CN118817669B - Intelligent in-vivo ingredient metabolism detection system - Google Patents
Intelligent in-vivo ingredient metabolism detection system Download PDFInfo
- Publication number
- CN118817669B CN118817669B CN202411296611.5A CN202411296611A CN118817669B CN 118817669 B CN118817669 B CN 118817669B CN 202411296611 A CN202411296611 A CN 202411296611A CN 118817669 B CN118817669 B CN 118817669B
- Authority
- CN
- China
- Prior art keywords
- body fluid
- detection
- images
- image
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 161
- 230000004060 metabolic process Effects 0.000 title claims abstract description 25
- 238000001727 in vivo Methods 0.000 title description 13
- 239000004615 ingredient Substances 0.000 title description 2
- 210000001124 body fluid Anatomy 0.000 claims abstract description 231
- 239000010839 body fluid Substances 0.000 claims abstract description 231
- 238000004458 analytical method Methods 0.000 claims abstract description 57
- 238000013210 evaluation model Methods 0.000 claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 6
- 201000010099 disease Diseases 0.000 claims description 44
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 44
- 239000003153 chemical reaction reagent Substances 0.000 claims description 23
- 238000003384 imaging method Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 13
- 238000011534 incubation Methods 0.000 claims description 12
- 238000010438 heat treatment Methods 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 7
- 238000005094 computer simulation Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 2
- 239000012916 chromogenic reagent Substances 0.000 description 25
- 238000000034 method Methods 0.000 description 18
- 230000008569 process Effects 0.000 description 12
- 239000000470 constituent Substances 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 208000000913 Kidney Calculi Diseases 0.000 description 4
- 208000009525 Myocarditis Diseases 0.000 description 4
- 206010029148 Nephrolithiasis Diseases 0.000 description 4
- 206010012601 diabetes mellitus Diseases 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 201000006938 muscular dystrophy Diseases 0.000 description 4
- 201000008383 nephritis Diseases 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 208000017667 Chronic Disease Diseases 0.000 description 2
- 206010020772 Hypertension Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000004064 dysfunction Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000005027 intestinal barrier Anatomy 0.000 description 2
- 230000007358 intestinal barrier function Effects 0.000 description 2
- KJFMBFZCATUALV-UHFFFAOYSA-N phenolphthalein Chemical compound C1=CC(O)=CC=C1C1(C=2C=CC(O)=CC=2)C2=CC=CC=C2C(=O)O1 KJFMBFZCATUALV-UHFFFAOYSA-N 0.000 description 2
- 208000001076 sarcopenia Diseases 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000004580 weight loss Effects 0.000 description 2
- 208000016261 weight loss Diseases 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 210000001175 cerebrospinal fluid Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 210000002381 plasma Anatomy 0.000 description 1
- 239000012286 potassium permanganate Substances 0.000 description 1
- 210000003296 saliva Anatomy 0.000 description 1
- 210000002966 serum Anatomy 0.000 description 1
- 238000002636 symptomatic treatment Methods 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
- 210000002700 urine Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/75—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated
- G01N21/77—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator
- G01N21/78—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator producing a change of colour
Landscapes
- Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Plasma & Fusion (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
本申请涉及一种智能体内成分代谢检测系统，属于医疗系统的技术领域，该系统包括移动终端、检测设备和服务器；移动终端用于录入并存储患者的特征信息；检测设备与移动终端绑定，用于对待测体液进行指定时间的恒温孵育后得到检测数据并将检测数据上传至服务器；服务器被配置为：获取检测数据；基于预设的分析模型对检测数据进行处理后获得与待测体液对应的指标浓度值；基于预设的评价模型对指标浓度值、同一患者不同检测时间的历史检测数据、患者的特征信息进行处理后得到分析结果；将分析结果发送至移动终端；移动终端用于获取并展示分析结果。本申请能够综合检测数据、历史检测数据和特征信息得出分析结果，使得到的分析结果更加全面和准确。
The present application relates to an intelligent body component metabolism detection system, which belongs to the technical field of medical systems. The system includes a mobile terminal, a detection device and a server; the mobile terminal is used to input and store the patient's characteristic information; the detection device is bound to the mobile terminal, and is used to obtain the detection data after the body fluid to be tested is incubated at a constant temperature for a specified time and upload the detection data to the server; the server is configured to: obtain the detection data; obtain the index concentration value corresponding to the body fluid to be tested after processing the detection data based on a preset analysis model; obtain the analysis result after processing the index concentration value, the historical detection data of the same patient at different detection times, and the patient's characteristic information based on a preset evaluation model; send the analysis result to the mobile terminal; the mobile terminal is used to obtain and display the analysis result. The present application can obtain the analysis result by integrating the detection data, historical detection data and characteristic information, so that the obtained analysis result is more comprehensive and accurate.
Description
Technical Field
The application relates to the technical field of medical systems, in particular to an intelligent in-vivo constituent metabolism detection system.
Background
The in vivo constituent metabolism detection technology can be used for detecting indexes of indicators of chronic diseases such as medical weight loss, sarcopenia, hypertension, intestinal barrier dysfunction and the like.
