Inspection delay monitoring device and inspection delay detection method

Abstract

The present invention enables detection and notification of an outlier delay at an early stage. To this end, this inspection delay monitoring device sorts specimens in chronological order of reception at a clinical laboratory (1), determines that an outlier delay has occurred in a determination target specimen when a difference between an elapsed time from reception of the determination target specimen at the clinical laboratory (1) to loading of the determination target specimen into a specimen conveyance device (3) or an automatic analysis device (4) and the maximum elapsed time of a prescribed number of comparison specimens before and after the determination target specimen exceeds a delay reference time, and issues an alarm notification regarding the determination target specimen determined to have the outlier delay.

Description

Inspection Delay Monitoring Device and Inspection Delay Detection Method

[0001] The present invention relates to an inspection delay monitoring device for monitoring inspection delays in a clinical laboratory and an inspection delay detection method therefor.

[0002] Patent Document 1 discloses an automatic analysis system that accurately predicts the stagnation status of specimens in an automatic analysis system and adjusts the timing and order of specimen loading and unloading in each device to prevent an increase in the TAT of emergency specimens.

[0003] Japanese Patent Application Laid-Open No. 2011-242154

[0004] In a clinical laboratory where a specimen transport device and an automatic analyzer are arranged to perform inspections on specimens, a clinical laboratory information system is often introduced to smoothly process a large number of specimens. The clinical laboratory information system holds specimen information including the inspection contents requested for each specimen, etc., and collects and reports the inspection results obtained in the clinical laboratory. Generally, the clinical laboratory information system has not only a function of reporting inspection results of specimens, but also many functions related to clinical inspections. For example, it has a time management function for specimen inspections so that results can be reported quickly. For example, it monitors the passage of time from the reception time of a specimen in the clinical laboratory to the loading time into the specimen transport device or automatic analyzer in the clinical laboratory, and when the elapsed time exceeds a predetermined threshold time, it detects it as an inspection delay and notifies the administrator of the clinical laboratory. In such a time management method, when a large number of specimens are received in the clinical laboratory in a short period of time or when a device failure occurs, a large number of specimens will be detected as inspection delays. On the other hand, although human error by clinical laboratory technicians is a typical example, there is also an inspection delay caused by a specimen being left unattended in the clinical laboratory for some reason and not being inspected.

[0005] While laboratory delays caused by laboratory capacity issues or equipment failures occur simultaneously for numerous samples, these particular delays occur only for one or more specific samples. Therefore, they are characterized by significantly longer delays compared to samples received at similar times, and we will refer to these as "outlier delays." Simple laboratory delay management systems fail to distinguish between outlier delays and delays caused by other factors, requiring clinical laboratory technologists to investigate the cause of a reported delay to identify it as an outlier delay. Early detection and notification of outlier delays to clinical laboratory technologists minimizes overall delays, provided the laboratory itself is operating normally.

[0006] One embodiment of the present invention is a test delay monitoring device that monitors test delays in a clinical laboratory where samples are received and the requested tests are performed using a sample transport device and / or an automated analyzer. The device comprises a detection unit, a delay notification unit that notifies of delays detected by the detection unit, and a storage unit that stores time management data. The time management data stores, for each sample received by the clinical laboratory, the reception time when the sample was received by the clinical laboratory, the delivery time when the sample was delivered to the sample transport device or automated analyzer, and the elapsed time which is the difference between the delivery time and the reception time. The detection unit sorts the samples in the order they were received by the clinical laboratory, and if the difference between the elapsed time of the sample to be judged and the maximum elapsed time of a predetermined number of comparison samples before and after it exceeds the delay standard time, it determines that a significant delay has occurred in the sample to be judged. The delay notification unit then issues an alarm notification for the sample to be judged that has been determined by the detection unit to have a significant delay.

[0007] This enables early detection and notification of protrusion delays. Other challenges and novel features will become apparent from the description and accompanying drawings herein.

