cases, if conventional prediction methods are used, such as using the integrated value of the lighting time, the sum of the luminous flux across all wavelengths (total radiant flux), or only data from wavelengths with sufficient light intensity, analysis may be performed in wavelengths where light intensity is attenuating before the predicted time for LED light source replacement arrives. Generally, a certain level of light intensity is required for high-precision optical analysis. Using data from wavelengths where light intensity is attenuating will reduce the accuracy of quantitative analysis. In the case of automated analyzers used for clinical testing, this could have a significant impact on the results of clinical tests.

[0020] Furthermore, similar to conventional prediction methods, using the integrated value of the lighting time or the sum of the luminous flux across all wavelengths (total radiant flux) for diagnosis may affect the diagnostic results, even if the light intensity in the wavelengths necessary for analysis is at a level that ensures accuracy, data from wavelengths that are attenuated and therefore unnecessary for analysis may influence the diagnostic results. In such cases, the time when LED light sources need to be replaced may be predicted earlier than the actual time. This encourages unnecessary replacement of LED light sources, resulting in extra equipment costs such as parts costs and the workload associated with replacement.

[0021] Furthermore, LED light sources may experience sudden changes in light intensity or electrical characteristics depending on changes in ambient temperature and humidity, as well as the state of ambient electrical noise. If conventional prediction methods simply monitor changes in light intensity and electrical characteristics over time from the start of operation and only identify a decrease in light intensity or electrical characteristics, there is a risk of misinterpreting sudden changes in light intensity or electrical characteristics as the time when the LED light source needs to be replaced.

[0022] Therefore, the present invention aims to provide an automated analyzer and analysis method that can accurately predict when an LED light source needs to be replaced, thereby reducing the equipment costs and workload associated with replacing the LED light source.

[0023] In other words, to solve the above problem, the automated analyzer according to the present invention comprises a container for containing a liquid to be analyzed, a light source for irradiating the liquid with light, a plurality of photodetectors for detecting light emitted from the liquid in different wavelength bands, a storage unit for recording measurement data over time that shows the measurement results of the amount of light detected by the photodetectors or the photocurrent generated by the light, and an analysis unit for analyzing the measurement data and diagnosing the state of the light source. The light source includes an LED light source, the storage unit records the measurement data in the steady state of the LED light source measured for light in one wavelength band detected by one of the plurality of photodetectors, and the energizing time data showing the energizing time which is the accumulated time that the LED light source was energized up to the time the light was detected for which the measurement data was taken, in correspondence with each other, and the analysis unit determines the correlation between the amount of light or the photocurrent and the energizing time based on the measurement data in the steady state of the LED light source and the energizing time, and predicts the timing of replacement of the LED light source based on the correlation.

[0024] Furthermore, the analysis method according to the present invention involves irradiating a liquid to be analyzed with light from an LED light source and detecting the light emitted from the liquid for each of different wavelength bands. The measurement results of the detected light intensity or the photocurrent generated by the light are recorded in correspondence with the steady-state measurement results of the LED light source measured for one wavelength band and the energizing time, which is the accumulated time that the LED light source was energized up to the time the light was detected for which the measurement results were taken. Based on the steady-state measurement results of the LED light source and the energizing time, a correlation relationship between the light intensity or photocurrent and the energizing time is determined, and the timing of replacement of the LED light source is predicted based on the correlation relationship.

[0025] According to the present invention, it is possible to accurately predict when an LED light source needs to be replaced, and an automated analysis device and analysis method can be provided that can reduce the equipment costs and workload associated with replacing LED light sources.

[0026] This is a diagram showing the configuration of an automated analyzer according to an embodiment of the present invention. This is a block diagram showing the configuration of the data processing unit of an automated analyzer according to an embodiment of the present invention. This is a diagram showing the configuration of the photodetector unit of an automated analyzer according to an embodiment of the present invention. This is a diagram schematically showing the configuration of a non-phosphor type LED light source. This is a diagram schematically showing the configuration of a phosphor type LED light source. This is a flowchart showing the startup operation of an automated analyzer according to an embodiment of the present invention. This is a flowchart showing the process for diagnosing the state of the LED light source. This is a diagram showing the results of analyzing the relationship between the amount of light emitted from the LED light source for the total radiant flux and the energizing time to the LED light source. This is a diagram showing the results of analyzing the relationship between the amount of light emitted from the LED light source for the wavelength band with a central wavelength of 450 nm and the energizing time to the LED light source. This is a diagram showing the results of analyzing the relationship between the amount of light emitted from the LED light source for the wavelength band with a central wavelength of 376 nm and the energizing time to the LED light source. This is a diagram showing the results of analyzing the relationship between the amount of light emitted from the LED light source for the wavelength band with a central wavelength of 450 nm and the energizing time to the LED light source. This is a diagram showing the results of calibrating the amount of light in the wavelength band with a central wavelength of 450 nm using the amount of light in the wavelength band with a central wavelength of 376 nm as a baseline. This figure shows the results of an analysis of the relationship between the ratio of the estimated and measured light intensity emitted from an LED light source and the power supply time to the LED light source.

[0027] The following describes an automated analyzer and an analysis method according to one embodiment of the present invention. In the following figures, common components are denoted by the same reference numerals, and redundant explanations are omitted.

[0028] Figure 1 is a diagram showing the configuration of an automated analyzer according to an embodiment of the present invention. Figure 1 schematically shows an example of the configuration of an automated analyzer according to an embodiment of the present invention. As shown in Figure 1, the automated analyzer 100 according to this embodiment comprises three types of disks 103, 106, and 109, a dispensing mechanism (110, 111) for dispensing liquid, and a control unit 201 that controls the drive mechanism for driving each disk, the dispensing mechanism, etc.

[0029] In the automated analyzer 100, three types of disks 103, 106, and 109 are provided: a sample disk 103, a reagent disk 106, and a reaction disk 109. A dispensing mechanism is provided, including a sample dispensing mechanism 110 and a reagent dispensing mechanism 111. A stirring unit 112, a photodetector unit 113, and a washing unit 114 are provided at predetermined positions on the reaction disk 109. A constant temperature bath 115 is formed on the reaction disk 109.

[0030] Furthermore, the automatic analyzer 100 according to this embodiment includes a light intensity measurement circuit 202, a data processing unit 203, an input unit 204 which is the input interface to the data processing unit 203, and an output unit 205 which is the output interface to the data processing unit 203. These components are connected to each other via buses, channels, etc.

[0031] The automated analyzer 100 is a device that automatically analyzes a liquid to be analyzed. In the automated analyzer 100, the components contained in the liquid to be analyzed are quantified by optical analysis. Optical analysis is performed using an LED light source and multiple photodetectors provided in the photodetector unit 113. In the automated analyzer 100, as an example of optical analysis, absorbance analysis is performed to measure the absorption of light. However, emission analysis, which measures the emission of light such as fluorescence, scattered light analysis, which measures scattered light, or a combination of these may also be performed.

[0032] Examples of liquids to be analyzed include reaction solutions, which are mixtures of a sample and a reagent. In the automated analyzer 100, the reaction solution is automatically prepared and then automatically analyzed. Examples of samples include blood, urine, cerebrospinal fluid, and standard solutions. Examples of components contained in the samples include proteins, sugars, lipids, enzymes, hormones, inorganic ions, and disease markers.

[0033] In addition to analyzing the liquid to be analyzed, the automated analyzer 100 also performs a process to diagnose the condition of the LED light source used for optical analysis. This process of diagnosing the condition of the LED light source is performed during the operation of the automated analyzer 100 for the purpose of maintenance and upkeep of the LED light source. The condition of the LED light source is a state corresponding to the changes in the LED light source over time, and includes a degradation state indicating the degree of deterioration of the LED light source, and a usage state indicating the degree of use of the LED light source.

[0034] In the process of diagnosing the condition of the LED light source, the condition of the LED light source is diagnosed by analyzing data collected during the operation of the automated analyzer 100. For example, it is possible to predict the recommended timing for future replacement of the LED light source for proper maintenance and upkeep. The process of diagnosing the condition of the LED light source can also estimate the current condition of the LED light source and estimate its future condition.

[0035] The process of diagnosing the state of an LED light source uses measurement data obtained by measuring the light emitted by the LED light source in a steady state. The measurement data used is either data showing the measurement result of the amount of light (radiant flux at the detection time) of light emitted from the LED light source in a steady state and detected by the photodetector, or data showing the measurement result of the photocurrent generated by the light emitted from the LED light source in a steady state and detected by the photodetector.

[0036] In the process of diagnosing the state of the LED light source, based on the measurement data of the LED light source in a steady state collected during the operation of the automatic analyzer 100, a correlation is determined between the measurement result of the light intensity or photocurrent detected by the photodetector and the energizing time, which is the cumulative time that power was supplied to the LED light source. Based on the correlation between the measurement results of light intensity or photocurrent and the energizing time, predictions are made about when the LED light source will need to be replaced in the future, the current state of the LED light source is estimated, and the future state of the LED light source is estimated.

[0037] In the automated analyzer 100, the reaction solution, which is the liquid to be analyzed, is prepared and optical analysis of the reaction solution is performed on the sample disc 103, reagent disc 106, and reaction disc 109. The sample disc 103, reagent disc 106, and reaction disc 109 are each driven to rotate by a drive mechanism (not shown). The rotation of each disc transports the material placed on each disc to a predetermined position.

[0038] The sample disk 103 is a disk for dispensing the sample 101. A sample cup 102 is detachably mounted on the sample disk 103. The sample cup 102 is a container for holding the sample 101. For example, a liquid sample 101, which is a specimen, is held in the sample cup 102. Multiple sample cups 102 can be mounted on the sample disk 103. Multiple sample cups 102 are arranged along the circumferential direction of the sample disk 103.

[0039] The reagent disk 106 is a disk for dispensing reagent 104. Reagent bottles 105 are detachably mounted on the reagent disk 106. The reagent bottles 105 are containers for holding reagent 104. Liquid reagent 104 is contained in the reagent bottles 105. For example, if the object of analysis is blood, a solution in which enzymes, measurement substrates, etc. are dissolved is used. Multiple reagent bottles 105 can be mounted on the reagent disk 106. Multiple reagent bottles 105 are arranged along the circumferential direction of the reagent disk 106.

[0040] The reaction disk 109 is a disk for preparing and analyzing the reaction solution 107. A reaction cell 108 is detachably mounted on the reaction disk 109. The reaction cell 108 is an optical analysis container that holds the reaction solution 107, which is the liquid to be analyzed. The reaction cell 108 contains the reaction solution 107, which is a mixture of the sample 101 and the reagent 104. Multiple reaction cells 108 can be mounted on the reaction disk 109. Multiple reaction cells 108 are arranged along the circumferential direction of the reaction disk 109.