In the related art, when in-vivo constituent metabolism detection is performed, a single detection index can be detected by collecting a sample of a patient through detection equipment, and a single analysis result is obtained. Since single analysis results can only analyze single-collected samples, the resulting report lacks systematicness. To compensate for the lack of systematics, doctors can more comprehensively analyze and judge according to personal experience by looking up past analysis results. The diagnosis result is inaccurate because the personal subjectivity is strong when the doctor judges according to personal experience. Moreover, the patient needs to go to the hospital for detection every time, and the social manpower, financial resources and material resources are greatly consumed on the premise of limited medical resources.
Disclosure of Invention
In order to enable analysis results to be more comprehensive and accurate, the application provides an in-vivo constituent metabolism detection system.
The application provides an in-vivo constituent metabolism detection system of an intelligent agent. The system comprises a mobile terminal, a detection device and a server;
The mobile terminal is used for inputting and storing characteristic information of a patient;
The detection equipment is bound with the mobile terminal and is used for obtaining detection data after constant temperature incubation for a specified time is carried out on body fluid to be detected and uploading the detection data to the server, wherein the body fluid to be detected comprises a chromogenic reagent and body fluid of a patient;
The server is configured to:
Acquiring the detection data;
processing the detection data based on a preset analysis model to obtain an index concentration value corresponding to the body fluid to be detected;
processing the index concentration value corresponding to the body fluid to be tested, the historical detection data of the same patient at different detection times and the characteristic information of the patient based on a preset evaluation model to obtain an analysis result;
sending the analysis result to the mobile terminal;
The mobile terminal is used for acquiring and displaying the analysis result.
Through adopting above-mentioned technical scheme, check out test set can carry out constant temperature incubation in order to obtain detection data to the body fluid that awaits measuring, and then comprehensive analysis detection data, history detection data and characteristic information obtain analysis result, make the analysis result that the patient obtained more comprehensive and accurate, and the patient can also look over the analysis result through mobile terminal directly perceivedly simultaneously, reduces the manpower consumption.
Optionally, the detection device is provided with a plurality of groups, each of the plurality of groups of detection devices incubates with a plurality of types of body fluids to be detected at constant temperature, and the concentrations of the chromogenic reagents corresponding to the same types of body fluids to be detected in different detection devices are different.
Through adopting above-mentioned technical scheme, set up the different concentrations of the corresponding chromogenic reagent of same kind's body fluid that awaits measuring among multiunit check out test set and the different check out test set to can observe the reaction condition of patient's body fluid in different concentrations, different kinds of chromogenic reagent better, ensure the accuracy of the analysis result that obtains.
Optionally, the server is configured to:
Acquiring multiple groups of body fluid images, wherein each group of body fluid images comprises multiple body fluid images with the same type and different concentrations of color-developing reagents;
Determining an abscissa of a preset coordinate system according to the concentration of the chromogenic reagent of the multiple body fluid images in each group of body fluid images;
determining an ordinate in a preset coordinate system according to the color depth of the color developing reagent of the multiple body fluid images in each group of body fluid images;
Combining the abscissa and the ordinate determined by the same body fluid image into a standard coordinate;
And connecting a plurality of standard coordinates corresponding to each group of body fluid images in series to obtain a standard curve formula, wherein each group of body fluid images corresponds to one standard curve formula.
By adopting the technical scheme, after the standard curve formula is established according to the plurality of body fluid images in each group of body fluid images, the established standard curve formula is convenient to directly call in the follow-up, and the index concentration value of each body fluid image is rapidly obtained through the mapping relation between the concentration of the color reagent and the gray value contained in the standard curve formula. And when the standard curve formula is established, the standard curve formula is obtained by arranging a plurality of body fluid images, wherein the concentration of the color reagent in each body fluid image is different, so that the accuracy of establishing the obtained standard curve formula is ensured.
Optionally, the server is further configured to:
the method comprises the steps of determining the ordinate in a preset coordinate system according to the color depth of a color developing reagent of a plurality of body fluid images in each group of body fluid images, wherein the determination process of the ordinate of any body fluid image in each group of body fluid images is as follows:
Preprocessing the body fluid image;
judging whether the color depth of the body fluid image after pretreatment meets the preset uniformity degree;
if yes, determining a corresponding ordinate according to the gray value of the body fluid image;
if not, a computational model is employed to predict the ordinate of the body fluid image.
By adopting the technical scheme, the method and the device are provided with a plurality of algorithms capable of determining the ordinate of each body fluid image under different scenes, so that the application scene of the method and the device is widened.
Optionally, the server is further configured to:
the employing a computational model to predict an ordinate of the body fluid image includes:
Dividing the body fluid image into a plurality of sub-regions according to color depth;
dividing the subareas into different chromaticity grades according to the color depth of the subareas, wherein the different chromaticity grades correspond to different gray values;
calculating a plurality of classification probabilities according to the division result of the subregions of the body fluid image, wherein one classification probability corresponds to one chromaticity grade, and the classification probability is used for representing the probability that the body fluid image belongs to the corresponding chromaticity grade;
selecting one classification probability with the largest classification probability as a target probability;
and matching according to the gray value corresponding to the target probability to obtain a corresponding ordinate.
By adopting the technical scheme, the server can further analyze the body fluid image, specifically, split the body fluid image into a plurality of subareas, then divide the subareas into different chromaticity grades respectively, and finally select one with the largest classification probability as the target probability for representing the gray value of the body fluid image, thereby ensuring the accuracy of the obtained standard curve formula.