[0008] This is an example of a clinical laboratory system configuration. This is an example of a hardware configuration for a laboratory delay monitoring device. This is a functional block diagram of the laboratory delay monitoring device. This is the protruding delay detection flow of the detection unit. This is an example of sample data and time management data. This is an example of a protruding delay detection alarm screen. This is an example of a protruding delay judgment parameter setting screen. This is an example of a support table. This is an example of a protruding delay judgment parameter setting screen. This is an example of delay performance data. This is an example of a learning model using a neural network.

[0009] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In principle, components having the same function are denoted by the same reference numerals in all the drawings used to describe these embodiments.

[0010] Figure 1 shows an example of the configuration of a clinical laboratory system. Clinical laboratory 1 is equipped with a reception device 2 that reads barcodes or RFID tags attached to samples brought into clinical laboratory 1 and accepts the samples, a sample transport device 3 that performs pre-processing and sub-sample division of the received samples, and an automated analyzer 4 that performs sample analysis. These devices are connected to a Laboratory Information System (LIS) 5 via a network 7. LIS 5 is an information system that manages the testing operations in clinical laboratory 1 and collects the progress of sample processing by these devices installed in clinical laboratory 1. In Figure 1, a typical flow of samples in clinical laboratory 1 is shown with thick arrows. LIS 5 collects time management information such as when samples are brought into each of the reception device 2, sample transport device 3, and automated analyzer 4, and when sample processing is completed in each of the devices. The test delay monitoring device 6 monitors the sample emergence delay using the sample information held by the LIS 5 and the time management information collected by the LIS 5 in the clinical laboratory 1.

[0011] In clinical laboratory 1, specimens are sometimes moved automatically using a conveyor-like transport mechanism, and sometimes manually by clinical laboratory technicians. Therefore, due to some error in the transport mechanism or human error, specimens received in clinical laboratory 1 may be left unattended without being fed into the next processing device, potentially causing significant delays in the testing of those specimens. Here, a detection delay caused by a specimen not being delivered to a device in clinical laboratory 1 for any reason is referred to as an "outburst delay." The testing delay monitoring device 6 detects outburst delays and notifies the administrator.

[0012] In clinical laboratory 1, it is common to have multiple specimen transport devices 3 and automated analyzers 4. Furthermore, while multiple identical automated analyzers 4 may be present, there may also be separate automated analyzers for different types of tests, such as one for biochemical testing and another for immunological testing. Other devices besides those illustrated in Figure 1 may also be present.

[0013] The inspection delay monitoring device 6 is implemented by an information processing device that primarily includes a processor (CPU) 11, memory 12, storage device 13, input device 14, output device 15, communication device 16, and bus 17, as shown in Figure 2. The processor 11 functions as a functional unit (functional block) that provides predetermined functions by executing processing according to a program loaded into the memory 12. The storage device 13 stores data and programs used by the functional unit. The storage device 13 uses a non-volatile storage medium such as an HDD (Hard Disk Drive) or SSD (Solid State Drive). The input device 14 is a keyboard, pointing device, etc., and the output device 15 is a display, etc. The communication device 16 enables communication with LIS 5 and other devices via the network 7. For example, sample information and time management information collected by LIS 5 can be obtained via the communication device 16. These are connected to each other via the bus 17 so that they can communicate with one another.

[0014] Furthermore, the inspection delay monitoring device 6 does not need to be implemented with a single information processing device; it may be implemented with multiple information processing devices. Also, some or all of the functions of the inspection delay monitoring device 6 may be implemented as a cloud-based application.

[0015] Figure 3 shows a functional block diagram of the inspection delay monitoring device 6. The protrusion delay monitoring unit 21 includes a condition setting unit 22 for setting conditions for detecting protrusion delays, and a detection unit 23 for detecting protrusion delays based on the conditions set by the condition setting unit 22. The protrusion delay detected by the detection unit 23 is notified to the administrator via the user interface 25 by the protrusion delay notification unit 24. The storage unit 26 stores the data necessary for processing by the protrusion delay monitoring unit 21.

[0016] Figure 4 shows the protrusion delay detection flow of the detection unit 23, and Figure 5 shows examples of sample data 31 and time management data 32 used by the detection unit 23 for detecting the protrusion delay. The detection unit 23 is assumed to periodically perform the detection of the protrusion delay.

