Unattended laboratory management system

Through an unattended laboratory management system, RFID and LSTM algorithms are used to realize real-time tracking of samples and automatic adjustment of quality control standards, solving the problem that existing systems cannot track and dynamic adjustment in real-time, and improving the accuracy of experimental results and Laboratory work efficiency.

CN118919041APending Publication Date: 2024-11-08QILU HOSPITAL(QINGDAO) CHEELOO COLLEGE OF MEDICINE SHANDONG UNIV +1

Patent Information

Application Number
CN202410990483.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-08

Smart Images

  • Figure CN118919041A_ABST
    Figure CN118919041A_ABST
Patent Text Reader

Abstract

The invention provides an unattended laboratory management system, which comprises the following modules: a quality control detection module used for managing quality control samples and analyzing quality control data, and when the samples enter the quality control detection module, automatically identifying sample information including sample numbers, acquisition time and patient information by scanning bar codes on the samples, the quality control detection module specifically comprises the following modules: an intelligent sample identification and tracking module for identifying bar code information on a sample by bar code scanning and realizing real-time tracking and monitoring of the sample in a detection process through an RFID tag, and a self-adaptive quality control standard module for automatically adjusting a quality control standard through historical data and statistical analysis, and a machine learning method is introduced to carry out deep mining on quality control data so as to find potential abnormal modes and trends. According to the method, the quality control standard is automatically adjusted through historical data and statistical analysis, so that the quality control standard is more scientific and reasonable, potential abnormal modes and trends can be found in time, and the reliability and accuracy of experimental results can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of management systems, and in particular to an unattended laboratory management system. Background Art

[0002] With the advancement of science and technology and the improvement of automation level, laboratory management is gradually moving towards unattended operation. Traditional laboratory management methods often rely on manual operation, which is not only inefficient but also prone to errors. In addition, the management of quality control samples and the analysis of quality control data are important links in laboratory work and are crucial to ensure the accuracy and reliability of experimental results. However, existing laboratory management systems often have many shortcomings in quality control, such as the inability to track sample status in real time, fixed quality control standards and the inability to dynamically adjust based on historical data.

[0003] Therefore, we made improvements and proposed an unattended laboratory management system. Summary of the invention

[0004] The purpose of the present invention is to address the problems that existing laboratory management systems often have many deficiencies in quality control, such as the inability to track sample status in real time, fixed quality control standards and the inability to dynamically adjust according to historical data.

[0005] In order to achieve the above-mentioned object of the invention, the present invention provides an unattended laboratory management system to improve the above-mentioned problem.

[0006] The specific application is as follows:

[0007] Includes the following modules:

[0008] The quality control detection module is used for the management of quality control samples and the analysis of quality control data. When a sample enters the quality control detection module, the sample information is automatically identified by scanning the barcode on the sample, including the sample number, collection time, and patient information;

[0009] The quality control detection module specifically includes the following modules:

[0010] Intelligent sample identification and tracking module uses barcode scanning to identify the barcode information on the sample, and uses RFID tags to achieve real-time tracking and monitoring of the sample during the testing process;

[0011] The adaptive quality control standard module automatically adjusts the quality control standards through historical data and statistical analysis, and introduces machine learning methods to conduct in-depth mining of quality control data to discover potential abnormal patterns and trends.

[0012] As the preferred technical solution of this application, the following modules are also included:

[0013] A blood collection module, including a window robot, is used to perform automatic blood collection operations;

[0014] The sample distribution module is used to receive the samples after blood collection, automatically identify the sample information, and classify the samples;

[0015] The automatic detection module receives the divided samples and sends them to the blood routine detection equipment and biochemical immunoassay detection equipment for automatic detection;

[0016] The automatic review module performs individualized automatic review on the results of automatic testing and generates a test report;

[0017] The post-processing module is used to automatically process samples after testing is completed in the laboratory.

[0018] As the preferred technical solution of this application, the adaptive quality control standard module adopts the LSTM algorithm, which specifically includes the following steps:

[0019] S1. Data collection: Collect historical quality control data, including sample number, test time, and test results. The test time is recorded as t and the test result is recorded as y. t For each sample, sort the results by detection time and construct a time series. The formula is: Y = (t1, y1), (t2, y2), ..., (t T ,y T ), T is the length of the time series;

[0020] S2. LSTM model construction and training:

[0021] The LSTM model is represented as: t , c t =LSTM(X t ,h t-1 , c t-1 ), where h t is the hidden state, c t is the cell state;

[0022] Use MSE (mean square error) as the loss function: in, is the predicted value of the LSTM model at time point t.

