TWI746991B - Self-learning portable fast screening urine detection system - Google Patents

Abstract

A portable quick-screening urine detection system with self-learning, which includes an in vitro detector for inserting a special quick-screening test paper for urine testing and shooting images, and then transmitting it to an electronic device for comparison and analysis, wherein the quick-screening test paper It has at least one observation area, at least one to 14 test groups, and an identification tag, which can be used to read the algorithm corresponding to the basic data of the urine test item. The electronic device, which has an analysis software and a self-learning unit, can compare the judgment data with a data parameter downloaded from the cloud to generate a measurement value for each test group, plus AI artificial Smart learning and display image correction, so that the subject's urine physiological response to the test item can produce an accurate test report.

Description

Self-learning portable fast screening urine detection system

The present invention relates to the technical field of in vitro detection systems, in particular to a portable quick-screen urine detection system with self-learning.

With the continuous advancement of detection technology, there are many quick screening test devices or quick screening test papers that can obtain the results instantly, such as pregnancy test sticks, flu, and urine quick screening test papers. Among them, the urine test paper uses rapid chemical analysis. In order to obtain a variety of physiological values, a plastic sheet is attached to the urine to interpret white blood cells (Leukocytes), nitrite (Nitrite), urobilinogen (Uro-bilinogen), microalbumin (microalbumin), creatinine (Creatinine), Calcium (Ca), Vitamin C (Vitamin C), Protein (Protein), pH, Occult blood, Specific gravity, Ketone, Bilirubin and Glucose (Glucose) reagent area. By using human or animal urine to react with the reagent area of the urine test paper, the test results of related physiological conditions can be quickly obtained.

Although the current urine quick screening test paper advertises simple and easy-to-use operation, it can only tell the subjects about the urine test response based on the results. It cannot reflect the accurate classification and grading test results, and cannot provide more subjects. Information. Especially during testing, if the concentration of the test substance contained in the test object is low or too turbid, the result of the quick screening test paper after the test will not be easily recognized by the naked eye, resulting in misjudgment and cognition, which will be provided to the testee. The information of this kind can easily lead to physical and mental panic of the subject and family members, or cause unnecessary medical waste.

Although, at present, it can be overcome by capturing the image of the urine quick screening test paper after the test, and then analyzing it through a specific device. For example, there are special imaging equipment for testing in the hospital, but this special equipment is huge and expensive, and it can be easily purchased or obtained by non-ordinary people, and the testees have to spend a lot of time in the hospital during testing; in addition, there are also available in the market. Some portable electronic devices can capture the displayed image of the quick screening test paper after the test, and then send it to a computing server or a compatible device such as a smart phone or computer to obtain the test result; but the disadvantage is that it can When using a portable electronic device, it’s troublesome to set up a stand for shooting, and it’s in an open space for shooting. The shooting images are likely to be different due to different environments, which may lead to errors in interpretation. In addition, different styles of Portable electronic devices have different pixels, focusing distances, or sensitivity, and the captured images will also have gaps. Moreover, the device itself does not have the function of self-learning intelligence, and the concentration of the test object contained in the test object is treated face-to-face When the turbidity is different, it is easy to cause errors in interpretation and cause a lot of inconvenience.

In particular, the existing quick screening test papers are disposable products. Each batch of built-in test papers is of different quality, which is easily affected by differences in materials and manufacturing processes, resulting in differences between batches, and it is also easy to cause the interpretation results to lose accuracy. sex. Although the existing quick screening test papers all have an identification label, the identification label of the conventional quick screening test paper is usually only used to provide basic information such as the name, serial number, and manufacturing date of the related product. In fact, it cannot be used Effectively solve the shortcomings in use of the test paper batch production differences that lead to different test results and loss of accuracy.

In order to solve the above-mentioned problems, a self-learning portable quick-screen urine detection system with self-learning, which can be carried around, know the reaction result and the degree of reaction in real time, was developed in the present invention.