[0017] First, the detection unit 23 acquires sample data 31 and time management data 32 from the LIS 5 and stores them in the storage unit 26 (S01). As shown in Figure 5, the sample data 31 and time management data 32 are data for each sample ID that uniquely identifies the sample. Here, the sample data 31 and time management data 32 are stored in a single table, but they may be stored in separate tables, or the data may be stored in a data structure other than a table.

[0018] The sample data 31 is data indicating the attributes of the sample that affect the timing of transport to the next processing device after the sample is received by the receiving device 2. Here, the sample data 31 includes two items: emergency information 41 and test items 42. The sample data 31 may include any attribute data selected to affect the transport timing, and may include attribute data other than the example, or may not include the example attribute data. The emergency information 41 indicates whether the sample is a normal sample or an emergency sample that will be tested with priority over normal samples. The test items 42 indicate the test items requested for the sample.

[0019] The time management data 32 includes three items: reception time 45, delivery time 46, and elapsed time 47. Reception time 45 is the time when the reception device 2 received the sample, and delivery time 46 is the time when the sample was delivered to the sample transport device 3 or automated analyzer 4. Elapsed time 47 is the time between reception time 45 and delivery time 46. However, if the sample has not been delivered to another device after reception, it will be the time between reception time 45 and the current time. For the current time, for example, the time when the test delay monitoring device 6 received time management information from LIS 5 can be used. In the example in Figure 5, the current time is assumed to be 09:02:00, and the elapsed time for sample ID 4 is calculated.

[0020] The detection unit 23 detects the protrusion delay using the sample data 31 and the time management data 32. First, it selects undetermined samples (S02). At this time, the sample population used for detecting and determining the protrusion delay may be all samples received in the clinical laboratory 1 (in this case, sample data 31 is not required), or it may be a subset grouped using the sample attribute data. For example, emergency samples are generally transported from the reception device 2 to the automated analyzer 4 by bypassing the sample transport device 3. Also, since different automated analyzers perform tests depending on the test item, there will be differences in the status of delivery to the automated analyzer depending on the request status for each test item. In this way, by narrowing down the sample population used for detecting and determining the protrusion delay by attribute, it is expected that there will be less variation in the elapsed time of the narrowed-down samples, and an improvement in the accuracy of detecting the protrusion delay can be expected. In step S02, the sample population used for detecting and determining the protrusion delay is sorted by reception time 45, and a determination is made as to whether or not a protrusion delay has occurred in order from the earliest reception time.

[0021] A determination is made (S03) as to whether a single outlier delay has occurred in the selected sample. A single outlier delay means that when the samples of the population to be judged are sorted according to the reception time 45, the elapsed time of one sample is significantly longer than the elapsed times of the samples before and after it. Specifically, if the sample to be judged is sample A, then it is determined that a single outlier delay has occurred in sample A when the following is true: Elapsed time of sample A - Max (elapsed time of m samples before and after sample A) > delay criterion time T ... (Equation 1).

[0022] Equation 1 indicates that a protruding delay is determined when the difference between the elapsed time of sample A and the maximum elapsed time of comparison samples received before and after sample A exceeds the delay threshold T. The delay threshold T is the threshold for determining a protruding delay. If sample A is the i-th sample sorted by reception time 45, then the comparison samples are the (i-m) to (i-1)th and (i+1) to (i+m)th samples. The specified number of comparison samples m and the delay threshold T can be set by the administrator. Here, the number of comparison samples is the same before and after the sample to be judged, but the number may be different. Equation 1 allows for the determination of a protruding delay, distinguishing it from delays caused by equipment failure or excessive sample reception, by looking at the difference between the elapsed time of sample A and the maximum elapsed time of the comparison samples.