Optionally, the server is further configured to:
the calculating to obtain a plurality of classification probabilities according to the division result of the subareas of the body fluid image comprises the following steps:
calculating the area of the subarea falling into each chromaticity grade;
And calculating the ratio of each area to the total area of the body fluid image, wherein the ratio is taken as the classification probability of the body fluid image belonging to the chromaticity grade.
Optionally, the server is further configured to:
The method for determining the ordinate in the preset coordinate system according to the color depth of the color developing reagent of the multiple body fluid images in each group of body fluid images comprises the step of predicting the ordinate of the body fluid images in the preset coordinate system by adopting a neural network model.
By adopting the technical scheme, the application is provided with a plurality of algorithms to predict the ordinate of each body fluid image, so that the body fluid image can be selected by patients in different scenes.
Optionally, the server is further configured to:
Judging whether the color depth of the body fluid image after pretreatment meets the preset uniformity degree or not comprises the following steps:
calculating gray value differences of areas with different color depths in the body fluid image;
When the maximum value of the gray value differences is smaller than a first preset value, determining that the color depth of the body fluid image meets the preset uniformity degree, or
Counting the number of areas with different color depths in the body fluid image;
and when the number is smaller than a second preset value, determining that the color depth of the body fluid image meets the preset uniformity degree.
By adopting the technical scheme, the body fluid image which does not meet the preset uniformity degree is determined according to the fact that the color depth difference is large or the color partition is serious, so that the selected algorithm for determining the abscissa is more adaptive to the body fluid image, and the accuracy of the obtained standard curve formula is ensured.
Optionally, the server is further configured to:
After the standard curve formula is obtained, the method comprises the following steps:
According to the type of the body fluid of the patient in the body fluid to be detected, the type and the concentration of the chromogenic reagent, a preset analysis model is adopted to match a standard curve formula corresponding to each body fluid image from a plurality of standard curve formulas;
and obtaining an index concentration value corresponding to the body fluid image according to the standard curve formula.
Optionally, the detection device comprises a detection card and an incubator;
the detection card is provided with a plurality of detection holes for accommodating body fluid to be detected;
The incubator is internally provided with a positioning groove, an imaging hole and a heating module, the imaging hole is positioned at the bottom of the positioning groove, the bottom of the imaging hole is provided with a cmos camera, the imaging sight line of the cmos camera is arranged towards the opening end of the imaging hole, and the detection card is clamped with the positioning groove;
The heating module is arranged close to the positioning groove and is used for carrying out constant-temperature incubation on the body fluid to be detected for a designated time, and the body fluid to be detected after constant-temperature incubation for the designated time is imaged by the cmos camera to obtain detection data.
Through adopting above-mentioned technical scheme, a plurality of detection holes on the detection card can place different grade type chromogenic reagent, and put into the same patient's body fluid in different chromogenic reagents to realize the effect of single to the multiple body fluid that awaits measuring analysis, be favorable to analyzing detection data more comprehensively.
In summary, the present application includes at least one of the following beneficial technical effects:
On the one hand, the detection equipment can incubate the body fluid to be detected at a constant temperature to obtain detection data, the server comprehensively analyzes the detection data, the historical detection data and the characteristic information to obtain analysis results, so that the analysis results obtained by a patient are more comprehensive and accurate, and meanwhile, the patient can also visually check the analysis results through the mobile terminal, and the manpower consumption is reduced. On the other hand, after a standard curve formula is established according to a plurality of body fluid images in each group of body fluid images, the established standard curve formula is conveniently and directly called later, and the index concentration value of each body fluid image is rapidly obtained through the mapping relation between the concentration of the color reagent and the gray value contained in the standard curve formula, so that rapid analysis is realized and an analysis result is obtained.
Drawings
FIG. 1 is a system schematic diagram of an in-vivo constituent metabolism detection system according to an embodiment of the present application.
Fig. 2 is a schematic structural diagram of a test card according to an embodiment of the present application.
Fig. 3 is a schematic structural view of an incubator according to an embodiment of the present application.
Fig. 4 is an exemplary diagram of classifying body fluid images according to an embodiment of the present application.
FIG. 5 is an exemplary diagram of a standard curve formula derived from concatenating multiple standard coordinates in accordance with an embodiment of the application.
The reference numerals indicate 1, a box body, 2, a positioning groove, 3, an imaging hole, 4, an upper cover, 5, a detection card, 6, a detection hole, 7, detection equipment, 8, a server, 9 and a mobile terminal.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings 1 to 5 and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The in vivo constituent metabolism detection technology can be used for detecting the indexes of the chronic diseases such as medical weight loss, sarcopenia, hypertension, intestinal barrier dysfunction and the like. In the actual detection process, a certain body fluid of a patient needs to be sampled, the collected sample is analyzed, and a doctor can perform symptomatic treatment on the patient according to the analysis result. However, since the analysis results can only be performed on a single sample, the resulting report is not systematic. Therefore, it is often necessary to continuously monitor the metabolic index of the internal components of the patient at regular intervals, and then determine the next treatment regimen of the patient according to the trend of the change of each combination index in a certain period and the clinical significance and the clinical indication of the patient. At present, doctors can only analyze and judge according to personal experience by checking past analysis results. However, the experience of different doctors is different, and the judgment is also different, so that the problem of inaccurate diagnosis results can easily occur, thereby causing the delay of the illness state of patients.