The purpose of the present invention is to provide a portable quick screening urine detection system with self-learning, including: an in vitro detector, a quick screening test paper, and an electronic device, wherein

The extracorporeal detector has a closed environment with a specimen box, a shooting image processor, a transmission interface control processor, and a power supply unit. The specimen box is provided with a specimen insertion port for inserting a specimen. The quick screening test paper to be tested;

The quick screening test paper is dedicated to urine testing. It has at least one observation area and an identification label, wherein there are at least 1 to 14 test groups in the observation area, and the identification label has a first read message, at least Including the algorithm corresponding to the quick screening test paper and its test paper batch, material and basic information of urine test items; and

The electronic device has an analysis software and a self-learning unit in it. The analysis software is used to receive the display image of the quick screening test paper after being tested from the transmission interface control processor to generate a parameter package and a Judgment data; the self-learning unit can judge the ambiguity of the displayed image and correct it with AI artificial intelligence learning;

Thereby, through the test paper type and basic data of urine test items conveyed in the first read message, the judgment data can be compared with a data parameter downloaded from the cloud to generate a measurement value for each test group , Coupled with AI artificial intelligence learning display image correction, so that the subject's urine physiological response to the test item can produce an accurate test report.

To achieve the above purpose, the technical means of the present invention will set up a comparison database in the cloud. The comparison database has a variety of different contents based on the first read information of the identification tag and the learning record of the self-learning unit. The detection image results of the physiological response of the degree, and the detection image results generate the data parameters through the cloud computing. When the user opens the analysis software in the electronic device, the analysis software will automatically download the data parameters and store them in the In the analysis software; when the detection is completed, the electronic device receives the display image transmitted from the in vitro detector, and through the analysis software, a parameter package with physiological response parameters and display images will be generated respectively, and a parameter package of the current detection result To judge the data, the analysis software will further send the parameter package to the cloud for data storage and collection. At the same time, the judgment data and the data parameter will be compared with the degree of response according to the type of detection, plus artificial intelligence Smart learning and display image correction, so that the subject's urine physiological response to the test item can produce an accurate test report.

In order to facilitate a further understanding and understanding of the technical means and operation process of the present invention, examples are given in conjunction with the diagrams, and the detailed description is as follows.

Please refer to Figures 1 to 3, the present invention relates to a portable quick screening urine detection system with self-learning, including an in vitro detector 10, a quick screening test paper 20, and an electronic device 30, in which:

The extracorporeal detector 10 has a closed environment with a specimen box 15, a shooting image processor 16, a transmission interface control processor 17, and a power supply unit 18. The specimen box 15 is externally provided with a specimen insert Port 101 is used to insert a quick screening test paper 20 to be tested;

The quick screening test paper 20 is dedicated for urine testing. It has an observation area 21 and an identification label 23. The observation area 21 has at least 1 to 14 test groups 22, and the identification label 23 has a first The read information 231 includes at least the algorithm corresponding to the quick screening test paper 20, its test paper batch, material, and basic data of urine test items. The identification label 23 can be a one-dimensional bar code or a two-dimensional bar code. Not limited to this; and

The electronic device 30 has an analysis software 31 and a self-learning unit 32 therein. The analysis software 31 is used to receive the display image of the quick screening test paper 20 after being tested from the transmission interface control processor 17, so as to Generate a parameter package and a judgment data; the self-learning unit 30 can judge the ambiguity of the displayed image and correct it with AI artificial intelligence learning;

Thereby, through the test paper type and basic data of urine test items conveyed in the first read message 231, the judgment data can be compared with a data parameter downloaded from the cloud 40 to generate one of the test groups The measured value, together with the analysis software 31 and the self-learning unit 32 in the electronic device 30, can perform AI artificial intelligence learning and display image correction, so that the testee's urine physiological response test item can generate an accurate test report.

With the above effective design, the present invention has the following advantages when testing and applying:

First, the first read message 231 of the identification tag 23 can be read by the in vitro detector 10, and the mobile phone APP connected via Bluetooth is downloaded to the cloud 40 and the algorithm represented by the first read message 231 is applied. , Perform the interpretation of the quick screening test paper 20.