[0023] If it is determined that no single outlier delay has occurred for the selected sample, a determination is made as to whether a continuum outlier delay has occurred (S04). A continuum outlier delay refers to a situation where, when the samples of the population to be judged are sorted by reception time 45, the elapsed time of several consecutive samples is significantly larger than the elapsed time of the samples before and after them. This assumes a scenario where multiple samples received at the same time are left unattended in the clinical laboratory. In this case, since the outlier delay cannot be detected by (Equation 1), the determination is made using the following procedure. Let the sample to be judged be sample A, and let sample A be the i-th sample sorted by reception time 45. Elapsed time of sample A - Max (elapsed time of m samples before sample A) > delay criterion time T ... (Equation 2) If (Equation 2) is true, sample A is considered a candidate for a continuum outlier delay, sample A is excluded from the population, and the sample immediately following sample A (the (i+1)-th sample) is used as the sample to be judged, and the determination of (Equation 2) is made. If (Equation 2) is true, the same determination is repeated. If the number of iterations R exceeds the maximum number of iterations and candidates for continuous emergence delay are extracted, it is determined that the delay is due to equipment failure or excessive sample reception, rather than continuous emergence delay, and it is determined that no continuous emergence delay has occurred. If (Equation 2) becomes false before the maximum number of iterations R is reached, it is determined that continuous emergence delay has occurred in multiple samples extracted as candidates for continuous emergence delay.

[0024] If a single or continuous protrusion delay is detected, the protrusion delay notification unit 24 notifies the administrator of the protrusion delay (S05). Figure 6 shows an example of the alarm screen. The alarm screen 50 displays a list of samples in which a protrusion delay has been detected. The list 51 includes items extracted from the sample data 31 and time management data 32, as well as information on whether the sample has been loaded into the sample transport device 3 or the automated analyzer 4, and the delay time. The delay time is calculated using the left-hand side of (Equation 1) or (Equation 2). A delete button 52 is displayed for each sample shown in the list 51. For example, once the clinical laboratory technician has completed the necessary actions for a sample displayed on the alarm screen 50, the display can be deleted by pressing the delete button 52.

[0025] With the protrusion delay determination for one (or more) samples now complete, the detection unit 23 checks if there are any other samples that have not yet been determined (S06). If there are any other samples that have not yet been determined, the detection unit 23 selects them and performs the protrusion delay determination. If there are no other samples, the detection unit waits until the timing for the next data acquisition (S01) (S07).

[0026] Prior to the outburst delay determination described above, it is necessary to pre-set the number of comparison samples m and the delay reference time T. These parameters are set by the condition setting unit 22. The simplest method is for the condition setting unit 22 to display a condition setting screen to the administrator and have the administrator set these parameters. Figure 7 shows an example of the outburst delay determination parameter setting screen. On the outburst delay determination parameter setting screen 60, the administrator enters the number of comparison samples m and the delay reference time T and presses the save button 61 to set these parameters. The set parameters are stored in the storage unit 26 as determination parameters 33.

[0027] Here, parameters such as the number of comparison samples specified m are difficult for administrators to intuitively set to appropriate values. Therefore, it is desirable to provide the condition setting unit 22 with a function to assist in setting the judgment parameters. Figure 8 shows a support table 34 for assisting in setting the number of comparison samples specified m. The support table 34 shows the detection rate (number of samples judged to have an outlier delay / number of samples received) detected when the outlier delay detection flow in Figure 4 is executed by appropriately changing the number of comparison samples specified m and the delay reference time T for past data of sample data 31 and time management data 32 (for example, data for the past month). Administrators can intuitively set the detection rate by considering the possibility of outlier delays occurring. Using the support table 34, administrators set an appropriate number of comparison samples specified m. Figure 9 shows an example of the outlier delay judgment parameter setting screen in this case. On the outlier delay judgment parameter setting screen 65, administrators input the detection rate and the delay reference time T. The condition setting unit 22 searches the support table 34 based on the input value, extracts the number of comparison samples m that gives the closest detection rate to the input detection rate with the delay reference time T closest to the input delay reference time, and displays that value on the protruding delay judgment parameter setting screen 65. When the administrator presses the save button 66, the parameters of the number of comparison samples m and the delay reference time T are set. The set parameters are stored in the storage unit 26 as judgment parameters 33. With this configuration, if there are many misjudgments in the judgment using the protruding delay detection flow in Figure 4, it becomes easy to make adjustments such as lowering the detection rate or lengthening the delay reference time T and resetting the number of comparison samples m.