The embodiment of the application discloses an in-vivo component metabolism detection system, which can enable analysis results to be more comprehensive and accurate.
Referring to fig. 1 to 3, the in-agent constituent metabolism detection system includes a mobile terminal 9, a detection device 7, and a server 8.
The detection device 7 is used for binding the mobile terminal 9 and obtaining detection data after incubating the body fluid to be detected for a specified time at constant temperature, wherein the body fluid to be detected comprises a chromogenic reagent and a patient body fluid, the patient body fluid is one of urine, serum, whole blood, cerebrospinal fluid, blood plasma and saliva, the display reagent such as potassium permanganate, phenolphthalein and the like can react with the patient body fluid and develop the chromogenic reagent, and the detection device 7 is also used for uploading the obtained detection data to the server 8.
The detection device 7 comprises two parts, a detection card 5 and an incubator.
The detection card 5 is provided with a plurality of detection holes 6 for containing the body fluid to be detected, and the body fluid of the patient placed in all detection holes 6 on the same detection card 5 is the same, but the types of the chromogenic reagents are different, for example, six detection holes 6 exist, and the chromogenic reagents placed in each detection hole 6 are different. In order to better observe the reaction condition of the body fluid of the patient in different concentrations and different kinds of chromogenic reagents, the present example is further provided with eight groups of detection cards 5, wherein the types of the chromogenic reagents placed in each detection card 5 in the eight groups of detection cards 5 are the same, but the concentrations of the chromogenic reagents placed at the same positions are different, for example, each detection hole 6 in the eight groups of detection cards 5 is given a number, and the types of the chromogenic reagents in the detection holes 6 with the same number are the same but the concentrations are different. Therefore, the patient needs to drop his body onto the detection hole 6 at the time of detection.
The incubator includes box 1 and upper cover 4, and upper cover 4 lid closes on box 1, is provided with in the box 1 with detect constant head tank 2 of 5 joint of card, and the bottom of constant head tank 2 is provided with imaging hole 3, and imaging hole 3's open end and constant head tank 2's bottom coincidence, imaging hole 3's bottom is provided with the cmos camera, and the imaging sight of cmos camera is towards imaging hole 3's open end setting, does not show the cmos camera in the drawing, and with its bottom that can fixed mounting at imaging hole 3 is standard. In addition, a heating module is further arranged in the incubator, the heating module is arranged close to the positioning groove 2, the heating module such as an air heater, an infrared heating lamp and the like can be used for realizing constant-temperature incubation of body fluid to be tested for a designated time, and similarly, the heating module is not shown in the drawing.
When the patient drops the body fluid to be tested into the test card 5, the test card 5 needs to be placed in the positioning slot 2, usually one test card 5 occupies one positioning slot 2, and the test card 5 is clamped with the positioning slot 2 and closes the upper cover 4. At this time, the heating module can incubate the body fluid to be tested at constant temperature, and the incubation time needs to reach a designated time, and the designated time is preset time. After incubation at constant temperature is completed, the cmos camera can image the body fluid to be detected through the imaging holes 3 to obtain detection data, so that the detection data reflect the state of the body fluid to be detected in each detection hole 6. It will be appreciated that, due to the different types of chromogenic reagents that are added to the different wells 6, there may be a difference in the color of each body fluid to be tested after incubation at constant temperature.
The server 8 is in communication connection with the detection device 7, and the server 8 is configured to process the detection data by using a preset analysis model to obtain an index concentration value corresponding to the body fluid to be detected, process the index concentration value corresponding to the body fluid to be detected by using a preset evaluation model, obtain analysis results by using historical detection data of the same patient at different detection times and characteristic information of the patient, and send the analysis results to the mobile terminal 9.
Specifically, the process of analyzing the detection data by the server 8 is as follows.
First, detection data is acquired, which is uploaded to the server 8 by the detection device 7.
Then, the region where each detection hole 6 included in the detection data is located is divided to obtain a body fluid image. The present example is provided with eight sets of detection cards 5, one detection card 5 is a set, and after six body fluid images corresponding to each other are obtained by the eight sets of detection cards 5 through the above manner, a standard curve formula is established according to the obtained body fluid images. Specifically, the procedure of creating the standard curve formula is as follows in steps S1 to S5.
Step S1, acquiring a plurality of groups of body fluid images, wherein each group of body fluid images comprises a plurality of body fluids to be tested with the same types and different concentrations of chromogenic reagents. As in the above example, it is assumed that each of the detection wells 6 on the eight sets of detection cards 5 is given a number, and the types of the color-developing reagents of the detection wells 6 of the same number are the same but the concentrations are different, so that each set of body fluid images is a body fluid image from the detection wells 6 of the eight sets of detection cards 5 having the same number. Thus, each set of body fluid images includes eight body fluid images.
In this example, a standard curve formula corresponding to each group of liquid images is established, and steps S2 to S5 are the establishment process of the standard curve formula of one group of liquid images.
And S2, determining an abscissa in a preset coordinate system according to the concentrations of the chromogenic reagent in the plurality of body fluid images in each group of body fluid images. Specifically, a coordinate system is set in advance, the abscissa of the coordinate system is the concentration of the color-developing reagent, and the ordinate is the gradation value of the body fluid image. And then, the concentration of the chromogenic reagent corresponding to each body fluid image in the group of body fluid images is called, wherein the concentration of the chromogenic reagent can be measured before the chromogenic reagent is put into the body fluid of a patient, or can be manually recorded when the chromogenic reagent and the body fluid of the patient are proportioned. That is, the corresponding color reagent concentrations in the set of body fluid images can be determined directly in the coordinate system, i.e., the abscissa.