Application note: The human or animal urine is reacted with multiple urine test groups 22 of the quick screening test paper 20, and then the quick screening test paper 20 is placed in the in vitro detector 10, and the in vitro detector 10 is used to photograph and scan an object to be tested The images of the test group 22 and the identification tag 23 on the quick screening test paper 20 are connected to an electronic device 30 in a wired or wireless manner, and the images are captured into the electronic device 30. When the in vitro detector 10 initially obtains the first reading message 231 by reading the identification tag 23 on the quick screening test paper 20, it can clearly know the type of test paper and the basic data of the urine test item, and it will read and download it from the cloud 40 The algorithm corresponding to the first read message 231 of the identification tag 23 is sent to an electronic in vitro detector 10, and the comparison result analysis of each test group 22 is performed to generate the detection result in real time.

Second, the electronic device 30 has a self-learning unit 32, so that when urine is detected, it can be intelligently learned to perform measurement calibration.

Application note: When the urine test result of the quick screening test paper 20 is interpreted through the analysis software 31 of the electronic device 30, the analysis software 31 performs color, brightness, and saturation on the displayed image of each urine test group 22 of the quick screening test paper 20 Interpretation, and read a parameter through the identification tag 20, and compare and analyze the chromaticity, brightness, and saturation of the image displayed on the test paper with the parameter to obtain the interpretation result of each urine test group 22 and store it. In addition, the electronic device 30 can also perform learning correction training with AI artificial intelligence through the self-learning unit 32, so that accurate semi-quantitative result judgments can be obtained in different fuzzy judgments.

Third, the corresponding content of the first read message 231 of the identification tag 23 can also be extended to reduce the difference between test paper batches and to control the validity period of the test paper.

Application note: Mainly because the results of the observation area 21 of the general quick screening test paper 20 may be affected by differences in materials and manufacturing processes, which may cause differences between batches, which will lead to differences in the system's interpretation of the results, so it can be used at this time The first read message 231 of the identification tag 23 is read to load different operation parameters, and the result of calibration and proofreading of the observation area 22 is achieved. Furthermore, the identification tag 23 of the quick screening test paper 20 can be used as the cloud 40 to record the expiration date of the test paper to limit the use of the expired test paper, and avoid possible misinterpretation of the expired test paper to confuse the user's physiological value interpretation.

As a result, as shown in Figures 1 to 4, when the system is implemented to perform the detection with the recognition tag compensation function, when the electronic device 30 captures the displayed images of each urine test group 22 of the quick screening test paper 20 for analysis The software 31 processes the displayed images of each urine test group 22. By reading the identification label 23 on the quick screening test paper 20, when the first reading message 231 is initially obtained, the type of the test paper and the basic information of the urine test items are clearly known , It reads from the cloud 40 and downloads the algorithm corresponding to the first read message 231 of the identification tag 23, compares the judgment data with a data parameter downloaded from the cloud 40, and compares each urine test group 22 Generate a measurement value, and use the self-learning unit 32 in the electronic device 30 to perform AI artificial intelligence learning on the response results of each urine test group 22 input and read, so that the turbidity of the displayed image can be appropriately adjusted and corrected according to The parameter index table pre-implanted in the identification label 23 matches the color of the urine test paper for chromaticity, brightness, and saturation, so that an accurate test report can be generated for the tested item of the urine physiological response of the subject. In order to achieve the detection items at least include but not limited to: urine test paper white blood cells (Leukocytes), nitrite (Nitrite), urobilinogen (Uro-bilinogen), microalbumin (microalbumin), creatinine (Creatinine), calcium ( Ca), Vitamin C, Protein, pH, Occult blood, Specific gravity, Ketone, Bilirubin and Glucose )...Etc. The semi-quantitative readings are displayed on an electronic device 30 for data storage.

More specifically, the system of the present invention described above can be applied to various convenient and portable extracorporeal detectors 10, wherein the extracorporeal detector 10 and the electronic device 30 can be connected in a wired or wireless manner, and the electronic device 30 can be a smart phone, a tablet computer, a desktop computer, a notebook computer, etc.

As shown in Fig. 1 and Fig. 5, the shape of the in vitro detector 10 of the present invention can be diversified, and its appearance is not limited to this, and it can also be designed in other different shapes.