[0028] The following describes some variations.

[0029] (Modification 1) If the automated analyzer experiences any malfunction, it will be necessary to stop the device, and the delivery of samples will be halted. In such cases, it is desirable to calculate the elapsed time excluding the effects of the device stoppage. That is, when calculating the elapsed time, the time during which the delivery was hindered due to the device malfunction should be excluded. Also, in the case of samples not being delivered, the time should be calculated from the time between the reception time of 45 and the current time, to the time when the automated analyzer scheduled to receive the samples stopped delivering them. This will help to suppress the extraction of incorrect candidates for protruding delays.

[0030] (Modification 2) In some time periods, the number of samples received may be small. In such cases, there is a risk that a considerable time difference may occur between the time of receipt of sample A to be evaluated and the time of receipt of the comparison sample. In such cases, the sample processing status (e.g., congestion level) at the time of receipt of the comparison sample and the sample processing status at the time of receipt of sample A may differ significantly, and the protruding delay may be detected as excessively or understated.

[0031] Therefore, the comparison samples are limited to those whose arrival time is less than or equal to a predetermined time difference from the reception time of sample A. For example, even if the comparison samples are defined as the (i-m) to (i-1)th and (i+1) to (i+m)th samples for the i-th sample A, if the arrival time of the comparison samples is not within the predetermined time range from the arrival time of sample A, they will not be used in the calculation of (Equation 1).

[0032] Alternatively, if the number of samples arriving around the time of sample A's receipt is small, the system may determine that there is a significant delay if the elapsed time is equal to or greater than a predetermined time, without applying (Formula 1).

[0033] (Modification 3) In the embodiment, an example was shown in which the administrator sets the number of comparison samples m and the delay reference time T. However, a learning model may be created to calculate the number of comparison samples m and the delay reference time T, and the system may automatically set the appropriate number of comparison samples m and the delay reference time T according to the status of the test request to the clinical laboratory 1. Any learning device can be used as the learning model, but here we show an example using a learning model based on a neural network.

[0034] To train the learning model, the administrator prepares delay performance data 35. The delay performance data 35 shown in Figure 10 consists of past sample data 31 and time management data 32 from the clinical laboratory 1, which the administrator has labeled to indicate whether or not the sample has an outstanding delay. The labeling is recorded in the judgment result column as either an outstanding delay or not (-). The learning model is then trained using this delay performance data 35.

[0035] Figure 11 shows an example of a learning model using a neural network. The learning model 36 is a learning model for obtaining the appropriate number of comparison samples m and delay reference time T for each time period. By training it using delay performance data 35, it becomes a trained model and is used for the automatic setting of the number of comparison samples m and delay reference time T.

[0036] The learning model 36 propagates information input to the input layer 71 to the hidden layer 72, and then to the output layer 73, which outputs an inference result based on the information input to the input layer 71. The input layer 71, hidden layer 72, and output layer 73 each have multiple input units, hidden units, and output units, respectively, indicated by circles. Note that while a neural network usually has multiple hidden layers, the diagram shows only the hidden layer 72 as a representative example. The information input to each input unit of the input layer 71 is weighted by the coupling coefficient between the input unit and the hidden unit and input to each hidden unit. The value of the hidden unit in the hidden layer 72 is calculated by adding the values ​​from the input units. The output from each hidden unit in the hidden layer 72 is weighted by the coupling coefficient between the hidden unit and the output unit and input to each output unit. The value of the output unit in the output layer 73 is calculated by adding the values ​​from the hidden units.

[0037] After the coupling coefficients are adjusted through training using delay performance data 35, the number of comparison samples m and the delay reference time T are output from the output layer 73 by inputting predetermined time information and the number of samples arriving per hour into the input layer 71. The detection unit 23 performs detection of protruding delays using the judgment parameters output from the learning model 36.