And S3, determining an ordinate in a preset coordinate system according to the color depth of the color developing reagent of the plurality of body fluid images in each group of body fluid images.
First, each body fluid image in the set of body fluid images is preprocessed to determine invalid regions in the body fluid images. In a specific example, the ineffective area is an edge shadow area of each detection hole 6, and the ineffective area can be determined by means of image recognition, wherein the image recognition is a neural network model obtained by training in advance. In other embodiments, the invalid region may be a reflective region and a flare region caused by reflection and flare, respectively, for example, the reflective region and the flare region in the body fluid image are identified, and even if some images still need to identify a floccule region, the floccule region refers to an invalid region caused by floccule generated by insufficient reaction of the body fluid to be tested. Thereafter, the identified invalid region is removed.
And then judging whether the color depth of the body fluid image after pretreatment meets the preset uniformity degree, for example, calculating the gray value difference value of different color depth areas in the body fluid image, wherein the different color depth areas are also identified by adopting a neural network model obtained by training in advance, and particularly, the different areas are divided according to the gray value obtained by identification. If the maximum value in the gray value difference is smaller than a first preset value, namely the color difference between the deepest color area and the shallowest color area of the body fluid image is small, the body fluid image is indicated to meet the preset uniformity degree, otherwise, the body fluid image does not meet the preset uniformity degree. Of course, the number of the areas with different color depths in the body fluid image can be counted, if the number of the areas with different color depths is smaller than a second preset value, the body fluid image is indicated to meet the preset uniformity degree, for example, the second preset value is 5, if the number of the areas with different color depths is smaller than five, the body fluid image is indicated to meet the uniformity degree, otherwise, the body fluid image is not indicated to meet the uniformity degree. The first preset value and the second preset value are values defined in advance.
If the body fluid image meets the preset uniformity degree, directly calculating the gray value of the body fluid image, and then determining the calculated gray value in a coordinate system, namely determining the ordinate.
If the body fluid image does not meet the preset uniformity, a calculation model is used to predict the ordinate of the body fluid image, and the specific prediction process is shown in step S31 to step S34:
Step S31, dividing the body fluid image into a plurality of subareas according to color depth by adopting an image segmentation algorithm, dividing the subareas into different chroma grades according to colors in the subareas, and setting multi-grade chroma grades in advance according to the example, wherein the color depth corresponding to each grade chroma grade is different, and the higher the chroma grade is, the deeper the color depth is, and the larger the corresponding gray value is. For the convenience of explanation of the classification process, for example, as shown in fig. 4, it is assumed that there is a body fluid image a, which is divided into 5 sub-areas according to color depths, denoted by a1, a2, a3, a4, a5, respectively, wherein the color depths of the sub-areas a1 and a3 are identical, the color depths of the sub-areas a2 and a5 are identical, different filling lines are used to represent different color depths in fig. 4, and the same lines represent the same color depth. Meanwhile, the color depth of the sub-areas a1 and a3 > the color depth of the sub-areas a2 and a5 > the color depth of the sub-area a4, and the color depths of the sub-areas a1 and a3 correspond to the first chromaticity level, the color depths of the sub-areas a2 and a5 correspond to the second chromaticity level, and the sub-area a4 corresponds to the third chromaticity level, then the sub-areas a1 and a3 are divided into the first chromaticity level, the sub-areas a2 and a5 are divided into the second chromaticity level, and the sub-area a4 is divided into the third chromaticity level.
Step S32, calculating the classification probability of the body fluid image according to the division result of the subregions of the body fluid image, specifically, the occupation ratio of the area of the subregions falling into different color levels to the total area of the body fluid image according to the body fluid image, and taking the occupation ratio as the classification probability of the body fluid image in the color level. As in the example of fig. 4, assuming that the areas corresponding to the sub-areas a1, a2, a3, a4, a5 are 2,4, 2,3, 5, respectively, in mm 2, the classification probability of the body fluid image a at the first chromaticity level is (2+2)/(2+4+2+3+5) =25%, the classification probability of the body fluid image at the second chromaticity level is (4+5)/(2+4+2+3+5) =56.25%, and the classification probability of the body fluid image at the third chromaticity level is (3)/(2+4+2+3+5) =18.75%. Of course, if the color depth in the body fluid image is consistent, only one sub-region is divided, and at the same time, the classification probability corresponding to the body fluid image on the chromaticity level of the sub-region is 100%.
And S33, selecting one classification probability with the largest classification probability as a target probability, and matching the corresponding standard curve formula through the target probability. Specifically, when there is only one classification probability of the body fluid image, that is, when colors of all sub-areas of the body fluid image are identical, the classification probability is directly taken as the target probability. When the body fluid image has a plurality of classification probabilities, one classification probability having the largest classification probability is selected as the target probability. Then, the gradation value of the chromaticity level corresponding to the target probability is defined in advance as the gradation value of the body fluid image.
Step S34, determining an ordinate in the coordinate system according to the gray value of the body fluid image.