Taking Fig. 5 as an example, a face of the extracorporeal detector 10 has a control panel 11, the control panel 11 is provided with a power button 111 and an image capture button 112, and the extracorporeal detector 10 has a specimen insertion port 101 and a USB connection slot 14, the specimen insertion port 101 and the USB connection slot 14 are respectively provided on different sides, the in vitro detector 10 has a viewing panel 13, and the viewing panel 13 is provided with a power indicator light 131 A wireless connection indicator light 132, a specimen display light 133, and a transmission display light 134 are used to remind the user whether the extracorporeal detector 10 is activated and operating conditions.

Please refer to Figures 1, 3, and 5, during application, the tested quick screening test paper 20 can be inserted into the sample box 15 from the sample insertion port 101, and when the tested quick screening test paper 20 touches When a sensor 151 is provided at one end of the specimen box 15, the specimen display light 133 of the inspection panel 13 will be activated, so that the user will know that the quick screening test paper 20 after the test is correctly inserted. And image capture can be started. The shooting image processor 16 is arranged on one side of the specimen box 15 and is electrically connected to the image capturing button 112, and the shooting image processor is controlled through the image capturing button 112 16 Capture the displayed image of the quick screening test paper 20 in the specimen box 15 after being tested, and send the displayed image to the transmission interface control processor 17, which can be wireless Bluetooth transmission Or wifi transmission, or wired transmission through the USB connection slot 14 electrically connected to it, and directly output the displayed image to the electronic device 30 for analysis in a wireless or wired manner. At the same time, the viewing panel 13 The transmission indicator light 134 will be activated.

Please refer to Figures 1 and 2, the electronic device 30 has an analysis software 31. When the detection is completed, the electronic device 30 receives the display image transmitted by the extracorporeal detector 10, and the analysis software 31 displays the image After performing back-removing, de-noising, and sharpness processing, a parameter package and a judgment data are respectively generated to interpret the displayed images of each test group 22. At the same time, a self-learning unit 32 in the electronic device 30 can be used to target The ambiguity of the displayed image is interpreted, corrected by AI artificial intelligence learning, and the relevant interpretation and learning correction records are transmitted to the cloud 40 of the present invention.

The cloud 40 is provided with a comparison database corresponding to the analysis software 31. The comparison database contains the corresponding content of the calibration record according to the first read information 231 of the identification tag 23 and the learning calibration record of the self-learning unit 32. The displayed images of different levels of physiological response testing. Due to the different levels of physiological response, the concentration of the components to be detected in the sample will be different, so the results displayed on the quick screening test paper 20 after the test will also be different in depth. When the user turns on the electronic device 30 in a wired environment, the analysis software 31 will automatically download the latest data parameters from the cloud 40, so that when the electronic device 30 is in a wireless situation, the ratio of the detection results can still be performed right.

After the cloud 40 receives the parameter package sent from the analysis software 31, it will classify and archive the data according to the physiological response parameters contained in the parameter package and the images of each test group 22. If the provided image is new, The control database will be updated so that the content corresponding to the first read message 231 on the identification label 23 of the quick screening test paper 20 can be downloaded by the analysis software 31 so that when the present invention is applied, the first read Take the type of test paper and basic data of urine test items conveyed in the message 231, and let the analysis software in the electronic device 30 generate a parameter package and a judgment data correspondingly, and then the judgment data and a data parameter downloaded from the cloud 40 Comparison to generate one of the measured values of each test group 22;

In addition, during the interpretation, the algorithm can be downloaded through the cloud 40 to compare a measurement value of each of the aforementioned test groups 22 with the measurement value of the learning calibration record of the system self-learning unit 32, and the calculation is performed based on the difference. The method generates compensatory corrections to the measured values of each test group 22. As a result, an accurate test report can be generated for the subject’s physiological response to the test item, which will not be affected by the different materials of the test strips and the measured turbidity, so as to achieve the functions of early detection, early prevention and planning.

The in vitro detection system of the present invention captures the displayed image of the quick screening test paper 20 after the test through the in vitro detector 10 in a closed environment. There is no need to worry about the difference in the environment or the shooting equipment that will affect the results of the image capture. The user can carry the extracorporeal detector 10 with them to perform urine testing anytime and anywhere; and send the displayed image to the electronic device 30 for analysis of the results, which can save the procedure of uploading data by the user; moreover, the extracorporeal The detection system uses the analysis software 31 and the self-learning unit 32 in the electronic device 30 to periodically download and upload updated data parameters from the cloud 40. Therefore, the display image can be analyzed and compared under wired or wireless conditions. In addition to knowing the physiological response results of various items in urine testing, it can also analyze the degree of physiological response and provide suggestions to achieve the functions of early detection, early prevention and planning.