[0038] The present invention is not limited to the embodiments described above, and various modifications are included. For example, although an example in which a clinical testing system is equipped with a test delay monitoring device 6 has been described, it is also possible to incorporate the functions of the test delay monitoring device 6 as functions of a clinical testing information system. Thus, the embodiments and modifications described above are explained in detail to make the present invention easier to understand, and are not necessarily limited to those that have all the described configurations. Furthermore, it is possible to replace a part of the configuration of one embodiment or modification with the configuration of another embodiment or modification, and it is also possible to add the configuration of another embodiment or modification to the configuration of one embodiment or modification. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment or modification with other configurations.

[0039] 1: Clinical laboratory, 2: Reception device, 3: Specimen transport device, 4: Automated analyzer, 5: Clinical laboratory information system, 6: Test delay monitoring device, 7: Network, 11: Processor (CPU), 12: Memory, 13: Storage device, 14: Input device, 15: Output device, 16: Communication device, 17: Bus, 21: Protrusion delay monitoring unit, 22: Condition setting unit, 23: Detection unit, 24: Protrusion delay notification unit, 25: User interface, 26: Recording Memory section, 31: Sample data, 32: Time management data, 33: Judgment parameters, 34: Support table, 35: Delay performance data, 36: Learning model, 41: Emergency information, 42: Test items, 45: Reception time, 46: Delivery time, 47: Elapsed time, 50: Alarm screen, 51: List, 52: Delete button, 60, 65: Outlier delay judgment parameter setting screen, 61, 66: Save button, 71: Input layer, 72: Hidden layer, 73: Output layer.

Claims

1. A test delay monitoring device for monitoring test delays in a clinical laboratory that receives samples and performs requested tests on the samples using a sample transport device and / or an automated analyzer, comprising: a detection unit; a delay notification unit that notifies of delays detected by the detection unit; and a storage unit that stores time management data, wherein the time management data stores, for each sample received by the clinical laboratory, the reception time when the sample was received by the clinical laboratory, the delivery time when the sample was delivered to the sample transport device or the automated analyzer, and the elapsed time which is the difference between the delivery time and the reception time; the detection unit sorts the samples in the order they were received by the clinical laboratory, and determines that a significant delay has occurred in the sample to be judged if the difference between the elapsed time of the sample to be judged and the maximum elapsed time of a predetermined number of comparison samples before and after it exceeds a delay standard time; and the delay notification unit notifies an alarm for the sample to be judged that has been determined by the detection unit to have a significant delay.

2. In claim 1, the detection unit sorts the samples in the order they were received by the clinical laboratory, and if the difference between the elapsed time of the first sample to be determined and the maximum elapsed time of a predetermined number of comparison samples prior to it exceeds the delay reference time, it determines that there is a possibility of a protruding delay in the first sample to be determined, and sets the next sample after the first sample to be determined as the second sample to be determined, and determines whether the difference between the elapsed time of the second sample to be determined and the maximum elapsed time of a predetermined number of comparison samples prior to it, excluding the first sample to be determined, exceeds the delay reference time, and if it does, it determines that there is a possibility of a protruding delay in the second sample to be determined, and repeats the determination process with the second sample determined to have a protruding delay as the first sample to be determined. The delay notification unit is an inspection delay monitoring device that, if the number of repetitions when it is determined that the difference between the elapsed time of the second sample to be judged and the maximum elapsed time of a predetermined number of comparison samples prior to the first sample to be judged does not exceed the delay reference time is less than a predetermined maximum number of repetitions, the detection unit determines that there is a possibility of a protruding delay occurring in multiple samples to be judged.

3. The inspection delay monitoring device according to claim 2, wherein the detection unit determines that no protrusion delay has occurred in the sample subject to determination when the number of repetitions of the determination process reaches the maximum number of repetitions, and terminates the determination process.

4. The inspection delay monitoring device according to any one of claims 1 to 3, wherein the detection unit periodically detects the protrusion delay.

5. The test delay monitoring device according to any one of claims 1 to 3, wherein the storage unit stores sample data, and the detection unit determines a population of samples to be used for detecting the protrusion delay based on the attributes of the samples stored as the sample data.