[0041] The sample dispensing mechanism 110 is a mechanism for dispensing the sample 101. The sample dispensing mechanism 110 sucks the sample 101 from the sample cup 102 and discharges it into the reaction cell 108, moving a predetermined amount of the sample from the sample cup 102 to the reaction cell 108. The sample dispensing mechanism 110 is composed of, for example, a nozzle for sucking and discharging liquid, a robot arm for moving the nozzle, a pump for driving the sucking and discharging of liquid, a liquid delivery channel for connecting the nozzle and the pump to each other, and the like.

[0042] The reagent dispensing mechanism 111 is a mechanism for dispensing the reagent 104. The reagent dispensing mechanism 111 sucks the reagent 104 from the reagent bottle 105 and discharges it into the reaction cell 108, moving a predetermined amount of the reagent from the reagent bottle 105 to the reaction cell 108. The reagent dispensing mechanism 111 is composed of, for example, a nozzle for sucking and discharging liquid, a robot arm for moving the nozzle, a pump for driving the sucking and discharging of liquid, a liquid delivery channel for connecting the nozzle and the pump to each other, and the like.

[0043] The stirring unit 112 is composed of a mechanism for stirring the liquid. In the stirring unit 112, the sample 101 and the reagent 104 dispensed into the reaction cell 108 are mixed by stirring, and the reaction solution 107 to be the object of optical analysis is prepared. The stirring unit 112 includes, for example, a stirrer for stirring the liquid, a moving mechanism for moving the stirrer into the reaction cell 108, and a driving mechanism for driving the rotation of the stirrer and the like. As the stirrer, a stirring blade, a spatula-shaped stirring rod, and the like are provided.

[0044] The light detection unit 113 is composed of a mechanism for detecting the light radiated from the reaction solution 107 and the like. The light detection unit 113 is provided with a light source for irradiating light to the liquid inside the reaction cell 108 and a plurality of photodetectors for detecting the light radiated from the liquid inside the reaction cell 108. The liquid inside the reaction cell 108 radiates light such as transmitted light and emitted light when irradiated with the light from the light source.

[0045] The light source is an optical light source that irradiates the liquid or other substance to be analyzed with light in a wavelength band having a predetermined central wavelength. The light source is provided with at least an LED light source. Preferably, the light source is provided with one or more types capable of emitting light over a wide range of wavelengths. In addition to the LED light source, lamp light sources such as xenon lamps and halogen lamps, or laser light sources, may be used as needed.

[0046] A photodetector detects light in a wavelength band with a predetermined central wavelength emitted from a light source or the object to be analyzed. The wavelength band typically refers to a range of wavelengths with a width of ±5 nm from the central wavelength. Photodetectors include photodiodes, phototransistors, and the like. A system may consist of multiple photodetectors, each detecting light emitted from the object to be analyzed in different wavelength bands with distinct central wavelengths. The central wavelengths detected by each photodetector are pre-selected before analysis, according to the analytical parameters required for the sample analysis.

[0047] In the photodetector 113, reaction cells 108 containing the reaction solution 107 are sequentially transported by the rotation of the reaction disk 109. The light source and the photodetector are arranged, for example, so as to face each other with the reaction cell 108 transported to the measurement position in between. In the photodetector 113, light of a predetermined wavelength band is irradiated from the light source onto the reaction solution 107 inside the reaction cell 108. The light emitted from the reaction solution 107, etc., is then detected by multiple photodetectors. Each photodetector detects light of wavelength bands with pre-selected and different central wavelengths.

[0048] The photodetector 113 can cover a wide range of measurement wavelengths necessary for the analysis of multiple components by detecting light in multiple wavelength bands using multiple photodetectors. The light source can be of a type capable of emitting light across a wide range of wavelengths to cover this wide range of measurement wavelengths. By analyzing the measurement results for each wavelength band using this wide range of measurement wavelengths, the quantitative determination of multiple components contained in the reaction solution 107 can be performed simultaneously.

[0049] For example, when performing absorbance analysis, multiple photodetectors each detect transmitted light in a wavelength band with a predetermined central wavelength that has passed through the reaction solution 107, for each wavelength band. Based on these detection results, the amount of light and photocurrent of the transmitted light for each wavelength band are measured. Based on these measurements of the amount of light and photocurrent of the transmitted light, the absorbance of the reaction solution 107 is determined for each wavelength band. Absorbance can be calculated as the common logarithm of the ratio of the intensity of incident light to the intensity of transmitted light, based on the Lambert-Beer law.

[0050] The cleaning unit 114 is comprised of a mechanism for cleaning the reaction cell 108. The cleaning unit 114 includes, for example, a nozzle for sucking liquid, a nozzle for discharging liquid, a moving mechanism for moving the nozzle, a vacuum pump for sucking liquid from inside the reaction cell 108, a cleaning pump for supplying cleaning solution to the inside of the reaction cell 108, a tank for storing the cleaning solution, a tank for collecting waste liquid, and a liquid transfer channel for connecting the nozzle, pump, and tank to each other.

[0051] In the cleaning unit 114, reaction cells 108 that have completed optical analysis are sequentially transported by the rotation of the reaction disk 109. In the cleaning unit 114, the liquid inside the reaction cells 108 is discharged by the operation of a vacuum pump. At the same time, cleaning solution is supplied into the reaction cells 108 by the operation of a cleaning pump. The reaction cells 108 are cleaned by repeatedly supplying and discharging the cleaning solution. After cleaning, the reaction cells 108 are reused by dispensing new samples 101 and reagents 104.

[0052] The constant temperature bath 115 is a bath that houses the reaction cell 108 and maintains it at a constant temperature. The constant temperature bath 115 is formed circumferentially on the reaction disk 109. The reaction cell 108 is held in a fluid inside the constant temperature bath 115. The fluid is controlled to maintain a constant temperature and flow rate. Water, air, etc., are used as the fluid. The reaction solution 107 inside the reaction cell 108 is kept at a constant temperature in the constant temperature bath 115 during transport and measurement.

[0053] The control unit 201 controls the equipment involved in the analysis and the processing of data analysis. The control unit 201 controls the drive mechanism that drives the rotation of each disk 103, 106, and 109, the sample dispensing mechanism 110, the reagent dispensing mechanism 111, the stirring unit 112, the photodetector 113, the washing unit 114, the constant temperature bath 115, and other equipment. The control unit 201 controls the forward drive current that lights up the LED light source provided in the photodetector 113. The control unit 201 can perform control to change the drive current according to the results of the diagnosis.

[0054] The light intensity measurement circuit 202 measures the light intensity and photocurrent based on the detection signals detected by the photodetectors. In the photodetection unit 113, light of multiple wavelength bands is detected for each wavelength band by multiple photodetectors. The detection signals detected by each photodetector are amplified and converted into electrical signals, which are then input to the light intensity measurement circuit 202.

[0055] In the light intensity measurement circuit 202, the light intensity of light detected by the photodetector and the photocurrent generated by the light detected by the photodetector are measured for each of the pre-selected, distinct wavelength bands. The measurement data showing the measurement results of light intensity and photocurrent is associated with data indicating the time when light was detected by the photodetector. The measurement data shows the measurement results of light intensity and photocurrent for each wavelength band and for each time of detection. The measurement data is transmitted to the data processing unit 203.

[0056] The data processing unit 203 performs processes such as storing data and analyzing data. Data analysis processes include analyzing the target liquid and diagnosing the state of the LED light source. Analysis result data is generated through these data analysis processes. This analysis result data indicates the results of the data analysis process. For example, the analysis result data is output to the output unit 205.

[0057] In Figure 1, the data processing unit 203 is connected to the server 208 via a communication network such as a public network 206 or a dedicated network 207. The public network 206 is formed by the Internet, etc. The dedicated network 207 is formed by a LAN (Local Area Network), HIS (Hospital Information Systems), etc. Multiple automated analyzers 100 can be connected to the public network 206 or the dedicated network 207.

[0058] The connection between each automated analyzer 100 and the server 208 via the communication network may be encrypted and certified using SSH (Secure Shell), SSL (Secure Sockets Layer), etc. Furthermore, the connection between the automated analyzer 100 and the server 208 via the communication network may be via only the public network 206 or only the dedicated network 207.

[0059] The input unit 204 is a user interface device for input, and is used for inputting data to the data processing unit 203. The input unit 204 is composed of input devices such as a keyboard, touch panel, and mouse. The input unit 204 is operated by an operator using the automatic analyzer 100. The start, operation, and stop of liquid analysis processes, as well as the start, operation, and stop of processes for diagnosing the state of the LED light source, can be instructed via the input unit 204.

[0060] The output unit 205 is a user interface device for output, and performs output of data from the data processing unit 203. The output unit 205 is composed of output devices such as a display, touch panel, and printer. Analysis results from liquid analysis processes and diagnostic results from LED light source status diagnosis processes can be output to the output unit 205. These results may be displayed as images, or output as printed materials, audio, etc.

[0061] The output unit 205 may also output via a GUI (Graphical User Interface). The GUI allows users to operate the automatic analyzer 100 and set analysis conditions in the data analysis process on the image. In addition to displays related to operations, the GUI can also display the operating status of the automatic analyzer 100, the operating status of the data analysis process, and the data and analysis results on the image.

[0062] Server 208 manages user authentication for multiple automated analyzers 100 connected to a communication network, manages the operation of the automated analyzers 100, and manages data. Server 208 can manage analysis content, consumables such as reagents, reagent specifications, and maintenance and upkeep of equipment and parts. Server 208 is operated, for example, by a service department.

[0063] The data processing unit 203 of each automated analyzer 100 can send and receive data related to liquid analysis, data related to the process of diagnosing the state of the LED light source, etc., with the server 208. For example, the data processing unit 203 can send measurement data, analysis result data, etc. It can also receive sample data related to the sample, analysis item data related to the analysis items, parameter data related to the parameters used in the analysis, reference data to be referenced during the analysis, etc.

[0064] Here, an example of control by the control unit 201 in the automated analyzer 100 will be described. In the automated analyzer 100, the quantitative determination of each component contained in the sample 101 is performed, for example, by the following procedure.

[0065] First, the control unit 201 controls the cleaning of the reaction cell 108 by the cleaning unit 114. In the cleaning unit 114, cleaning solution is discharged into the reaction cell 108 by a nozzle. Then, the cleaning solution is drawn out of the reaction cell 108 by the nozzle. By repeating this supply and discharge of cleaning solution, the reaction cell 108 is cleaned before the sample 101 and reagent 104 are dispensed.