However, the above are only preferred embodiments of the present invention, and should not be used to limit the scope of implementation of the present invention; therefore, all simple equivalent changes and modifications made in accordance with the scope of the patent application of the present invention and the content of the description of the invention , Should still fall within the scope of the invention patent.

10: In vitro detector 101: Specimen insertion port 11: Control Panel 111: Power button 112: Image capture button 13: View panel 131: Power indicator 132: Wireless connection indicator 133: Specimen display light 134: Transmission indicator light 14: USB connection slot 15: Specimen box 151: Sensor 16: shooting image processor 17: Transmission interface control processor 18: Power supply unit 20: Quick screening test paper 21: Observation area 22: Test group 23: Identify the label 231: First read message 30: Electronic device 31: Analysis software 32: Self-study unit 40: Cloud

Figure 1 is a configuration diagram of the system of the present invention. Figure 2 is a plan view of the quick screening test paper of the present invention. Figure 3 is the internal structure diagram of the in vitro detector of the present invention. Figure 4 is a flow chart of the test application of the present invention. Figure 5 is an external view of another in vitro detector of the present invention.

10: In vitro detector

101: Specimen insertion port

20: Quick screening test paper

21: Observation area

22: Test group

23: Identify the label

231: First read message

30: Electronic device

31: Analysis software

32: Self-study unit

40: Cloud

Claims (4)

A portable quick screening urine detection system with self-learning, comprising: an in vitro detector, a quick screening test paper, and an electronic device, wherein the in vitro detector has a closed environment and is provided with a specimen box, a A photographing image processor, a transmission interface control processor and a power supply unit. The sample box is provided with a sample insertion port for inserting the quick screening test paper to be tested. The electronic device has an analysis software in it The analysis software is used to receive the display image of the quick screening test paper after being tested from the transmission interface control processor to generate a parameter package and a judgment data; the electronic device is a smart phone, a tablet, a desk One of a PC and a notebook computer; and the transmission interface control processor can respectively perform wired transmission through a USB connection slot, and wireless transmission through Bluetooth or WIFI; it is characterized in that the external detector is inserted into the The quick screening test paper is dedicated to urine testing. It has at least one observation area and an identification label, wherein the observation area has at least 1 to 14 test groups, and the identification label has a first read message, including at least The quick screening test paper has an algorithm corresponding to the test paper batch, material, and basic data of urine testing items; the electronic device has a self-learning unit that can determine the ambiguity of the displayed image and use AI artificial intelligence learning is used to calibrate. Through the type of test paper and basic data of urine test items conveyed in the first read message, the judgment data can be compared with a data parameter downloaded from the cloud to generate each test group One of the measured values, coupled with AI artificial intelligence learning and display image correction, allows the subject's urine physiological response to the test item to generate an accurate test report; and; the cloud has a built-in control corresponding to the analysis software A database, the control database is based on the first read information of the identification tag and the content corresponding to the learning record of the self-learning unit, and has a plurality of different degrees of physiological response detection and display images, and displays the images The image generates the data parameter through calculation and classification, and the cloud will receive the physiological response parameter with the detection result in the parameter package and the image of each test group, and store and collect the calibration data by AI artificial intelligence learning. The self-learning portable quick-screen urine detection system as described in claim 1, wherein one of the external detectors has a display panel, on which a power indicator light, a wireless connection indicator light, and a specimen are respectively provided Indicator light and a transmission indicator light. The self-learning portable fast screening urine detection system as described in claim 1, wherein the corresponding content of the first read message of the identification tag is further used to reduce the difference between test strips and provide different calculation parameters for downloading for proofreading Observe the results of the area. As described in claim 1, a portable quick-screen urine detection system with self-learning, wherein the corresponding content of the first read message of the identification tag is further provided as the expiration control of the test paper.

Images