6. The inspection delay monitoring device according to any one of claims 1 to 3, wherein the storage unit stores the difference between the current time and the reception time as the elapsed time for samples that have not been transported to the sample transport device or the automated analyzer.

7. The inspection delay monitoring device according to claim 6, wherein the storage unit calculates the elapsed time excluding the period during which the sample transport device or the automatic analyzer is stopped.

8. The test delay monitoring device according to any one of claims 1 to 3, wherein the detection unit is a sample that, as the comparison sample, is included in a predetermined number of samples sorted in order of receipt, either before or after the sample to be determined, and whose receipt time falls within a predetermined time range from the receipt time of the sample to be determined.

9. An inspection delay monitoring device according to any one of claims 1 to 3, comprising a condition setting unit for setting determination parameters for detection of protrusion delay by the detection unit, wherein the determination parameters include a predetermined number of comparison samples and the delay reference time.

10. The inspection delay device according to claim 9, wherein the storage unit stores a support table in which a detection rate is calculated for the proportion of samples in which the detection unit determines that an outlier delay has occurred, by changing a predetermined number of comparison samples and the delay reference time with respect to the past time management data, and the condition setting unit sets a predetermined number of comparison samples by referring to the support table based on the specified detection rate and the delay reference time.

11. The inspection delay device according to claim 9, wherein the storage unit stores a trained learning model which takes a date and time and the number of samples received per hour as input and outputs a predetermined number of comparison samples and the delay reference time, and the storage unit is trained using delay performance data which has registered the determination result of whether or not an outlier delay occurred with respect to past time management data; and the condition setting unit sets the predetermined number of comparison samples and the delay reference time using the trained learning model.

12. A method for detecting a test delay in a clinical laboratory that receives samples and performs requested tests on the samples using a sample transport device and / or an automated analyzer, wherein the test delay monitoring device includes a detection unit, a delay notification unit that notifies of the delay detected by the detection unit, and a storage unit that stores time management data, wherein the time management data stores, for each sample received by the clinical laboratory, the reception time when the sample was received by the clinical laboratory, the delivery time when the sample was delivered to the sample transport device or the automated analyzer, and the elapsed time which is the difference between the delivery time and the reception time, wherein the detection unit sorts the samples in the order they were received by the clinical laboratory, and determines that a significant delay has occurred in the sample to be judged if the difference between the elapsed time of the sample to be judged and the maximum value of the elapsed time of a predetermined number of comparison samples before and after it exceeds a delay standard time, and the delay notification unit notifies an alarm for the sample to be judged that has been determined by the detection unit to have a significant delay.

13. In claim 12, the detection unit sorts the samples in the order they are received by the clinical laboratory, and if the difference between the elapsed time of the first sample to be determined and the maximum elapsed time of a predetermined number of comparison samples prior to it exceeds the delay criterion time, it determines that there is a possibility of a protruding delay in the first sample to be determined, and sets the next sample after the first sample to be determined as the second sample to be determined, and determines whether the difference between the elapsed time of the second sample to be determined and the maximum elapsed time of a predetermined number of comparison samples prior to it, excluding the first sample to be determined, exceeds the delay criterion time, and if it does, it determines that there is a possibility of a protruding delay in the second sample to be determined, and repeats the determination process with the second sample determined to have a protruding delay as the first sample to be determined. The delay notification unit provides an alarm notification for multiple samples to be judged that the detection unit has determined may be experiencing a protruding delay, provided that the number of repetitions in which the difference between the elapsed time of the second sample to be judged and the maximum elapsed time of a predetermined number of comparison samples prior to the first sample to be judged does not exceed the delay reference time is less than a predetermined maximum number of repetitions.

14. The inspection delay detection method according to claim 13, wherein the detection unit determines that no protrusion delay has occurred in the sample subject to determination when the number of repetitions of the determination process reaches the maximum number of repetitions, and terminates the determination process.

15. The inspection delay detection method according to any one of claims 12 to 14, wherein the detection unit periodically detects the protrusion delay.

Images