[0066] Next, the control unit 201 controls the dispensing of the sample 101 by the sample dispensing mechanism 110. When dispensing the sample 101, the drive mechanism that drives each disk is controlled to drive the rotation of the sample disk 103 and the reaction disk 109. A predetermined sample cup 102 on the sample disk 103 and a predetermined reaction cell 108 on the reaction disk 109 are transported to a predetermined dispensing position. At the dispensing position, a predetermined amount of sample 101 is dispensed from the sample cup 102 into the predetermined reaction cell 108.

[0067] Next, the control unit 201 controls the dispensing of reagent 104 by the reagent dispensing mechanism 111. When the reagent 104 is dispensed, the drive mechanism that drives each disk is controlled to drive the rotation of the reagent disk 106 and the reaction disk 109. A predetermined reagent bottle 105 on the reagent disk 106 and a predetermined reaction cell 108 on the reaction disk 109 are transported to a predetermined dispensing position. At the dispensing position, a predetermined amount of reagent 104 is dispensed from the reagent bottle 105 into the predetermined reaction cell 108.

[0068] Next, the control unit 201 controls the stirring of the liquid by the stirring unit 112. During liquid stirring, the drive mechanism that drives the reaction disk 109 is controlled, and the rotation of the reaction disk 109 is driven. The predetermined reaction cell 108, in which the sample 101 and reagent 104 have been dispensed onto the reaction disk 109, is transported to the stirring unit 112. In the stirring unit 112, the rotation of the stirring bar is driven, and the sample 101 and reagent 104 inside the reaction cell 108 are stirred. Through mixing by stirring, the reaction solution 107 to be analyzed is prepared.

[0069] Next, the control unit 201 controls the optical analysis of the reaction solution 107 by the photodetector 113. During liquid analysis, the drive mechanism that drives the reaction disk 109 is controlled, and the rotation of the reaction disk 109 is driven. The illumination of the light source and detection by the photodetector are also controlled. As the reaction cell 108 containing the prepared reaction solution 107 rotates, it passes through the measurement position where the photodetector 113 is located. Each time it passes through the measurement position, the light emitted from the reaction solution 107 is detected for each wavelength band.

[0070] Light detection in the photodetector 113 is performed for each position on the reaction disk 109. Based on the detection results in the photodetector 113, the amount of light and photocurrent emitted from the reaction solution 107 are measured. Measurement data showing the measurement results of the amount of light and photocurrent are acquired for each reaction solution 107, each wavelength band, and each detection time, and recorded in the data processing unit 203. The multiple measurement data acquired for each reaction solution 107 constitute time-series data showing the reaction process, etc.

[0071] While collecting multiple measurement data, additional reagents 104 can be dispensed into the reaction cell 108 containing the reaction solution 107. The reaction cell 108 to which additional reagents 104 have been added can be re-mixed by the stirring unit 112 and then re-analyzed by the photodetector 113. Dispensing additional reagents 104 allows for combinations of pretreatment and analysis, as well as analysis of sequential reactions due to the addition of reagents 104.

[0072] Figure 2 is a block diagram showing the configuration of the data processing unit of an automated analyzer according to an embodiment of the present invention. As shown in Figure 2, the data processing unit 203 of the automated analyzer 100 according to this embodiment includes a storage unit 2031 and an analysis unit 2032. The storage unit 2031 includes a main storage unit 2033 and an auxiliary storage unit 2034. The analysis unit 2032 includes a processor 2035 and an input / output interface 2036.

[0073] The storage unit 2031 is composed of a storage device that stores information such as data and programs. The storage unit 2031 stores data related to the data analysis process, programs used in the data analysis process, analysis result data generated by the data analysis process, and reference data acquired via a communication network. The data processing unit 203 may acquire reference data to be referenced during analysis via a communication network.

[0074] Data involved in the data analysis process includes historical data 2034a, measurement data, correlation data showing the correlation between the measurement results of light intensity and photocurrent and the energizing time, analysis result data, sample data, analysis item data, parameter data, and identification data that identifies the mounted objects at each position on each disk. Analysis result data includes analysis result data showing the results of the liquid analysis process and diagnostic result data showing the results of the LED light source state diagnosis process. This data can be stored in the auxiliary storage unit 2034.

[0075] Examples of programs used for data analysis include a diagnostic program 2034b used for diagnosing the state of an LED light source, and an analysis program used for analyzing a target liquid. These programs can be stored in the auxiliary storage unit 2034.

[0076] Reference data includes historical data 2034a collected from other LED light sources of the same type as the LED light source being diagnosed, which are provided in other automated analyzers 100 connected to the communication network; correlation data showing the correlation between the measurement results of light intensity and photocurrent determined based on such historical data 2034a and the energizing time; and threshold data used for judgment. This data can be received via the communication network and stored in the auxiliary storage unit 2034.

[0077] The main memory unit 2033 is used as a workspace for the processor 2035. The main memory unit 2033 is composed of DRAM (Dynamic Random Access Memory), etc. The auxiliary memory unit 2034 stores various data and programs nonvolatilously. The auxiliary memory unit 2034 is composed of HDD (Hard Disk Drive), SSD (Solid State Drive), semiconductor flash memory, or a combination thereof.

[0078] The analysis unit 2032 performs data analysis according to a program. The analysis unit 2032 reads data and programs stored in the auxiliary storage unit 2034, as well as data input from peripheral devices, and analyzes measurement data, etc., by executing a predetermined program. Based on the analysis of the measurement data, etc., processes such as liquid analysis and LED light source status diagnosis are performed.

[0079] The processor 2035 loads programs and performs calculations on data. The processor 2035 consists of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), and other components.

[0080] The input / output interface 2036 is a hardware interface that connects the data processing unit 203 and peripheral devices to enable communication between them. Peripheral devices include the control unit 201, the light intensity measurement circuit 202, the input unit 204, the output unit 205, and communication devices that communicate via the public network 206 or the dedicated network 207.

[0081] As shown in Figure 2, the auxiliary storage unit 2034 stores history data 2034a and a diagnostic program 2034b, which are used for processing the state of the LED light source. The history data 2034a is recorded over time by the storage unit 2031 during the operation of the automatic analyzer 100. The diagnostic program 2034b is stored in the auxiliary storage unit 2034 in advance, such as before the automatic analyzer 100 starts operation.

[0082] The historical data 2034a is time-series data collected from an LED light source for optical analysis in a steady state, and shows the history of the measured light intensity or photocurrent in relation to the energization time to the LED light source. The historical data 2034a is recorded by associating measurement data, which shows the measured light intensity or photocurrent for each wavelength band and each detection time in the steady state of the LED light source, with energization time data, which shows the energization time up to the detection of the light for which the measurement data was taken.

[0083] In this specification, "power-on time" refers to the cumulative time during which power is supplied to the LED light source for illumination. Power-on time data can be recorded, for example, using data from a time counter that measures the time during which power is supplied to the LED light source. Power-on time is calculated from the start of use of a new LED light source. Power-on time is not accumulated during the off period when power is not supplied to the LED light source, but is accumulated during the on period when power is supplied to the LED light source.

[0084] The historical data 2034a can be collected while the automated analyzer 100 is in operation, before optical analysis, for example, during the warm-up process when the automated analyzer 100 is started up. The historical data 2034a may be collected periodically or irregularly according to a pre-set schedule. Preferably, the historical data 2034a is collected at least each time the automated analyzer 100 is started up.

[0085] The measurement data constituting the historical data 2034a is collected for multiple different energizing times. The measurement data constituting the historical data 2034a can be obtained by detecting blank light that has not passed through the liquid to be analyzed, emitted from the LED light source, with a photodetector. The blank light may be light that has passed through the reaction cell 108, or light that has not passed through the reaction cell 108.

[0086] The measurement data constituting the history data 2034a may be data showing relative light intensity values ​​with the light intensity of the LED light source at a reference time set to 100%, or it may be data showing absolute light intensity values ​​obtained by referring to the measurement results of the total radiant flux emitted by the LED light source. The reference time can be set to the time when the LED light source is first used, when the power supply time to the LED light source is 0 hours. However, the reference time is not particularly limited.

[0087] The measurement data constituting the historical data 2034a is measured for light in at least one wavelength band detected by one of the multiple photodetectors provided in the photodetector unit 113. The process of diagnosing the state of the LED light source can be performed based solely on the historical data 2034a collected for one wavelength band. By selecting an appropriate wavelength band, it is possible to accurately diagnose the overall state of the LED light source through the process described later.

[0088] The measurement data constituting the historical data 2034a is acquired by detecting the light emitted in the steady state of the LED light source. In this specification, the steady state of the LED light source means the state after the LED light source is turned on, in which the radiant flux emitted by the LED light source no longer substantially exhibits the large fluctuations inherent at the start of operation. In the steady state, the amount of fluctuation per unit time of the radiant flux in each wavelength band emitted by the LED light source is less than or equal to a predetermined value.

[0089] Specifically, a steady state for an LED light source is defined as the state after 30 minutes or more have elapsed since the LED light source was turned on. For example, after 30 minutes to 1 hour has elapsed since the LED light source was turned on, light in a predetermined wavelength band can be detected, and the light intensity and photocurrent of that light can be measured. Such measurement results of light intensity and photocurrent can be collected each time the automatic analyzer 100 is started up.

[0090] Diagnostic program 2034b is a program that performs a diagnostic process for the state of an LED light source. The state of the LED light source is diagnosed in accordance with the changes in the LED light source over time. By diagnosing the state of the LED light source, it is possible to determine the recommended future replacement timing for the LED light source for proper maintenance and upkeep, as well as the current state of the LED light source and its future state.

[0091] When a diagnostic request is received via the input unit 204, the diagnostic program 2034b is read and executed by the analysis unit 2032. When the diagnostic program 2034b is executed, the history data 2034a stored in the auxiliary storage unit 2034, etc., is read. The analysis unit 2032 then determines the correlation between the measured light intensity and photocurrent and the energizing time based on the history data 2034a recorded over time, that is, the measurement data in the steady state of the LED light source and the energizing time data corresponding to this measurement data.

[0092] Next, the analysis unit 2032 predicts the timing of future LED light source replacement based on the derived correlation and the performance conditions that the LED light source should possess. Alternatively, it estimates the current and future state of the LED light source based on the derived correlation and the predetermined time conditions. Diagnostic result data showing these diagnostic results is stored in the auxiliary storage unit 2034, etc. The diagnostic result data can be output to the output unit 205 in response to requests or programs via the input unit 204.

[0093] The functions of the data processing unit 203 may be implemented by a single piece of hardware or by multiple pieces of hardware. Some or all of the functions of the data processing unit 203 may be implemented by cloud computing via a communication network. The data processing unit 203 can also be operated as a remote system. The functions of the data processing unit 203 can also be performed remotely via a communication network.

[0094] According to such an automated analysis device 100, historical data 2034a collected for LED light sources in a steady state is used in the process of diagnosing the state of the LED light source. Specifically, measurement data and energization time data for a single wavelength band detected by one of the multiple photodetectors provided in the photodetector unit 113 are used. This data is collected when the amount of variation per unit time of the radiant flux in each wavelength band emitted by the LED light source falls below a predetermined value.

[0095] Generally, the radiant flux emitted by an LED light source tends to fluctuate significantly when the LED light source is first switched on. Immediately after the LED light source is switched on, the radiant flux becomes unstable due to reasons such as an unstable forward voltage. As time passes immediately after the switch on, the radiant flux in one of the wavelength bands may increase or decrease significantly. The time it takes for the radiant flux to stabilize varies depending on the degree of change over time in the components of the LED light source.

[0096] Fluctuations in the radiant flux emitted by an LED light source pose a greater problem in phosphor-based systems, where the wavelengths of light emitted from the LED light source span a wide range. In phosphor-based systems, the time it takes for the radiant flux emitted by the LED light source to stabilize often differs for each wavelength band. This is because the degree of change over time differs for each light-emitting material, such as the excitation light source or phosphor. If the radiant flux in any wavelength band fluctuates, the total radiant flux emitted from the LED light source also fluctuates. As with conventional prediction methods, it is difficult to accurately diagnose the state of the LED light source based on total radiant flux or electrical characteristics.

[0097] In contrast, the automatic analyzer 100 according to this embodiment uses measurement data collected on an LED light source in a steady state to diagnose the state of the LED light source. Therefore, it is possible to avoid the influence of inherent fluctuations in the radiant flux emitted by the LED light source that occur when it starts to light up. Thus, systematic errors can be suppressed between detection times where the measurement results of light intensity and photocurrent differ from each other. Furthermore, when the wavelength of light emitted from the LED light source is wide, systematic errors can be suppressed between wavelength bands where the measurement results of light intensity and photocurrent differ from each other.

[0098] Specifically, by using measurement data collected from LED light sources in a steady state, it is possible to eliminate the influence of differences in the degree of change over time, which varies depending on the excitation light source and the light-emitting material such as the phosphor. This reduces the likelihood of baseline variations across wavelength bands and errors in measured values ​​relative to the baseline in the time-series measurement data used to diagnose the state of the LED light source. Therefore, it becomes possible to accurately diagnose the state of the LED light source.

[0099] Furthermore, by using measurement data collected on LED light sources in a steady state, even when the wavelengths of light emitted from the LED light source are wide-ranging, such as when using a phosphor-type LED light source, it becomes possible to accurately diagnose the overall state of the LED light source based on the measurement results of light in one wavelength band detected by one of multiple photodetectors.

[0100] For example, it is possible to avoid performing analysis based on measurement data in wavelength bands where light intensity is attenuating before the predicted time for LED light source replacement has arrived. Therefore, highly accurate analysis can be performed based on sufficiently large light intensity and photocurrent. In addition, it is possible to avoid predicting the time when LED light source replacement is needed earlier than the actual time when replacement is needed, based on measurement data in wavelength bands that are unnecessary for analysis where light intensity and photocurrent are attenuating. Therefore, unnecessary replacement of LED light sources can be suppressed, reducing equipment costs such as parts costs and the workload associated with replacement.

[0101] Figure 3 is a diagram showing the configuration of the photodetector unit of an automated analyzer according to an embodiment of the present invention. Figure 3 shows an example of the configuration of the photodetector unit 113 of the automated analyzer 100 according to this embodiment. As shown in Figure 3, the photodetector unit 113 may include an LED light source 401 for optical analysis, a spectrometer 305 that spectrally separates light into wavelength bands, and a detector array 3052 formed by a plurality of photodetectors that detect light into wavelength bands.

[0102] When acquiring measurement data, the LED light source 401 emits irradiation light 301 in a predetermined wavelength band under control by the control unit 201. The irradiation light 301 travels along a predetermined optical axis and enters the focusing lens 303. The irradiation light 301 is focused by the focusing lens 303 into a predetermined range on the irradiation surface perpendicular to the optical axis. After being focused, the irradiation light 301 is irradiated onto the reaction solution 107 inside the reaction cell 108.

[0103] In Figure 3, a slit 302 is positioned between the focusing lens 303 and the LED light source 401. By positioning the slit 302 on the light source side, the beam width of the illumination light 301 emitted from the LED light source 401 can be limited. By limiting the range of the radiant flux, the uniformity of the light intensity distribution on the illumination surface perpendicular to the optical axis can be improved. However, the slit 302 on the light source side may be omitted.

[0104] When the reaction solution 107 is irradiated with light 301, light is emitted from the reaction solution 107. For example, when performing absorbance analysis, transmitted light that has passed through the reaction solution 107 and been attenuated is emitted from the reaction solution 107. The light emitted from the reaction solution 107 is incident on the diffraction grating 3051 built into the spectrometer 305. The diffraction grating 3051 spectrally separates the non-monochromatic light into light in predetermined wavelength bands. The light spectrally separated by the diffraction grating 3051 is detected by the detector array 3052 for each wavelength band.

[0105] In Figure 3, a slit 304 is positioned between the reaction cell 108 and the spectrometer 305. By positioning the slit 304 on the spectrometer side, stray light that has not passed through the reaction solution 107, etc., can be prevented from entering the spectrometer 305. Since stray light that causes noise is less likely to be detected, optical analysis can be performed with higher precision. However, the slit 304 on the spectrometer side may be omitted.

[0106] The detector array 3052 is a device in which multiple photodetectors are arranged in an array. The detector array 3052 detects light of multiple different wavelength bands for each wavelength band using multiple photodetectors. In Figure 3, the detector array 3052 includes a first photodetector 3052a that detects light of a first wavelength band, a second photodetector 3052b that detects light of a second wavelength band, and a third photodetector 3052c that detects light of a third wavelength band.

[0107] The first photodetector 3052a, the second photodetector 3052b, and the third photodetector 3052c each detect light in wavelength bands having different central wavelengths. Each central wavelength is selected in advance before analysis, according to the pre-set analysis items and components contained in the reaction solution 107. The detector array 3052 may be equipped with three types of photodetectors, or it may be equipped with any number of photodetectors, such as four or more types.

[0108] When light is detected by the detector array 3052, the optical signals for each wavelength band are converted into electrical signals by a converter. The optical or electrical signals are amplified by an amplifier as needed. The electrical signals indicating the detection results are input to the light intensity measurement circuit 202 and measured as the light intensity, which is the radiant flux at a predetermined detection time, and the corresponding photocurrent. The measurement results for each wavelength band and for each detection time are stored as measurement data in the auxiliary storage unit 2034.

[0109] In addition, the automated analyzer 100 may perform emission analysis such as fluorescence analysis instead of absorbance analysis. For example, if the reaction solution 107 contains a fluorescent component capable of emitting fluorescence, the reaction solution 107 can be irradiated with excitation light from the LED light source 401. When the fluorescent component in the reaction solution 107 is irradiated with excitation light, fluorescence is emitted when the fluorescent component excited by the excitation light returns to its ground state. By detecting such fluorescence, the fluorescent component can be quantified based on the fluorescence intensity.

[0110] The central wavelength of light detected by the photodetector 113 can be in the ultraviolet, visible, or near-infrared regions. Specific examples of central wavelengths include 340 nm, 376 nm, 405 nm, 415 nm, 450 nm, 480 nm, 505 nm, 546 nm, 570 nm, 600 nm, 660 nm, 700 nm, 750 nm, and 800 nm. These central wavelengths can be used for biochemical and immunological tests using blood as the analysis target.

[0111] Optical analysis in the photodetector 113 requires irradiation with light of wavelengths corresponding to each component in the reaction solution 107. However, if the wavelength range of light emitted from the LED light source is wide, shorter wavelength light is more likely to incident on the components of the LED light source. The components of the LED light source are prone to changes over time due to the high-energy, short-wavelength light. Therefore, measurement data measured in an appropriate wavelength range should be used to diagnose the state of the LED light source.

[0112] Figures 4A and 4B illustrate the configuration of an LED light source. Figure 4A schematically shows the configuration of a non-phosphor type LED light source that does not use phosphors. Figure 4B schematically shows the configuration of a phosphor type LED light source that uses phosphors. In Figures 4A and 4B, the differences in the light emission methods of LED light sources are explained using SMD (Surface Mount Device) type LED light sources as an example.

[0113] As shown in Figures 4A and 4B, in the SMD type LED light sources 401a and 401b, the LED chip 4011 is mounted on the surface of the substrate. The LED chip 4011 is a semiconductor light-emitting element that emits light of a predetermined wavelength. The LED chip 4011 is mounted on the surface of the package 4012. The LED chip 4011 is electrically connected to electrodes (not shown) via bonding wires or the like.

[0114] Package 4012 functions as a substrate or packaging for supporting and sealing the components of the LED light source. Package 4012 is formed from, for example, ceramics or a metal such as aluminum. Package 4012 is soldered to a printed circuit board or the like. The LED light sources 401a and 401b are controlled to light up on the printed circuit board or the like.

[0115] The LED chip 4011 is sealed on the surface of the package 4012 by being covered with a sealing material 4013. The sealing material 4013 is made of a transparent resin with high light transmittance. Epoxy resin, silicone resin, etc., are used as the sealing material 4013. The inner wall of the package 4012 may be provided with a reflector function to reflect light.

[0116] As shown in Figure 4A, in the non-phosphor type LED light source 401a, no phosphor is used, and only an LED chip 4011 is provided as the light-emitting element. On the other hand, as shown in Figure 4B, in the phosphor type LED light source 401b, an LED chip 4011 which is an excitation light source that emits excitation light, and a phosphor 4014 which emits fluorescence when irradiated with excitation light. The phosphor 4014 is dispersed in a encapsulating material 4013, for example.

[0117] In the phosphor-type LED light source 401b, the LED chip 4011 emits excitation light of a predetermined wavelength. The excitation light emitted by the LED chip 4011 irradiates the phosphor 4014, exciting the phosphor 4014. As the excited phosphor 4014 returns to its ground state, it emits fluorescence of a wavelength corresponding to its band gap. Such fluorescence is emitted from the phosphor-type LED light source 401b along with the excitation light.

[0118] The photodetector 113 of the automatic analyzer 100 may be equipped with a non-phosphor type LED light source or a phosphor type LED light source. From the viewpoint of irradiating the liquid to be analyzed with light of a wide range of wavelengths and from the viewpoint of making the process of diagnosing the state of the LED light source more effective, it is preferable to be equipped with at least a phosphor type LED light source. This is because in the case of a phosphor type, high-energy light, which causes large changes in the LED light source over time, is more likely to be emitted.

[0119] The light detection unit 113 of the automatic analyzer 100 may be equipped with a bullet-shaped LED light source or a COB (Chip On Board) type LED light source instead of an SMD type LED light source.

[0120] In the case of a bullet-shaped LED light source, the LED chip is mounted on the surface of a cup-shaped metal component on which a lead frame is formed. The LED chip is covered and sealed with a encapsulant on the surface of the metal component. The excitation chip and metal component are covered with an external encapsulant. The external encapsulant is formed, for example, from a bullet-shaped molded resin that functions as a lens. In the case of a phosphor-type bullet-shaped LED light source, the phosphor can be dispersed in the encapsulant, etc.

[0121] In the case of COB type LEDs, LED chips are mounted on the surface of a substrate on which wiring and terminals are formed. The LED chips are covered and sealed on the surface of the substrate with a encapsulant. The encapsulant is filled so as to cover a wide area of ​​the substrate surface, such as inside a bank formed on the substrate. In the case of phosphor-type COB LED light sources, the phosphor can be dispersed in the encapsulant, etc.

[0122] In the photodetector 113 of the automatic analyzer 100, it is preferable to detect light in the wavelength band with a central wavelength within the range of 340 nm to 800 nm. For the liquid to be analyzed, it is preferable to irradiate it with light in the wavelength band within the range of 340 nm to 800 nm. As the LED light source, one or more types that emit light in all wavelength bands included in this range, or one or more types that emit light in some wavelength bands included in this range can be used.

[0123] The LED light source constituting the light detection unit 113 may consist of a single LED chip or multiple LED chips with different emission wavelength ranges. When multiple LED chips are used, the drive current for each LED chip can be controlled by the control unit 201. By controlling the drive current, the amount of light emitted from each wavelength band from the LED light source can be adjusted.

[0124] The LED light source constituting the photodetector 113 may comprise one type of phosphor, or it may comprise multiple types of phosphors with different emission wavelengths. The phosphors can be excited by direct excitation using an excitation light source, or by cascade excitation using fluorescence emitted by other phosphors as excitation light. By using an LED chip that emits excitation light of a predetermined wavelength and a suitable combination of one or more types of phosphors, a wide range of wavelengths of light can be irradiated.

[0125] The analytical method performed by the automated analyzer 100 includes not only the analysis of the liquid to be analyzed, but also a process to diagnose the state of the LED light source for optical analysis. The analytical method according to this embodiment is a method for analyzing a liquid to be analyzed, such as a reaction solution in which a sample and a reagent are mixed. Examples of samples include blood, urine, cerebrospinal fluid, and standard solutions.

[0126] The analysis method according to this embodiment involves irradiating the liquid to be analyzed with light from an LED light source, detecting the light emitted from the liquid for each of different wavelength bands, and recording the measurement results of the light intensity of the detected light or the photocurrent generated by the detected light, by correlating the measurement result in the steady state of the LED light source measured for light of one wavelength band with the energization time, which is the sum of the time that the LED light source was energized up to the time the measured light was detected.

[0127] In the analysis method according to this embodiment, the correlation between light intensity or photocurrent and energizing time is determined based on the measurement results and energizing time recorded in the steady state of the LED light source. Then, the state of the LED light source is diagnosed based on the determined correlation. By diagnosing the state of the LED light source, it is possible to predict when the LED light source needs to be replaced and to estimate the current and future state of the LED light source. As the LED light source, a phosphor type comprising an excitation light source that emits excitation light and a phosphor that emits fluorescence when irradiated with excitation light is preferred.

[0128] The analysis method according to this embodiment and the process for diagnosing the state of the LED light source in the automatic analyzer 100 will be described below in accordance with the operation of the automatic analyzer 100.

[0129] Figure 5 is a flowchart illustrating the startup operation of an automated analyzer according to an embodiment of the present invention. As shown in Figure 5, the automated analyzer 100 can perform processes for diagnosing the state of the LED light source and collecting measurement results used in the process for diagnosing the state of the LED light source during the warm-up process when the automated analyzer 100 is started up. Figure 5 illustrates a case in which the process for diagnosing the state of the LED light source predicts the timing of future replacement of the LED light source.

[0130] The automatic analyzer 100 starts up when the power is switched ON (step S501). The auxiliary storage unit 2034 of the automatic analyzer 100 stores in advance historical data 2034a and diagnostic programs 2034b that have been collected in the past. In addition, performance conditions to be used for diagnosis are specified in advance. Condition data indicating the conditions to be used for diagnosis is stored in the auxiliary storage unit 2034, etc.

[0131] When the automatic analyzer 100 is started, a warm-up process is initiated (step S502). During the warm-up process, various programs are started and the operation of various devices is checked. The time required for the warm-up process is called the warm-up time.

[0132] Next, during the warm-up process, the LED light source for optical analysis is switched ON (step S503). Generally, an LED light source requires time for the radiant flux emitted into the surroundings to stabilize after power is turned on. Therefore, analysis using the LED light source is not started until at least a predetermined warm-up time has elapsed.

[0133] Next, the light intensity or photocurrent of the LED light source for optical analysis is measured in a steady state (step S504). The light emitted from the LED light source in a steady state is detected for each wavelength band by at least one of the multiple photodetectors provided in the photodetector 113. The measurement data of the LED light source in a steady state, measured for at least one wavelength band, is then input to the data processing unit 203. In addition, energization time data corresponding to this measurement data is acquired.

[0134] Next, new history data 2034a is recorded based on the measurement data of the LED light source in a steady state and the energizing time data corresponding to this measurement data (step S505). The storage unit 2031 records the measurement data of the LED light source in a steady state, measured for light in at least one wavelength band, and the energizing time data, which indicates the energizing time up to the detection of the light for which this measurement data was measured, in correspondence with each other. The time-series history data 2034a is updated by the additional recording of new history data 2034a.

[0135] Next, a process to diagnose the state of the LED light source is executed (step S506). The analysis unit 2032 determines the correlation between light quantity or photocurrent and energization time based on the measurement data of the LED light source in a steady state, measured for light in one wavelength band, and energization time data indicating the energization time up to the detection of the light for which the measurement data was taken. Then, based on the derived correlation and the predetermined performance conditions that the LED light source should possess, it predicts the timing of future replacement of the LED light source.

[0136] For example, in a process for diagnosing the condition of an LED light source, it is possible to predict the remaining time from the current time until the next time the LED light source needs to be replaced. The remaining time is determined as the time corresponding to the limit value that indicates the performance conditions that the LED light source should meet. As the performance conditions that the LED light source should meet, the minimum allowable values ​​of light intensity and photocurrent that are permissible for use in the optical analysis of existing LED light sources can be specified. The minimum allowable values ​​can be specified, for example, so that the light intensity in the wavelength band required for analysis is at a level that ensures accuracy.

[0137] Next, based on the predicted timing of the LED light source replacement, it is determined whether there is sufficient remaining time for the LED light source to be used before the future replacement (step S507). For example, the predicted remaining time until the future replacement of the LED light source can be compared with a preset threshold to determine whether the predicted remaining time is greater than or equal to the threshold.

[0138] The threshold value can be set in advance, for example, before the automatic analyzer 100 starts operation, taking into account the grace period and safety factor until the replacement of existing LED light sources. The threshold value can be set to any time depending on the specifications of the LED light source and the maintenance schedule. For example, if the automatic analyzer 100 is expected to be used daily, a threshold of 24 hours can be set.

[0139] If the assessment determines that the predicted remaining time is below a threshold and that there is insufficient remaining time for the existing LED light source to be used before the recommended replacement time (step S507; No), a warning to that effect is issued (step S508). In this case, the existing LED light source is replaced before the optical analysis is started (step S509).

[0140] When it is determined that there is not enough time remaining for the existing LED light source to be used (step S507; No), the output unit 205 can output an image or the like indicating that it is time to replace the LED light source, that its service life has ended, or that the existing LED light source cannot be used any longer. Alternatively, the control unit 201 can control the illumination of a warning light or the emission of a warning sound to indicate this. Based on the diagnosis, the control unit 201 may also control the forward drive current used to illuminate the LED light source. By performing such control, the amount of light that has been attenuating due to changes over time can be increased by increasing the drive current.

[0141] On the other hand, if the assessment determines that the predicted remaining time is above a threshold and that there is sufficient time remaining for the existing LED light source to be used before the recommended time for replacing the LED light source (step S507; Yes), then optical analysis using the existing LED light source is started (step S510). In this case, the existing LED light source is used for optical analysis without being replaced.

[0142] When it is determined that there is sufficient remaining time for the use of the existing LED light source (step S507; Yes), the output unit 205 can output an image or other information indicating that the analysis can be started normally, that it is time to replace the LED light source, or that the end of its service life has not yet been reached. The history data 2034a for the existing LED light source is continuously updated by the addition of new data.

[0143] According to the flow chart in Figure 5, during the warm-up process that is performed each time the automated analyzer 100 is started, a process to diagnose the state of the LED light source and the collection of measurement results used in the LED light source diagnosis process are performed. Therefore, it is possible to know whether an existing LED light source can be used for optical analysis before the optical analysis is performed. Thus, proper maintenance and upkeep of LED light sources for optical analysis becomes possible.

[0144] Furthermore, according to the flow chart shown in Figure 5, it is determined whether there is sufficient remaining time for use of the LED light source before the next replacement time. Therefore, the lifespan of the existing LED light source and whether the existing LED light source is in a condition that ensures the accuracy of the analysis can be taken into consideration in the diagnosis. Thus, maintenance and upkeep of the LED light source can be carried out while ensuring the accuracy of optical analysis. This enables practical operation that is suitable for reducing costs such as the cost of parts required each time the LED light source is replaced, and the workload for the operator of the automated analyzer.

[0145] Furthermore, according to the flow shown in Figure 5, the output of an optical analysis warning is switched depending on the result of the judgment. If there is sufficient remaining time for use of the existing LED light source, it can be used for optical analysis without replacing the existing LED light source. On the other hand, if there is not enough remaining time for use of the existing LED light source, a warning can be issued to the operator to ensure that replacement is performed. Therefore, a certain level of light intensity or higher can be ensured from the LED light source, enabling stable and highly accurate analysis.

[0146] Figure 6 is a flowchart showing the process for diagnosing the state of an LED light source. Figure 6 shows a detailed flow of the operation in the process for diagnosing the state of an LED light source (step S506) shown in Figure 5. The process shown in Figure 6 is performed by the analysis unit 2032 of the data processing unit 203 executing the diagnostic program 2034b.

[0147] When the diagnostic program 2034b is loaded, the analysis unit 2032 refers to the history data 2034a stored in the auxiliary storage unit 2034 (step S601). The history data 2034a read out is a combination of measurement data of the steady state of the LED light source for light of one wavelength band detected by one of the multiple photodetectors, and the corresponding energization time data.

[0148] The historical data 2034a is data acquired for light in a predetermined wavelength band, and multiple data sets collected for multiple different energizing times are read out. When an LED light source emits light of a wide range of wavelengths, especially when using a phosphor-type LED light source, it is preferable to read out data acquired for light in the shorter wavelength band from among the light detected by multiple photodetectors.

[0149] Next, the analysis unit 2032 derives the correlation between the measurement results of light intensity and photocurrent and the energizing time based on the referenced historical data 2034a (step S602). The analysis unit 2032 performs regression analysis on the read-out historical data 2034a. Through regression analysis, correlation data corresponding to the approximation line showing the correlation between the measurement results of light intensity and photocurrent and the energizing time is approximately obtained. The correlation data may be generated as a relational expression, a data table, or the like.

[0150] The correlation between measurement results and energization time is preferably derived based on measurement results obtained for the short-wavelength band of light detected by multiple photodetectors, especially when an LED light source emits light of a wide range of wavelengths and a phosphor-type LED light source is used. The short-wavelength band is preferably a wavelength band where the center wavelength is shorter than the median of the measurement wavelengths, and more preferably a wavelength band where the center wavelength is within the range of 400 nm to 500 nm. Furthermore, the wavelength band of fluorescence emitted by the phosphor is also preferred.

[0151] By using measurement results obtained in the short-wavelength band, it becomes possible to accurately determine when an LED light source needs to be replaced, given that LED light sources are prone to changes over time. The short-wavelength band has high energy, which easily causes changes over time in the components of the LED light source, such as the LED chip, phosphor, and encapsulant. Furthermore, phosphors such as blue phosphors tend to change more rapidly over time compared to the LED chip, which is the excitation light source. Therefore, by using these measurement results, it is possible to reflect the changes over time that necessitate replacement of the LED light source in the diagnosis and predict with high accuracy when the LED light source needs to be replaced.

[0152] Furthermore, by using measurement results obtained for the shorter wavelength band, the overall state of the LED light source can be diagnosed comprehensively based on the measurement results for light of one wavelength band detected by one of the multiple photodetectors. The overall state of the LED light source is easily reflected in the measurement results for the shorter wavelength band, which is prone to changes over time. Therefore, by using such measurement results, the overall state of the LED light source and the timing of its replacement can be accurately determined without using measurement results obtained for multiple wavelength bands.

[0153] The correlation between the measurement result and the energizing time can be derived by fitting multiple historical data 2034a, obtained in the steady state of the LED light source, to an approximate straight line or approximate curve, which are historical data 2034a for light of one wavelength band detected by one of the multiple photodetectors. The fitting can be performed using the least squares method or the like. The data corresponding to the approximate line can be derived by any method such as exponential approximation, linear approximation, logarithmic approximation, polynomial approximation, power approximation, or moving average approximation.

[0154] The correlation between the measurement results and the energizing time may be regressively derived based on measurement results obtained by integrating the measurement results collected for the LED light source under diagnosis with the measurement results collected for other LED light sources of the same type as the LED light source under diagnosis. The measurement results collected for other LED light sources can be received as reference data, for example, from another automated analyzer 100 connected to a communication network.

[0155] Other LED light sources that have the same specifications as the LED light source being diagnosed, or belong to the same lot as the LED light source being diagnosed, are preferred. Furthermore, LED light sources with a longer energized time than the LED light source being diagnosed are preferred. The measurement results collected for other LED light sources are preferably those measured after a longer energized time than the measurement results collected for the LED light source being diagnosed.

[0156] Integrating these measurement results increases the number of samples used to derive correlations. Therefore, the fitting process for determining correlations can be performed with greater accuracy. In particular, adding measurement results collected for other LED light sources, specifically those taken after a longer energization period than the measurement results collected for the LED light source under diagnosis, allows for more accurate prediction of the future state of the LED light source under diagnosis. Depending on the additional measurement results, it becomes possible to accurately predict the state even after long energization periods.

[0157] The analysis unit 2032 may derive the correlation between the measurement result and the energizing time based on the calibrated historical data 2034a. The historical data 2034a stored in the auxiliary storage unit 2034, that is, the steady-state measurement results of the LED light source for light of one wavelength band detected by one of the multiple photodetectors, can be calibrated in advance before deriving the correlation.

[0158] The calibration of the measurement results can be performed by referring to historical data 2034a for light of multiple wavelength bands detected by multiple photodetectors. When performing calibration of the measurement results, measurements are taken for light of multiple wavelength bands during the measurement of light intensity and photocurrent (step S504). Then, during the recording of historical data 2034a (step S505), the measurement data measured for light of multiple wavelength bands is recorded. Subsequently, during the referencing of historical data 2034a (step S601), the historical data 2034a recorded for light of multiple wavelength bands is referred to. During the derivation of the correlation (step S602), the baseline of the measurement results of light intensity and photocurrent is calibrated, and the correlation between the calibrated measurement results and the energization time is determined.

[0159] For example, calibration of measurement results can be performed using the measured light intensity or photocurrent for one wavelength band detected by one of multiple photodetectors as a baseline. Using the measured light intensity or photocurrent for one wavelength band as a baseline, the measured light intensity or photocurrent for other wavelength bands can be calibrated, and the correlation can be approximately derived using the calibrated light intensity or photocurrent results.

[0160] The baseline wavelength band is preferably the longer wavelength band among the light detected by multiple photodetectors. The longer wavelength band is preferably one whose center wavelength is longer than the median of the measurement wavelengths, and more preferably one whose center wavelength is within the range of 500 nm or more.

[0161] Alternatively, when using a phosphor-type LED light source, the calibration of the measurement results can be performed using the measured light intensity or photocurrent of the excitation light emitted from the excitation light source as a baseline. Using the measured light intensity or photocurrent of the excitation light as a baseline, the measured light intensity or photocurrent of the fluorescence emitted from the phosphor can be calibrated, and the correlation can be approximately derived using the calibrated light intensity or photocurrent measurement results.

[0162] The excitation light used as the baseline preferably has a central wavelength in the ultraviolet region, and more preferably within the range of 400 nm or less. For example, measurement results for excitation light with a central wavelength in the range of 200 nm to 400 nm, preferably within the range of 350 nm to 400 nm, can be used as the baseline.

[0163] Methods for calibrating measurement results include, for example, calculating the ratio of measurement results with different wavelength bands, scaling measurement results with different wavelength bands, and normalizing measurement results with different wavelength bands.

[0164] For example, when recording history data 2034a (step S505), the storage unit 2031 records steady-state measurement data of the LED light source measured for light in a first wavelength band detected by one of the multiple photodetectors, and steady-state measurement data of the LED light source measured for light in a second wavelength band detected by another of the multiple photodetectors. Then, when deriving the correlation (step S602), the analysis unit 2032 calculates the ratio of the measurement result of the light intensity or photocurrent of the light in the first wavelength band to the measurement result of the light intensity or photocurrent of the light in the second wavelength band. When such a ratio is calculated, the measurement result for the second wavelength band forms a baseline, and the measurement result for the first wavelength band is calibrated against the baseline. The analysis unit 2032 can approximately determine the correlation between the calculated ratio and the energizing time.

[0165] Calibrating the measurement results reduces systematic errors that occur when multiple photodetectors detect light in different wavelength bands, as these errors occur between different wavelength bands. Even if the LED light source experiences sudden changes in light intensity or photocurrent due to changes in ambient temperature and humidity, or ambient electrical noise, the effects of these changes can be eliminated, and only the effects of changes over time can be reflected.

[0166] In particular, the longer wavelength band detected by one of the multiple photodetectors, and the excitation light emitted from the excitation source, tend to show relatively small changes over time. By using the measurement results for the longer wavelength band and the measurement results for the excitation light as a baseline, the effects of sudden changes in light intensity and photocurrent, as well as measurement errors between detectors, can be eliminated. Therefore, it is possible to determine a correlation that accurately reflects the effects of changes over time with high precision.

[0167] The analysis unit 2032 can also determine the correlation between the measurement result and the energizing time by correcting it with a correction coefficient. The correlation between the measurement result and the energizing time may be derived based on the corrected historical data 2034a, or it may be corrected with a correction coefficient after the derivation. The correction coefficient can be stored in advance in the auxiliary storage unit 2034, etc., before the automatic analyzer 100 starts operation.

[0168] As correction coefficients, coefficients can be used that correct for environmental conditions such as temperature and humidity in the environment where the LED light source is placed, or coefficients that correct for the power supply conditions to the LED light source. These correction coefficients can be determined regressively by collecting measurement results of light intensity and photocurrent for other LED light sources of the same type as the LED light source being diagnosed, while changing environmental conditions and power supply conditions, and then performing regression analysis on the collected measurement results.

[0169] By correcting the correlation using these correction coefficients, even when environmental conditions and power supply conditions for LED light sources differ, the influence of these differences can be eliminated, allowing for more accurate correlation-based diagnosis. Since the influence of differences in environmental conditions and power supply conditions on the measurement results is reduced, highly accurate diagnostic results that mainly reflect the effects of changes over time can be obtained.

[0170] Furthermore, the correction coefficient can also be a characteristic specific to the type of LED light source, specifically the rate of change of light intensity or photocurrent over time. Such a correction coefficient can be determined based on reference data acquired via a communication network. For example, it can be determined regressively by collecting measurement results of light intensity and photocurrent for each wavelength band and at each detection point for other LED light sources of the same type as the LED light source being diagnosed, and then performing regression analysis on the collected measurement results.

[0171] Such correction coefficients are preferably determined regressively using other LED light sources with longer energizing times compared to the energizing time of the LED light source being diagnosed. Such correction coefficients may be derived by the automated analyzer 100 performing the diagnosis by referring to reference data, by another automated analyzer 100 connected to a communication network, or by a service department operating the server 208.

[0172] By correcting correlations using such rates of change, the future state of an LED light source can be predicted with high accuracy. Even if the power supply time to the LED light source under diagnosis is short, the state after a long period of time can be estimated with high accuracy by referring to the characteristics measured for other LED light sources with longer power supply times. Furthermore, it is possible to eliminate the influence of lot differences and individual differences in LED light sources and obtain highly accurate diagnostic results that reflect the effects of changes over time that are specific to the type of LED light source.

[0173] Next, the analysis unit 2032 calculates the timing of future LED light source replacement based on an approximation line showing the correlation between the measured light intensity and photocurrent and the energizing time (step S603). For example, the analysis unit 2032 extrapolates the allowable minimum values ​​of light intensity and photocurrent to the derived correlation. Through extrapolation, the future energizing time corresponding to the allowable minimum values ​​of light intensity and photocurrent is determined. Then, based on the energizing time to date and the future energizing time, the remaining time until the future LED light source replacement is calculated.

[0174] The minimum permissible values ​​for light intensity and photocurrent can be pre-set before the automatic analyzer 100 is put into operation or when a new LED light source is installed, to ensure accuracy in the wavelength band used for optical analysis. The minimum permissible values ​​can be set arbitrarily, taking into account the specifications of the automatic analyzer 100 and the LED light source, as well as the safety margin for use. For example, the light intensity and photocurrent values ​​can be set to the point where the light intensity in the wavelength band used for optical analysis emitted by the LED light source decreases to 70% of the initial value.

[0175] Next, the analysis unit 2032 outputs the calculation result of the timing for future replacement of the LED light source (step S604). For example, the analysis unit 2032 outputs diagnostic result data showing the result of the process of diagnosing the state of the LED light source, such as predicted result data showing the timing for future replacement of the LED light source and predicted result data showing the remaining time, and stores them in the auxiliary storage unit 2034. The determination of whether or not there is sufficient time remaining for the use of the LED light source (step S507) is made based on such diagnostic result data.

[0176] For example, if it is determined that there is insufficient remaining time for the existing LED light source to be used by the recommended replacement time (step S507; No), or if it is determined that there is insufficient remaining time for the existing LED light source to be used by the recommended replacement time (step S507; Yes), in addition to the results of these determinations, predicted result data indicating the future replacement time of the LED light source and predicted result data indicating the remaining time may be output to the output unit 205 as an image or the like.

[0177] As shown in the flow chart in Figure 6, the timing of future LED light source replacement and remaining time are predicted based on the correlation between the measured light intensity and photocurrent and the energizing time. Therefore, the future state of the LED light source can be predicted with high accuracy by reflecting the changes over time that occur as the energizing time of the LED light source is elapsed. Even if the LED light source experiences sudden changes in light intensity or photocurrent in response to changes in the surrounding environment such as temperature and humidity, or the state of surrounding electrical noise, predictions are made based on the correlation that shows changes over time. Therefore, it is prevented to mistakenly identify sudden changes in light intensity or photocurrent as the time when the LED light source needs to be replaced.

[0178] Furthermore, as shown in the flow chart in Figure 6, the future state of the LED light source can be predicted with high accuracy, allowing the frequency of LED light source replacement to be kept to the minimum necessary for analysis. This enables the realization of automated analyzers and analytical methods that are suitable for reducing equipment costs such as parts costs associated with replacement, and for reducing the workload associated with replacement. For example, it prevents unnecessary premature replacement from being prompted simply because the light intensity in wavelength bands not needed for analysis has decreased, even though the light intensity in the wavelength bands necessary for analysis has reached a level that ensures high accuracy. This also leads to a reduction in the environmental burden associated with the disposal of existing LED light sources.

[0179] Although embodiments of the present invention have been described above, the present invention is not limited to the embodiments described above, and various modifications are included as long as they do not depart from the technical scope. For example, the embodiments described above are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace some of the configurations of one embodiment with other configurations, or to add other configurations to the configuration of one embodiment. It is also possible to add other configurations, delete configurations, or substitute configurations for some of the configurations of one embodiment.

[0180] For example, in the automated analyzer 100 described above, the data processing unit 203 performs processes such as determining the correlation between the measurement results of light intensity and photocurrent and the energizing time, and diagnosing the state of the LED light source. However, the means by which such processes are performed are not particularly limited. Such processes can also be performed in analysis methods that do not use the data processing unit 203.

[0181] Furthermore, in the aforementioned automatic analyzer 100, the processes for determining the correlation between the measurement results of light intensity and photocurrent and the energizing time, and the processes for diagnosing the state of the LED light source are executed during the warm-up process at startup. However, these processes may also be executed during maintenance of the automatic analyzer 100 or at a predetermined time according to a set schedule after the automatic analyzer 100 has started operation.

[0182] Furthermore, in the automated analyzer 100 described above, the process for diagnosing the state of the LED light source predicts the time for replacing the LED light source. However, the current state of degradation of the LED light source, the current state of use of the LED light source, the future state of degradation of the LED light source, the future state of use of the LED light source, etc., may also be estimated. These states may be quantitatively estimated as a degradation rate or usage rate relative to a reference state, or they may be qualitatively estimated as to whether or not a predetermined state of degradation or use is present.

[0183] The present invention will be described in detail below with reference to examples, but the technical scope of the present invention is not limited to these examples.

[0184] <Example 1> Example 1 demonstrates that when the wavelengths of light emitted from an LED light source are wide, the future state of the LED light source can be accurately diagnosed by using the measurement results for light in one wavelength band emitted from the LED light source. Generally, a certain level of light intensity is required to perform optical analysis with high precision. Therefore, if even one of the multiple wavelength bands of light emitted from an LED light source used for optical analysis falls below a certain level, the LED light source needs to be replaced.

[0185] Figures 7A and 7B show the results of an analysis of the relationship between the amount of light emitted from an LED light source and the time the LED light source is energized. Figure 7A shows the relationship between the amount of light emitted from the LED light source and the energizing time for the total radiant flux (all wavelength range). Figure 7B shows the relationship between the amount of light emitted from the LED light source and the energizing time for the wavelength band with a central wavelength of 450 nm. The dashed lines in the figures show the approximation curve obtained by exponential approximation from the six plots in each figure.

[0186] In the analysis of the relationship between light intensity and energizing time, a phosphor-type white LED was used as the LED light source. The white LED comprises an LED chip that emits blue light with a central wavelength of 385 nm and a phosphor that is excited by this blue light. The drive current for driving the LED light source is 600 mA. The light intensity emitted from the LED light source was measured each time the energizing time to the LED light source reached 0 hours, 1000 hours, 5000 hours, 10000 hours, 15000 hours, and 20000 hours.

[0187] Figures 7A and 7B plot the measurement results in a steady state where the variation in the total radiant flux emitted from the LED light source per unit time is below a predetermined value. In each figure, the vertical axis represents the relative light intensity value (%) at each energization time, with the light intensity in the initial state (energy duration 0 hours) set to 100%. In each figure, the horizontal axis represents the energization time (h), which is the cumulative time during which power is supplied to the LED light source to turn it on.

[0188] As shown in Figures 7A and 7B, the relationship between the amount of light emitted from the LED light source and the energizing time of the LED light source varies significantly depending on the measured wavelength. Compared to the decrease in the amount of light with energizing time for the total radiant flux, the decrease in the amount of light with energizing time for the wavelength band with a central wavelength of 450 nm was found to be larger. In the wavelength band with a central wavelength of 450 nm, it is thought that the change in the LED light source over time was larger due to the high-energy, short-wavelength light.

[0189] Table 1 shows the remaining time until future LED light source replacement, predicted based on an approximation curve using exponential approximation. The remaining time was calculated as the time until the amount of light emitted by the LED light source in the wavelength range used for optical analysis decreases to 70% of the initial level.

[0190]

[0191] As shown in Table 1, while sufficient remaining time is ensured for the total radiant flux, a situation may occur where the remaining time is shortened for some wavelength bands. As shown in Figures 7A and 7B, during energization time when the light intensity for the total radiant flux is above a certain level, the light intensity for some wavelength bands, such as the wavelength band with a central wavelength of 450 nm, may decrease significantly.

[0192] In such conditions, the wavelength band with significantly reduced light intensity may be the wavelength band required for analysis. In such cases, if the diagnosis of the LED light source is based on measurement results for total radiant flux or measurement results for other wavelength bands with sufficient light intensity, there is a risk that analysis will be performed in the wavelength band with reduced light intensity before the predicted time for replacing the LED light source arrives. If a certain level of light intensity cannot be ensured, this leads to a problem of reduced accuracy in the analysis.

[0193] For example, when using a wavelength band with a central wavelength of 450 nm for analysis, the light intensity decreases significantly, and the LED light source should ideally be replaced. However, if the measurement results for the total radiant flux are used, as with conventional LED light sources for illumination, there is a risk that the time when the LED light source needs to be replaced will be overlooked. If the wavelength band with a central wavelength of 450 nm is used for analysis after 34,200 hours of power supply to the LED light source, it will become impossible to quantify the components contained in the sample with high accuracy.

[0194] Furthermore, in such conditions, other wavelength bands with sufficient light intensity may correspond to the wavelength bands required for analysis. In such cases, if the measurement results for total radiant flux are used to diagnose the condition of the LED light source, the time when the LED light source needs to be replaced will be predicted earlier than the time when it actually needs to be replaced. If LED light sources are replaced unnecessarily early, the frequency of replacements will increase, leading to problems such as extra equipment costs including parts and increased workload associated with replacements.

[0195] For example, when using wavelength bands other than the wavelength band with a central wavelength of 450 nm for analysis, even if the light intensity in the wavelength band with a central wavelength of 450 nm is significantly reduced, it may still be possible to perform the analysis with high accuracy if sufficient light intensity is maintained in other wavelength bands. However, if the measurement results for the total radiant flux are used, as with conventional LED light sources for illumination, the replacement of the LED light source may be prompted based on the measurement results for the wavelength band where the light intensity is significantly reduced, even though high-precision analysis can be continued.

[0196] In contrast, by using measurement results for a single wavelength band appropriately selected in the process of diagnosing the state of the LED light source, the overall state of the LED light source can be accurately diagnosed. Because the state of the LED light source can be accurately diagnosed, proper maintenance and upkeep of the LED light source become possible before optical analysis. In other words, the amount of light emitted by the LED light source for optical analysis can be maintained above a certain level. Therefore, the performance of automated analyzers that automatically analyze liquids can be kept at a high level.

[0197] <Example 2> Example 2 demonstrates that by using measurement results obtained for light in multiple wavelength bands, the state of an LED light source can be diagnosed with high accuracy by eliminating the influence of environmental conditions such as temperature and humidity in the environment where the LED light source is placed, as well as electrical noise. In Example 2, the measurement result for one wavelength band detected by one of the multiple photodetectors is calibrated using the measurement result for a wavelength band that shows little change over time as a baseline.

[0198] Figures 8A, 8B, and 8C show the results of an analysis of the relationship between the amount of light emitted from an LED light source and the energizing time of the LED light source. Figure 8A shows the relationship between the amount of light emitted from the LED light source and the energizing time for the wavelength band with a central wavelength of 376 nm. Figure 8B shows the relationship between the amount of light emitted from the LED light source and the energizing time for the wavelength band with a central wavelength of 450 nm. Figure 8C shows the results of calibrating the relationship between the amount of light and the energizing time.

[0199] Figure 8C shows the results of calibrating the light intensity in the wavelength band with a central wavelength of 450 nm, using the light intensity in the wavelength band with a central wavelength of 376 nm as the baseline, based on the results shown in Figures 8A and 8B. The light intensity in the wavelength band with a central wavelength of 450 nm is shown as a relative value with the initial light intensity after calibration set to 100%.

[0200] In the analysis of the relationship between light intensity and energizing time, a phosphor-type white LED was used as the LED light source. The white LED consists of an LED chip that emits blue light with a central wavelength of 385 nm and a phosphor that is excited by this blue light. The drive current for driving the LED light source is 600 mA. The light intensity emitted from the LED light source was measured every 2000 hours from 0 hours to 20000 hours of energizing time to the LED light source.

[0201] Figures 8A, 8B, and 8C plot the measurement results in a steady state where the variation in the total radiant flux emitted from the LED light source per unit time is below a predetermined value. In each figure, the vertical axis represents the relative light intensity (%) at each energization time, with the light intensity in the initial state (energy duration 0 hours) set to 100%. In each figure, the horizontal axis represents the energization time (h), which is the cumulative time during which power is supplied to the LED light source to turn it on.

[0202] In Figures 8A and 8B, as indicated by the arrows in each figure, the measured light intensity shows an upward deviation in the data where the LED light source has been energized for 16,000 hours. Such sudden fluctuations can be caused by, for example, changes in environmental conditions such as temperature and humidity in the environment where the LED light source is placed, or by the influence of electrical noise. The effects of such sudden fluctuations can be eliminated by performing calibration using measurement results measured across multiple wavelength bands.

[0203] In Figure 8C, based on the results shown in Figures 8A and 8B, the light intensity for the wavelength band with a central wavelength of 450 nm is calibrated using the light intensity for the wavelength band with a central wavelength of 376 nm as the baseline. The vertical axis of Figure 8C shows the ratio of the light intensity for the wavelength band with a central wavelength of 450 nm to the light intensity for the wavelength band with a central wavelength of 376 nm.

[0204] Light in the wavelength band with a central wavelength of 376 nm is emitted by an excitation light source that exhibits little change over time. On the other hand, light in the wavelength band with a central wavelength of 450 nm is emitted by a phosphor that is prone to changes over time. By performing calibration using measurement results obtained for light in these multiple wavelength bands, the influence of sudden fluctuations in the light emitted from the LED light source can be eliminated. Therefore, the state of the LED light source can be diagnosed with high accuracy using approximation lines.

[0205] <Example 3> Example 3 demonstrates that the state of an LED light source can be diagnosed with high accuracy even when using measurement results collected over a relatively short period of time. Generally, when measurement results collected over a long period of time are used, an accurate approximation line showing the correlation between the light intensity measurement results and the energizing time can be derived, enabling high-precision diagnosis. However, by appropriately correcting the measurement results, high-precision diagnosis is possible even when using measurement results collected over a relatively short period of time, such as from the start of energizing up to 3000 hours.

[0206] Figure 9 shows the results of an analysis of the relationship between the ratio of estimated to measured light intensity emitted from an LED light source and the power supply time to the LED light source. Figure 9 shows the results for the wavelength band with a central wavelength of 480 nm emitted from the LED light source. The vertical axis in Figure 9 shows the ratio of estimated light intensity to measured light intensity, based on an approximation line showing the correlation between measured light intensity and power supply time. The horizontal axis in Figure 9 shows the power supply time (h), which is the cumulative time that power was supplied to the LED light source in order to light it up.

[0207] In the analysis of the relationship between light intensity and energizing time, a phosphor-type white LED was used as the LED light source. The white LED consists of an LED chip that emits blue light with a central wavelength of 385 nm and a phosphor that is excited by this blue light. The drive current for driving the LED light source is 600 mA. The measured value of light intensity was taken when the energizing time reached 20,000 hours.

[0208] The estimated light intensity was calculated by deriving an approximation curve showing the correlation between the measured light intensity and the energizing time, and then fitting each energizing time to this approximation curve. An exponential approximation curve based on the Arrhenius equation was used as the approximation curve. Eighteen samples were used to derive the approximation curve. The plots in Figure 9 show the average values ​​across all samples. The error bars in Figure 9 show the standard deviations across all samples.

[0209] For example, the plot for a 3000-hour energization period was obtained by deriving an approximation line showing the correlation between the measured light intensity and the energization period using measurement results from 400 hours to 3000 hours of energization, then applying the 3000-hour energization period to this approximation line to calculate an estimated value, and finally calculating the ratio of the estimated value to the measured value.

[0210] As shown in Figure 9, the ratio of estimated to measured light intensity was approximately 0.8 to 0.9 from the start of power supply to the LED light source until 3000 hours. In contrast, after more than 3000 hours of power supply to the LED light source, the ratio of estimated to measured light intensity asymptotically approached 1. Therefore, it can be said that the condition of the LED light source can be diagnosed with higher accuracy by using measurement results collected over a long period of time.

[0211] On the other hand, when using measurement results collected over a relatively short period, such as from the start of power-on to 3000 hours, it is possible to correct the measurement results. Correction can be performed using measurement results collected for other LED light sources of the same type as the LED light source being diagnosed. If the other LED light sources have the same specifications and lot as the LED light source being diagnosed, and measurement results have been collected for the other LED light sources over a longer period than for the LED light source being diagnosed, they can be added as samples for regressively determining the approximation line.

[0212] For example, Figure 9 uses measurement results with a sample size of 18 and a total energizing time of up to 20,000 hours. However, it is also possible to add measurement results from other LED light sources of the same type as the LED light source being diagnosed, with energizing times exceeding 20,000 hours. By increasing the number of samples used to derive the correlation, it is possible to improve the accuracy of diagnosing the state of the LED light source. It is preferable to collect measurement results from other LED light sources at a longer energizing time than the LED light source being diagnosed. For example, measurement results at 10,000 hours, 20,000 hours, 30,000 hours, etc., can be collected.

[0213] 100: Automatic analyzer 101: Sample 102: Sample cup 103: Sample disk 104: Reagent 105: Reagent bottle 106: Reagent disk 107: Reaction solution 108: Reaction cell 109: Reaction disk 110: Sample dispensing mechanism 111: Reagent dispensing mechanism 112: Stirring unit 113: Light detection unit 114: Washing unit 115: Constant temperature bath 201: Control unit 202: Light intensity measurement circuit 203: Data processing unit 2031: Storage unit 2032: Analysis unit 2033: Main storage unit 2034: Auxiliary storage unit 2034a: History data 2034b: Diagnostic program 2035: Processor 2036: Input / output interface 204: Input unit 205: Output unit 206: Public network 207: Dedicated network 208: Server 301: Irradiation light 302: Slit 303: Focusing lens 304: Slit 305: Spectrometer 3051: Diffraction grating 3052: Detector array 3052a: First photodetector 3052b: Second photodetector 3052c: Third photodetector 401: LED light source 4011: LED chip 4012: Package 4013: Encapsulation material 4014: Phosphor

Claims

A container for holding the liquid to be analyzed, A light source that irradiates the aforementioned liquid with light, Multiple photodetectors for detecting light emitted from the liquid in different wavelength bands, A storage unit that records measurement data over time, showing the measurement results of the amount of light detected by the photodetector or the photocurrent generated by the light, The system includes an analysis unit that analyzes the measurement data to diagnose the state of the light source, The aforementioned light source includes an LED light source. The storage unit records, in correspondence with each other, the measurement data of the LED light source in a steady state, measured for light of one wavelength band detected by one of the plurality of photodetectors, and the energizing time data, which represents the time during which power was supplied to the LED light source up to the time the light was detected for which the measurement data was taken. The analysis unit is an automated analyzer that determines the correlation between the light quantity or photocurrent and the energizing time based on the measurement data in the steady state of the LED light source and the energizing time data, and predicts the timing of replacement of the LED light source based on the correlation. 　 In the automated analyzer described in claim 1, The aforementioned wavelength band is an automated analyzer in which the central wavelength is within the range of 400 nm to 500 nm. 　 In the automated analyzer described in claim 1, The LED light source is an automated analyzer that uses a phosphor system, comprising an excitation light source that emits excitation light and a phosphor that emits fluorescence when irradiated with the excitation light. 　 In an automated analyzer according to any one of claims 1 to 3 　 The storage unit records the measurement data of the LED light source in a steady state, measured for light in a first wavelength band detected by one of the plurality of photodetectors, and the measurement data of the LED light source in a steady state, measured for light in a second wavelength band detected by another of the plurality of photodetectors. The analysis unit calculates the ratio of the light intensity or photocurrent of light in the first wavelength band to the light intensity or photocurrent of light in the second wavelength band, calibrates the baseline of the measurement results based on the calculation result of the ratio, determines the correlation between the calibrated light intensity or photocurrent and the energizing time, and predicts the timing of replacement of the LED light source based on the correlation. 　 In the automated analyzer according to claim 4, The LED light source is a phosphor type comprising an excitation light source that emits excitation light and a phosphor that emits fluorescence when irradiated with the excitation light. The analysis unit is an automated analyzer that calibrates the measurement results using the measured light quantity or photocurrent for the excitation light as a baseline. 　 In the automated analyzer according to claim 4, The analysis unit is an automated analyzer that calibrates the measurement results using the amount of light or the photocurrent measured for light of one wavelength band detected by one of the plurality of photodetectors as a baseline. 　 In the automated analyzer according to any one of claims 1 to 3, The memory unit stores reference data to be used when predicting the timing of replacement of the LED light source. The reference data includes the measurement data of the steady state of the LED light source collected for other LED light sources of the same type as the LED light source being diagnosed, and the energization time data associated with the measurement data. The analysis unit is an automated analyzer that determines the correlation between the light quantity or photocurrent and the energizing time based on the measurement data and energizing time data of the LED light source to be diagnosed in a steady state and the reference data, and predicts the timing of replacement of the LED light source to be diagnosed based on the correlation. 　 In the automated analyzer according to claim 7, The analysis unit regressively determines the rate of change of the light quantity or photocurrent with respect to the energizing time based on the reference data, corrects the correlation obtained for the LED light source to be diagnosed based on the rate of change, and predicts the timing of replacement of the LED light source based on the corrected correlation. 　 In the automated analyzer according to any one of claims 1 to 3, An automatic analyzer comprising a control unit that controls the drive current for illuminating the aforementioned LED light source. 　 In the automated analyzer according to claim 9, The control unit is an automatic analyzer that changes the drive current according to the prediction of when the LED light source should be replaced. 　 The liquid to be analyzed is irradiated with light from an LED light source, and the light emitted from the liquid is detected for each of the different wavelength bands. The measurement results of the light intensity of the detected light or the photocurrent generated by the light are recorded in correspondence with each other, specifically the measurement results in the steady state of the LED light source measured for light in one wavelength band and the energizing time, which is the sum of the time the LED light source was energized up to the time the light was detected for which the measurement results were taken. An analytical method for determining the correlation between the light quantity or photocurrent and the energizing time, based on the measurement results in the steady state of the LED light source and the energizing time, and predicting the timing of replacement of the LED light source based on the correlation.

Images