Therefore, each body fluid image can be predicted to obtain a corresponding gray value through the steps S31 to S34, and then the ordinate in the coordinate system is locked according to the gray value.
Of course, in addition to the above-described prediction of the gradation value of the body fluid image in the manner of step S31 to step S34, a neural network model mainly having a YOLOv network structure may be used to predict the gradation value of the body fluid image. Specifically, a YOLOv network structure is adopted to finely adjust a body fluid image, invalid areas such as a shadow area, a reflection area and a facula area in the image are removed, then a multi-layer convolution layer is adopted to carry out convolution operation on the image, data output by the last layer of convolution layer is extracted to serve as a target image, the gray value of the target image serves as the gray value of the body fluid image, and then the ordinate in a coordinate system is locked according to the gray value.
Finally, the body fluid image is reduced to one-dimensional data according to the weight of the neural network model, and then a standard curve formula is called, wherein the standard curve formula is trained in advance and used for reflecting the mapping relation between the index concentration value and the one-dimensional data. Therefore, after the one-dimensional data is obtained, the index concentration value corresponding to the one-dimensional data is obtained through the mapping relation of the standard curve formula, so that the index concentration value corresponding to each body fluid image is obtained.
And S4, combining the abscissa and the ordinate obtained for the same body fluid image into a standard coordinate.
And S5, connecting a plurality of standard coordinates corresponding to each group of body fluid images in series to obtain a standard curve formula.
For convenience of explanation, for example, as shown in fig. 5, it is assumed that a coordinate system with an abscissa representing the concentration of the color-developing reagent and an ordinate representing the gray value exists, and the standard coordinates corresponding to eight body fluid images in one set of body fluid images are a point, b point, c point, d point, e point, f point, g point, and h point, respectively, and the abscissa and the ordinate representing the a point correspond to the same body fluid image, and the b point, c point, d point, e point, f point, g point, and h point correspond to one body fluid image respectively. And sequentially connecting a plurality of standard coordinates in series according to a descending order sorting mode or an ascending order sorting mode of the concentration of the color reagent, namely sequentially connecting an a point, a b point, a c point, a d point, an e point, an f point, a g point and an h point, so as to obtain a curve which is a standard curve formula corresponding to the group of body liquid images.
It should be noted that, in the example of fig. 5, the standard curve formula is a curve close to a straight line, and in practical application, the standard curve formula may be a curved curve, specifically obtained in actual standard coordinates.
Based on the obtained standard curve formula, in the subsequent use process, the server 8 can match the standard curve formula corresponding to each body fluid image from a plurality of standard curve formulas by adopting a preset analysis model according to the type of the body fluid of the patient in the body fluid to be detected, the type and the concentration of the chromogenic reagent, namely, lock the abscissa of the standard curve formula according to the concentration of the chromogenic reagent corresponding to the body fluid image, and lock the ordinate corresponding to the abscissa, namely, the index concentration value of the body fluid image, if the abscissa is locked to be 250 in the example of fig. 5, the corresponding ordinate 15000 is the index concentration value.
Based on the obtained index concentration value corresponding to each body fluid image, the characteristic information of the patient and the historical detection data of different detection times of the patient are obtained. Wherein the patient characteristic information includes, but is not limited to, the patient's height, weight, and age. The history detection data includes index concentration values, arithmetic relationships among the index concentration values and analysis results during each detection, and the arithmetic relationships among the index concentration values include a proportional relationship among the index concentration values, which may specifically be a proportional relationship between one index concentration value and a plurality of index concentration values, a proportional relationship between a plurality of index concentration values and all index concentration values, or other proportional relationships, which are not limited herein.
Then, the characteristic information of the patient, the historical detection data of different detection times of the patient, the index concentration values and the arithmetic relation between the index concentration values are processed by adopting an evaluation model to determine an analysis result. The evaluation model is a trained neural network model. When the model is trained, the actual clinical data is mainly used as sample data for training, namely, detection data of each patient in each detection and disease types obtained by expert analysis according to the detection data are used as samples for training the neural network.
Based on the trained evaluation model, the characteristic information of the patient, historical detection data of different detection times of the patient, each index concentration value and the arithmetic relation of the index concentration values are input into the evaluation model, and then the analysis result of the patient can be obtained.
It should be noted that, the analysis result includes the disease type corresponding to the patient and the prediction probability of the disease type, where the calculation process of the prediction probability is:
Assume that the existing features in the evaluation model are Label(s)The specific example sample is 20000 cases, the characteristics are 15 items, 6 items are body fluid to be tested, 9 items are characteristic information of patients, the diseases set by the label are 6 types, such as nephritis, myocarditis, muscular dystrophy, diabetes, kidney stones and no abnormality, and the data set in the evaluation model can be expressed as follows:
,
Wherein, Y i e { nephritis, myocarditis, muscular dystrophy, diabetes, kidney stones, no anomaly }, the table structure in the dataset is 20000 x 16.
The evaluation model calculates the predictive probability of any one sample (disease type) as:
,,
f k is a base learner, the evaluation model is composed of a plurality of base learners, K is the number of the base learners, and K represents a set of the base learners in the evaluation model. The objective function of the evaluation model is:
,
is the predictive probability of the first t-1 learners for the samples, Is the predictive probability of the current learner to the samples, n is the number of samples, l is the loss function, and y i andThe established function of the loss is that,Is the regularization term of the t-th learner.