[0023] S3. Model evaluation and anomaly detection: For a new quality control data point X t′ , use the trained LSTM model to make predictions and get the predicted values Then calculate the anomaly score St′, the formula is:

[0024] S4. Adaptive adjustment and trend analysis of quality control standards: Dynamically adjust quality control standards based on anomaly scores and the number of consecutive anomaly points.

[0025] As a preferred technical solution of the present application, in S1, the feature vector construction and data normalization are performed, and for each time point t, the feature vector X is constructed. t , including the detection results at the current and past time points: X t =[y t-n ,y t-(n-1) , ..., y t-1 ,y t ], where n is the selected time window size.

[0026] As a preferred technical solution of this application, the data is normalized so that each feature has zero mean and unit variance. μ is the mean of the sequence and σ is the standard deviation of the sequence.

[0027] As a preferred technical solution of the present application, S3 also includes setting an abnormal threshold, setting a threshold T h , when S t′ >T h , then the data at time point t′ is judged to be abnormal.

[0028] As the preferred technical solution of this application, it also includes a user interface module for displaying various laboratory data, management information and providing a user operation interface, so that users can remotely monitor the laboratory status, query test results and adjust system settings.

[0029] As a preferred technical solution of the present application, the user interface module also includes an alarm module for triggering an alarm notification to promptly remind relevant personnel to handle abnormal situations to ensure the safe operation of the laboratory.

[0030] As a preferred technical solution of the present application, the sample allocation module automatically identifies sample information and specifically includes the following steps:

[0031] Image acquisition: Scan each sample tube to take 360-degree surround images;

[0032] Image preprocessing: remove noise and impurities in the image;

[0033] Information recognition and extraction: Use OCR technology to recognize text information on the test tube, including patient name, number, and blood collection time, and use image analysis technology to extract the color, shape, and texture characteristics of the sample.

[0034] As a preferred technical solution of the present application, the sample allocation module automatically identifies sample information and further includes the following steps:

[0035] Sample classification: Use support vector machine learning algorithm to classify the extracted sample features;

[0036] Result output and storage: The classification results are output in the form of digital signals, labels and reports for easy viewing and management by medical staff.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] In the scheme of this application:

[0039] 1. Automatically adjust quality control standards through historical data and statistical analysis to make them more scientific and reasonable, and be able to promptly discover potential abnormal patterns and trends, which helps to improve the reliability and accuracy of experimental results. Introducing machine learning methods to conduct in-depth mining of quality control data can discover hidden information and associations in the data, providing laboratory managers with more valuable decision support. By automatically identifying and tracking samples, the need for manual operation and record keeping is reduced, thereby greatly improving the efficiency and accuracy of laboratory work. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of an unattended laboratory management system provided for this application. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0042] As described in the background art, existing laboratory management systems often have many deficiencies in quality control, such as the inability to track sample status in real time, fixed quality control standards, and the inability to dynamically adjust based on historical data.

[0043] In order to solve this technical problem, the present invention provides an unattended laboratory management system, which is applied to the field of management systems.

[0044] Specifically, please refer to Figure 1 , the unattended laboratory management system specifically includes:

[0045] The quality control detection module is used for the management of quality control samples and the analysis of quality control data. When a sample enters the quality control detection module, the sample information is automatically identified by scanning the barcode on the sample, including the sample number, collection time, and patient information;

[0046] The quality control detection module specifically includes the following modules:

[0047] Intelligent sample identification and tracking module uses barcode scanning to identify the barcode information on the sample, and uses RFID tags to achieve real-time tracking and monitoring of the sample during the testing process;

[0048] The adaptive quality control standard module automatically adjusts the quality control standards through historical data and statistical analysis, and introduces machine learning methods to conduct in-depth mining of quality control data to discover potential abnormal patterns and trends.

[0049] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0050] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions in the embodiments may be combined with each other.

[0051] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0052] Example 1, please refer to Figure 1 , an unattended laboratory management system, which also includes the following modules:

[0053] A blood collection module, including a window robot, is used to perform automatic blood collection operations;

[0054] The sample distribution module is used to receive the samples after blood collection, automatically identify the sample information, and classify the samples;

[0055] The automatic detection module receives the divided samples and sends them to the blood routine detection equipment and biochemical immunoassay detection equipment for automatic detection;

[0056] The automatic review module performs individualized automatic review on the results of automatic testing and generates a test report;

[0057] The post-processing module is used to automatically process samples after testing is completed in the laboratory.

[0058] Example 2 further optimizes the unattended laboratory management system provided in Example 1. Specifically, Figure 1 As shown, the adaptive quality control standard module adopts the LSTM algorithm, which specifically includes the following steps:

[0059] S1. Data collection: Collect historical quality control data, including sample number, test time, and test results. The test time is recorded as t and the test result is recorded as y. tFor each sample, sort the results by detection time and construct a time series. The formula is: Y = (t1, y1), (t2, y2), ..., (t T ,y T ), T is the length of the time series;

[0060] S2. LSTM model construction and training:

[0061] The LSTM model is represented as: t , c t =LSTM(X t ,h t-1 , c t-1 ), where h t is the hidden state, c t is the cell state;

[0062] Use MSE (mean square error) as the loss function: in, is the predicted value of the LSTM model at time point t, and the weights and biases of the LSTM model are updated through back propagation and optimization algorithms;

[0063] Updating the weights and biases of the LSTM model through backpropagation and optimization algorithms is an iterative process that mainly includes the following steps:

[0064] S200, forward propagation:

[0065] In the LSTM model, the output of each time step is first calculated through forward propagation. This involves using the current input, the hidden state and cell state of the previous time step, and the weights and biases of the model to calculate the hidden state and cell state of the current time step;

[0066] Calculate the loss function (such as mean square error MSE) based on the output of LSTM and the actual label data;

[0067] S201, Error Back Propagation:

[0068] Starting from the loss function, calculate the gradient of the loss with respect to the LSTM output layer;

[0069] These gradients are then back-propagated back to the LSTM units to calculate the gradients of each LSTM unit’s internal states (such as hidden state and cell state);

[0070] Next, we calculate the gradients of the weights and biases, which involves taking partial derivatives with respect to each weight and bias parameter to understand their contribution to the loss function.

[0071] S202, optimization algorithm application:

[0072] Once the gradients of weights and biases are obtained, optimization algorithms (such as stochastic gradient descent SGD, Adam, etc.) can be used to update these parameters;

[0073] S203, the optimization algorithm adjusts the value of each weight and bias according to the gradient and learning rate to minimize the loss function;

[0074] S204, Iteration Update:

[0075] The above process will be iterated multiple times on the entire training dataset until a certain stopping condition is reached (such as a fixed number of iterations, the loss function reaches a preset threshold, or the performance on the validation set no longer improves significantly).

[0076] S3. Model evaluation and anomaly detection: For a new quality control data point X t′ , use the trained LSTM model to make predictions and get the predicted values Then calculate the anomaly score St′, the formula is:

[0077] S4. Adaptive adjustment of quality control standards and trend analysis: Dynamically adjust the quality control standards (i.e., threshold T h ), if anomalies are detected multiple times in succession (S t′ >T h ), which can moderately increase T h To reduce false alarms, if no abnormality is detected for a long time, T can be lowered. h To increase the sensitivity of detection, by observing the predicted value of the LSTM model and the actual value y t The long-term trend and periodic changes of quality control data can be analyzed by analyzing the changing trends of the quality control data. This helps the laboratory to give early warning and prevent potential problems.

[0078] In S1, feature vector construction and data normalization are performed. For each time point t, a feature vector X is constructed. t , including the detection results at the current and past time points: X t =[y t-n ,y t-(n-1) , ..., y t-1 ,y t ], where n is the selected time window size.

[0079] Normalize the data so that each feature has zero mean and unit variance. μ is the mean of the sequence and σ is the standard deviation of the sequence.

[0080] S3 also includes setting an abnormal threshold, setting a threshold T h , when S t′>T h , then the data at time point t′ is judged to be abnormal.

[0081] Example 3 further optimizes an unattended laboratory management system provided in Example 1 or 2. Specifically, Figure 1 As shown, the sample allocation module automatically identifies sample information and specifically includes the following steps:

[0082] Image acquisition: Scan each sample tube to take 360-degree surround images;

[0083] Image preprocessing: remove noise and impurities in the image;

[0084] Information recognition and extraction: Use OCR (Optical Character Recognition) technology to recognize text information on the test tube, including the patient's name, number, and blood collection time, and use image analysis technology to extract the color, shape, and texture characteristics of the sample.

[0085] The sample allocation module automatically identifies sample information and further comprises the following steps:

[0086] Sample classification: Use support vector machine learning algorithm to classify the extracted sample features;

[0087] Result output and storage: The classification results are output in the form of digital signals, labels and reports for easy viewing and management by medical staff.

[0088] Embodiment 4, an unattended laboratory management system provided in the above embodiments is further optimized, such as Figure 1 As shown, it also includes a user interface module for displaying various laboratory data, management information and providing a user operation interface, so that users can remotely monitor the laboratory status, query test results and adjust system settings.

[0089] The user interface module also includes an alarm module for triggering alarm notifications to promptly remind relevant personnel to handle abnormal situations to ensure the safe operation of the laboratory.

[0090] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection, an electrical connection, or communication with each other; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0091] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific embodiments, or to perform equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of the present invention, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. An unattended laboratory management system, characterized in that: Includes the following modules: The quality control detection module is used for the management of quality control samples and the analysis of quality control data. When a sample enters the quality control detection module, the sample information is automatically identified by scanning the barcode on the sample, including the sample number, collection time, and patient information; The quality control detection module specifically includes the following modules: Intelligent sample identification and tracking module uses barcode scanning to identify the barcode information on the sample, and uses RFID tags to achieve real-time tracking and monitoring of the sample during the testing process; The adaptive quality control standard module automatically adjusts the quality control standards through historical data and statistical analysis, and introduces machine learning methods to conduct in-depth mining of quality control data to discover potential abnormal patterns and trends.

2. The unattended laboratory management system according to claim 1, characterized in that: Also includes the following modules: A blood collection module, including a window robot, is used to perform automatic blood collection operations; The sample distribution module is used to receive the samples after blood collection, automatically identify the sample information, and classify the samples; The automatic detection module receives the divided samples and sends them to the blood routine detection equipment and biochemical immunoassay detection equipment for automatic detection; The automatic review module performs individualized automatic review on the results of automatic testing and generates a test report; The post-processing module is used to automatically process samples after testing is completed in the laboratory.

3. The unattended laboratory management system according to claim 1, characterized in that: The adaptive quality control standard module adopts the LSTM algorithm, which includes the following steps: S1. Data collection: Collect historical quality control data, including sample number, test time, and test result. The test time is recorded as t, and the test result is recorded as yt. For each sample, sort the results by test time and construct a time series. The formula is: Y = (t1, y1), (t2, y2), ..., (t T ,y T ), T is the length of the time series; S2. LSTM model construction and training: The LSTM model is represented as: t , c t =LSTM(X t ,h t-1 , c t-1 ), where h t is the hidden state, c t is the cell state; Use MSE (mean square error) as the loss function: in, is the predicted value of the LSTM model at time point t. S3. Model evaluation and anomaly detection: For a new quality control data point X t′ , use the trained LSTM model to make predictions and get the predicted values Then calculate the anomaly score St′, the formula is: S4. Adaptive adjustment and trend analysis of quality control standards: Dynamically adjust quality control standards based on anomaly scores and the number of consecutive anomaly points.

4. The unattended laboratory management system according to claim 3, characterized in that: In S1, feature vector construction and data normalization are performed. For each time point t, a feature vector X is constructed. t , including the detection results at the current and past time points: X t =[y t-n ,y t-(n-1) ,…,y t-1 ,y t ], where n is the selected time window size.

5. The unattended laboratory management system according to claim 4, characterized in that: Normalize the data so that each feature has zero mean and unit variance. μ is the mean of the sequence and σ is the standard deviation of the sequence.

6. The unattended laboratory management system according to claim 3, characterized in that: S3 also includes setting an abnormal threshold, setting a threshold T h , when S t′ >T h , then the data at time point t′ is judged to be abnormal.

7. The unattended laboratory management system according to claim 6, characterized in that: It also includes a user interface module for displaying various laboratory data, management information, and providing a user operation interface, so that users can remotely monitor laboratory status, query test results, and adjust system settings.

8. The unattended laboratory management system according to claim 7, characterized in that: The user interface module also includes an alarm module for triggering alarm notifications to promptly remind relevant personnel to handle abnormal situations to ensure the safe operation of the laboratory.

9. The unattended laboratory management system according to claim 2, characterized in that: The automatic identification of sample information by the sample allocation module specifically includes the following steps: Image acquisition: Scan each sample tube to take 360-degree surround images; Image preprocessing: remove noise and impurities in the image; Information recognition and extraction: Use OCR technology to recognize text information on the test tube, including patient name, number, and blood collection time, and use image analysis technology to extract the color, shape, and texture characteristics of the sample.

10. The unattended laboratory management system according to claim 9, characterized in that: The sample allocation module automatically identifies sample information and further comprises the following steps: Sample classification: Use support vector machine learning algorithm to classify the extracted sample features; Result output and storage: The classification results are output in the form of digital signals, labels and reports for easy viewing and management by medical staff.

Citation Information

Patent Citations

  • Automatic blood sampling and automatic blood detection system and method

    CN115429267A

  • Urine detection and analysis system of intelligent closestool

    CN117147544A

  • Sample automatic processing method and laboratory automatic assembly line

    CN117391113A

  • Animal tissue sampling tube traceability warehousing management method

    CN117787864A

Cited By

  • Electronic tag structure capable of preventing liquid from transferring

    CN119647510A

  • Sample image processing method and device based on multi-channel sample tube recognition model

    CN121095104A