Taylor quadratic expansion is performed on the objective function, and the form is as follows:
,
Wherein, ,。
Therefore, after the analysis result is obtained, analyzing the disease type and the prediction probability in the analysis result, if the prediction probability is lower than a preset probability threshold, the analysis result obtained by the current evaluation model is inaccurate, and the evaluation model needs to be optimized at the moment.
Firstly, a disease map library is prepared, wherein the disease map library comprises a plurality of disease types and body fluid standard images corresponding to each disease type, the body fluid standard images are standard images for representing body fluid of a disease when the disease is suffered from a certain type, in general, each disease type corresponds to a plurality of body fluid standard images, and the body fluid standard images contain states of different types of body fluid. Then, the similarity between each body fluid standard image and the body fluid image corresponding to each detection hole 6 is calculated, specifically, a disease type and a plurality of body fluid standard images corresponding to the disease type are selected from a disease map library, and then the body fluid standard images and the body fluid images of the same kind of body fluid are matched and the similarity is calculated. For disease types with multiple body fluid standard images, the average similarity of multiple sets of images is typically calculated as the similarity of the disease of the patient to the disease type, each set of images including a body fluid standard image and a body fluid image.
Repeating the above process until the similarity between the disease of the patient and each disease in the disease map library can be determined in the current detection data.
Finally, the disease type with the highest similarity is selected as an analysis result of the current detection of the patient, and the disease type is simultaneously led into an evaluation model, so that the purpose of dynamically supplementing the richness of the disease type in the evaluation model is realized, and the disease type set by the label is expanded from 6 types to 10 types, namely y i epsilon { nephritis, myocarditis, muscular dystrophy, diabetes, kidney stones, no abnormality and new disease 1, new disease 2, new disease 3, new disease 4}, new disease 1, new disease 2, new disease 3 and new disease 4 are diseases of different types.
And (3) recalibrating the sample label with low accuracy without changing the correct sample label, collecting relevant characteristic indexes of the sample label, and training the evaluation model to enable the evaluation model to be more perfect. For the concrete description, the above example is still adopted, that is, there are 20000 samples in the evaluation model, the characteristics are increased to 18 items, 9 items are body fluids, 9 items are characteristic information of the patient, the evaluation model is trained after the data set is newly added with the disease type, and the data set at this time is:
Wherein, Y i epsilon { nephritis, myocarditis, muscular dystrophy, diabetes, kidney stones, no abnormality, new disease 1, new disease 2, new disease 3, new disease 4}. Therefore, the effect of optimizing the evaluation model is achieved by supplementing the data set in the evaluation model, so that the evaluation model can output more accurate analysis results in the subsequent processing flow.
Similarly, for the situation that indexes and labels in the evaluation model are not contained in the disease map library, the node and relationship of the disease map library can be newly built and updated.
Further, the analysis result obtained by the analysis by the server 8 may be transmitted to the mobile terminal 9.
In the above-described process, the acquired detection data, the index concentration value of each body fluid to be measured, are stored in the server 8. The server 8 may be deployed locally or at the cloud.
The mobile terminal 9 is in communication connection with the server 8 and is used for binding with the detection equipment 7, setting the characteristic information of the patient, detecting time delay and checking the detection result. In some specific embodiments, the mobile terminal 9 may be a cell phone, tablet, mini-letter applet or APP software.
In the embodiment of the present application, the mobile terminal 9 and the server 8 may communicate with each other in a wired manner, such as a network cable, or in a wireless manner, such as wifi, bluetooth, 4G or 5G network.
The system is further described below by way of illustrating the manner in which the in-vivo constituent metabolism detection system is used.
Firstly, the patient needs to register an account number by using the mobile terminal 9, then the mobile terminal 9 and the detection device 7 are bound through a network cable or Bluetooth/wifi, and meanwhile, characteristic information of the patient is recorded.
Then, the information of the test card 5 is scanned, and the body fluid to be tested is dropped into the test hole 6 of the test card 5.
Subsequently, the detection card 5 is placed in an incubator, and the upper cover 4 is closed to perform incubation at a constant temperature for a prescribed time. The detection device 7 can automatically collect detection data after the incubation at constant temperature is completed, and upload the detection data to the server 8.
Finally, the server 8 analyzes the detection data to generate an analysis result, and returns the analysis result to the mobile terminal 9, so that the patient can check through the mobile terminal 9.
When the patient detects again, the characteristic information of the patient can be directly input to detect.
The principle of implementation of the in-vivo constituent metabolism detection system of the embodiment of the application is that the detection equipment 7 can incubate body fluid to be detected at constant temperature to obtain detection data, the server 8 can analyze by combining the detection data, the historical detection data and the characteristic information of a patient and obtain an analysis result, so that the patient can check the analysis result through the mobile terminal 9. Because the historical detection data and the characteristic information are combined when the detection data are analyzed, the obtained analysis result is more comprehensive and accurate. The in-vivo component metabolism detection system can realize qualitative, semi-quantitative and quantitative detection and monitoring of body fluid to be detected.
The system has perfect body composition metabolism detection indexes, corresponding reagent consumables, equipment, algorithms and patient terminal software, so that the clinical indication meaning of the detection result is clear, the equipment is small and portable, the operation is simple and convenient, the field limitation is avoided, and the patient can see the detection result and the health evaluation condition at any time and any place, so that the system is applicable to two detection scenes of hospitals and families.
The foregoing description of the preferred embodiments of the application is not intended to limit the scope of the application in any way, including the abstract and drawings, in which case any feature disclosed in this specification (including abstract and drawings) may be replaced by alternative features serving the same, equivalent purpose, unless expressly stated otherwise. That is, each feature is one example only of a generic series of equivalent or similar features, unless expressly stated otherwise.
Claims (9)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411296611.5A CN118817669B (en) | 2024-09-18 | 2024-09-18 | Intelligent in-vivo ingredient metabolism detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411296611.5A CN118817669B (en) | 2024-09-18 | 2024-09-18 | Intelligent in-vivo ingredient metabolism detection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118817669A CN118817669A (en) | 2024-10-22 |
| CN118817669B true CN118817669B (en) | 2024-12-20 |
Family
ID=93077021
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411296611.5A Active CN118817669B (en) | 2024-09-18 | 2024-09-18 | Intelligent in-vivo ingredient metabolism detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118817669B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119178873A (en) * | 2024-11-22 | 2024-12-24 | 北京中生金域诊断技术股份有限公司 | Method and system for monitoring metabolism of components in intelligent body |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724972A (en) * | 2019-02-25 | 2019-05-07 | 深圳市象形字科技股份有限公司 | A kind of portable intelligent urine detection instrument and detection method |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201239360A (en) * | 2011-03-25 | 2012-10-01 | Middleland Sensing Technology Inc | An analyzing system and method for a test strip |
| CN106645706A (en) * | 2015-12-29 | 2017-05-10 | 江苏戴格诺思生物技术有限公司 | Method for quantitatively analyzing colloidal gold detection reagent by mobile terminal |
| CN108279298B (en) * | 2018-02-09 | 2024-08-23 | 中山市漫鑫信息科技有限公司 | Intelligent detection system |
| EP3667301B1 (en) * | 2018-12-10 | 2021-12-01 | Roche Diabetes Care GmbH | Method and system for determining concentration of an analyte in a sample of a bodily fluid, and method and system for generating a software-implemented module |
| CN109900686A (en) * | 2019-02-27 | 2019-06-18 | 山东大学齐鲁医院 | A method, system and terminal for improving the accuracy of semi-quantitative poison detection |
| CN110349155A (en) * | 2019-07-26 | 2019-10-18 | 重庆德方信息技术有限公司 | Health detection image analysis system |
| CN114527116A (en) * | 2022-01-24 | 2022-05-24 | 太原理工大学 | Ambient light correction array type colorimetric analysis system and method based on intelligent equipment |
| CN116429702A (en) * | 2022-12-22 | 2023-07-14 | 江苏省食品药品监督检验研究院 | Sample color development rapid analysis and detection method and detection kit based on intelligent equipment |
| CN117990930A (en) * | 2024-02-01 | 2024-05-07 | 太原理工大学 | Portable intelligent analyzer based on digital picture colorimetric method and detection method |
-
2024
- 2024-09-18 CN CN202411296611.5A patent/CN118817669B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724972A (en) * | 2019-02-25 | 2019-05-07 | 深圳市象形字科技股份有限公司 | A kind of portable intelligent urine detection instrument and detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118817669A (en) | 2024-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3612963B1 (en) | Biochemical analyser based on a machine learning algorithm using test strips and a smartdevice | |
| CN118817669B (en) | Intelligent in-vivo ingredient metabolism detection system | |
| CN106991679A (en) | One kind quantifies recognition methods based on cloud platform urine test paper physical signs | |
| KR102091832B1 (en) | Portable In Vitro Diagnostic Kit Analyzer Using Multimedia Information | |
| JP2007052774A (en) | Computer assisted method for diagnosing disease | |
| CN111429990A (en) | Intelligent management system for interpretation, comparison analysis and reminding of inspection results | |
| JP2002022748A (en) | Clinical test system | |
| JP2023016928A (en) | Automated validation of medical data | |
| CN108231193A (en) | A kind of clinical decision auxiliary system and method | |
| CN108593643A (en) | Urine detection method based on intelligent mobile terminal | |
| US20130096878A1 (en) | State inference in a heterogeneous system | |
| EP4026140A1 (en) | Adjustment method for an analytical determination of an analyte in a bodily fluid | |
| WO2009063185A1 (en) | Test results reading method and apparatus | |
| CN101311726B (en) | Image detection method of detection disk | |
| JP6791170B2 (en) | Analytical instruments, analytical methods, and analytical systems | |
| Higgins et al. | Reference intervals: theory and practice | |
| CN113130050B (en) | A medical information display method and display system | |
| CN113990485A (en) | Method for diagnosing diseases based on conventional test data and knowledge graph | |
| JP2021174201A (en) | Program, information processor, information processing method, and learning model generation method | |
| CN117524456A (en) | Auxiliary diagnosis method and system | |
| Zeb et al. | Towards the selection of the best machine learning techniques and methods for urinalysis | |
| Flores et al. | A step forward in the diagnosis of urinary tract infections: from machine learning to clinical practice | |
| CN105223194A (en) | A kind of integrated urine examination method and system | |
| CN119178873A (en) | Method and system for monitoring metabolism of components in intelligent body | |
| JP6805492B2 (en) | Analyzer and analysis method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |