CN117522834A - Blood sample lipidemia and hemolysis quality monitoring method based on deep learning method - Google Patents
Blood sample lipidemia and hemolysis quality monitoring method based on deep learning method Download PDFInfo
- Publication number
- CN117522834A CN117522834A CN202311568523.1A CN202311568523A CN117522834A CN 117522834 A CN117522834 A CN 117522834A CN 202311568523 A CN202311568523 A CN 202311568523A CN 117522834 A CN117522834 A CN 117522834A
- Authority
- CN
- China
- Prior art keywords
- blood sample
- model
- image
- monitoring
- hemolysis
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 101
- 210000004369 blood Anatomy 0.000 title claims abstract description 89
- 239000008280 blood Substances 0.000 title claims abstract description 89
- 238000012544 monitoring process Methods 0.000 title claims abstract description 45
- 206010018910 Haemolysis Diseases 0.000 title claims abstract description 33
- 230000008588 hemolysis Effects 0.000 title claims abstract description 33
- 238000013135 deep learning Methods 0.000 title claims abstract description 24
- 238000012549 training Methods 0.000 claims abstract description 35
- 238000011156 evaluation Methods 0.000 claims abstract description 20
- 238000002372 labelling Methods 0.000 claims abstract description 18
- 238000001514 detection method Methods 0.000 claims abstract description 15
- 238000005457 optimization Methods 0.000 claims abstract description 15
- 238000013480 data collection Methods 0.000 claims abstract description 13
- 230000008713 feedback mechanism Effects 0.000 claims abstract description 12
- 230000002708 enhancing effect Effects 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims description 13
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 6
- 238000003672 processing method Methods 0.000 claims description 6
- 230000035945 sensitivity Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 4
- 150000002632 lipids Chemical class 0.000 claims description 4
- 230000009897 systematic effect Effects 0.000 claims description 3
- 238000000844 transformation Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 2
- 210000002966 serum Anatomy 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 206010011409 Cross infection Diseases 0.000 description 1
- 102000001554 Hemoglobins Human genes 0.000 description 1
- 108010054147 Hemoglobins Proteins 0.000 description 1
- 206010029803 Nosocomial infection Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000601 blood cell Anatomy 0.000 description 1
- 238000009534 blood test Methods 0.000 description 1
- 238000005119 centrifugation Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000004737 colorimetric analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 210000003743 erythrocyte Anatomy 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to the technical field of blood detection, in particular to a blood sample lipidemia and hemolysis quality monitoring method based on a deep learning method, which comprises data collection and labeling, pretreatment and enhancement, model selection and training, model evaluation and optimization, real-time monitoring and a feedback mechanism, wherein the data collection and labeling are used for collecting blood sample pictures and labeling the pictures, the pretreatment and enhancement are used for processing the collected pictures and enhancing the diversity of the pictures, the model selection and training are used for training the labeled data, the model evaluation and optimization are used for detecting a model, the reliability of the model is ensured, the real-time monitoring is used for detecting the blood sample and giving corresponding advice, and the feedback mechanism is used for feeding back the problems in the detection process, and the invention has the advantages that: realizes noninvasive detection of lipidemia and hemolysis, and greatly improves timeliness and safety.
Description
Technical Field
The invention relates to the technical field of blood detection, in particular to a blood sample lipidemia and hemolysis quality monitoring method based on a deep learning method.
Background
In the current quality control of blood samples, there are also problems in the determination of quality monitoring of blood, blood hemolysis, etc.
For lipidemia, it is usually referred to as that the fat content in blood is too high, which affects the accuracy of blood detection results, and at present, a common method is to separate serum from blood cells by centrifugation, and then judge by artificial contrast color card or perform biochemical detection on serum index. However, the manual comparison of the color cards is performed by manually judging individual variability; again, this is time consuming and does not allow for real-time monitoring. In addition, the accuracy of the results is also affected by the different specific gravities of the different lipids. Although serum index detection is good in terms of accuracy of judgment of results, detection time is long, and a certain economic cost of detection reagents is required.
For hemolysis, it refers to the release of hemoglobin into the serum caused by rupture of red blood cells, which also affects the accuracy of the blood test results. At present, the common methods are to judge by adopting naked eye observation or test paper colorimetric methods, and the methods are high in subjectivity and cannot be used for quantitative measurement.
Therefore, the existing monitoring and judging method for the quality of the blood sample such as the lipidemia, the hemolysis and the like has a plurality of defects, and a more accurate, rapid and simple method is needed for monitoring and controlling, and the deep learning method is used as an advanced machine learning method and can play a great role in the field.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: there is a need for a more accurate, rapid, and simple method of monitoring and control.
The invention adopts the following technical scheme to solve the technical problems: the method comprises the steps of data collection and labeling, pretreatment and enhancement, model selection and training, model evaluation and optimization, real-time monitoring and feedback mechanism, wherein the data collection and labeling are used for collecting blood sample pictures and labeling the pictures, the pretreatment and enhancement are used for processing the collected pictures and enhancing the diversity of the pictures, the model selection and training are used for training the labeled data, the model evaluation and optimization are used for detecting the model, the reliability of the model is guaranteed, the real-time monitoring is used for detecting the blood sample and giving corresponding suggestions, and the feedback mechanism is used for feeding back the problems in the detection process.
Preferably, the method for acquiring the data collection and labeling blood sample picture is shooting by a camera, and the shooting method is as follows: the shooting light is adjusted to be in dark, normal and bright modes for shooting, each group of photos is respectively shot in front of, behind, left of, right of and top view, the camera is one to one distance away from the blood sample, and each photo under the shooting light needs to shoot the photo of blood just extracted, one hour after extraction and three hours after extraction.
Preferably, the method for collecting and labeling data comprises the following steps: firstly, a blood sample is required to be identified by a large number of specialized doctors and experimenters, the blood sample is marked, and then the corresponding photographed picture is marked to indicate whether the blood sample in the image has the lipid blood or the hemolysis.
Preferably, the preprocessing and enhancing image processing method comprises the following steps: firstly, preprocessing an image, carrying out graying treatment on the image, taking the brightness of three components in a color image as the gray values of the three gray images, selecting one gray image according to application requirements, then carrying out weighted average on the three components according to different weights according to importance and other indexes, and carrying out weighted average on the three components of RGB according to the following formula because the sensitivity of human eyes to green is highest and the sensitivity to blue is lowest, so that a more reasonable gray image can be obtained, wherein f (i, j) =0.30R (i, j) +0.59G (i, j) +0.11B (i, j), and fk (i, j) (k=1, 2, 3) is the gray value of the converted gray image.
Preferably, the method for enhancing the image in preprocessing and enhancing comprises the following steps: the acquired images are processed through geometric transformations such as translation, transposition, mirroring, rotation, scaling and the like, so as to correct systematic errors of an image acquisition system and random errors of instrument positions, and a gray interpolation algorithm is also needed, because the pixels of the output images are possibly mapped onto non-integer coordinates of the input images according to the transformation relation, and then the processed images of all views are synthesized to form the 3D image.
Preferably, the training method for model selection and training comprises the following steps: training the marked image by using a computer through a convolutional neural network deep learning model, wherein in the training process, a supervised learning method is adopted, and a large amount of marked data and unmarked data are used for training, wherein the supervised learning method is used for carrying out data learning according to the marking of the marked data, the unmarked data is trained by comparing the unmarked data with the marked data, if the unmarked data is not marked, the marking is carried out, and if the marked data is judged, whether the data is accurate is judged manually.
Preferably, the model evaluation and optimization apple method is that a computer evaluates the model, adopts cross verification, ROC curve and other evaluation methods to determine the accuracy and reliability of the model, and optimizes according to the evaluation result, wherein the optimization method is that the learning efficiency of the model is changed and the model is subjected to countermeasure training.
Preferably, the real-time monitoring method is to apply the trained model to real-time blood sample quality monitoring. When a new blood sample image is input, the model can automatically judge whether the blood sample contains the lipidemia or the hemolysis, and give corresponding results and suggestions.
Preferably, the feedback method of the feedback mechanism is as follows: judging the blood sample, and sending out an alarm to remind a inspector if the blood sample does not meet the requirements, displaying the reasons of the blood sample, and prompting a corresponding processing method, wherein the alarm mode comprises the steps of displaying an alarm, a lamplight alarm and a sound alarm on a visual screen.
Compared with the prior art, the invention provides a blood sample lipidemia and hemolysis quality monitoring method based on a deep learning method, which has the following beneficial effects:
1. according to the blood sample lipidemia and hemolysis quality monitoring method based on the deep learning method, a traditional blood detection method generally needs to collect a blood sample and analyze the blood sample by using a laboratory instrument, so that pain of a patient is increased, risks such as cross infection and the like can be possibly caused, and the deep learning method can acquire information by analyzing images, so that noninvasive lipidemia and hemolysis detection is realized, and timeliness and safety are greatly improved.
2. According to the blood sample lipidemia and hemolysis quality monitoring method based on the deep learning method, the deep learning method can realize on-line monitoring of the quality of the blood sample, and the blood sample does not need to be sent into a laboratory for off-line analysis like a traditional method, so that the real-time performance and accuracy of monitoring can be improved, the sample processing time can be greatly shortened, the detection efficiency can be improved, the deep learning method can quantitatively measure lipidemia and hemolysis by analyzing the image or spectrum information of the blood sample, and the quality condition of the blood sample can be accurately reflected, and more accurate basis can be provided for clinical diagnosis and treatment.
3. According to the blood sample lipidemia and hemolysis quality monitoring method based on the deep learning method, the blood sample can be automatically classified and judged by the deep learning method without manual intervention. The method can not only improve the accuracy and consistency of discrimination, but also greatly reduce the labor cost, improve the working efficiency, has strong expandability, can easily integrate and fuse with other technologies, can combine the deep learning with the technologies of spectrum analysis, image processing and the like, and realizes more accurate and efficient blood sample quality monitoring.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
In the figure: 1. data collection and labeling; 2. pretreatment and enhancement; 3. model selection and training; 4. model evaluation and optimization; 5. monitoring in real time; 6. feedback mechanism.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
Referring to fig. 1, a blood sample lipidemia and hemolysis quality monitoring method based on a deep learning method comprises a data collection and labeling 1, a preprocessing and enhancement 2, a model selection and training 3, a model evaluation and optimization 4, a real-time monitoring 5 and a feedback mechanism 6, wherein the data collection and labeling 1 is used for collecting blood sample pictures and labeling the pictures, the preprocessing and enhancement 2 is used for processing the collected pictures and enhancing the diversity of the pictures, the model selection and training 3 is used for training the labeled data, the model evaluation and optimization 4 is used for detecting the model, the reliability of the model is ensured, the real-time monitoring 5 is used for detecting the blood sample and giving corresponding suggestions, and the feedback mechanism 6 is used for feeding back the problems in the detection process.
In this embodiment, the method for acquiring the data collection and labeling 1 blood sample picture is camera shooting, and the shooting method is as follows: the shooting light is adjusted to be in dark, normal and bright modes for shooting, each group of pictures respectively shoots the front, the back, the left, the right and the top view, the distance between the camera and the blood sample is one to one, and each picture under the shooting light needs to shoot the picture of blood just extracted, one hour after extraction and three hours after extraction.
In this embodiment, the method for collecting data and labeling 1 includes: firstly, a blood sample is required to be identified by a large number of specialized doctors and experimenters, the blood sample is marked, and then the corresponding photographed picture is marked to indicate whether the blood sample in the image has the lipid blood or the hemolysis.
In this embodiment, the image processing method of preprocessing and enhancing 2 includes: firstly, preprocessing an image, carrying out graying treatment on the image, taking the brightness of three components in a color image as the gray values of the three gray images, selecting one gray image according to application requirements, then carrying out weighted average on the three components according to different weights according to importance and other indexes, and carrying out weighted average on the three components of RGB according to the following formula because the sensitivity of human eyes to green is highest and the sensitivity to blue is lowest, so that a more reasonable gray image can be obtained, wherein f (i, j) =0.30R (i, j) +0.59G (i, j) +0.11B (i, j), and fk (i, j) (k=1, 2, 3) is the gray value of the converted gray image at i, j.
In this embodiment, the image enhancement method in preprocessing and enhancing 2 includes: the acquired images are processed through geometric transformations such as translation, transposition, mirroring, rotation, scaling and the like, so as to correct systematic errors of an image acquisition system and random errors of instrument positions, and a gray interpolation algorithm is also needed, because the pixels of the output images are possibly mapped onto non-integer coordinates of the input images according to the transformation relation, and then the processed images of all views are synthesized to form the 3D image.
In this embodiment, the training method of model selection and training 3 is as follows: training the marked image by using a computer through a convolutional neural network CNN deep learning model, wherein in the training process, a supervised learning method is adopted, and training is carried out by using a large amount of marked data and unmarked data, wherein the supervised learning method is to carry out data learning according to the marking of the marked data, the unmarked data is trained by comparing the unmarked data with marked data, if the unmarked data is not marked, the marking is carried out, and if the marked data is judged, whether the data is accurate is judged manually.
In this embodiment, the apple method of model evaluation and optimization 4 is that the computer evaluates the model, adopts the evaluation methods such as cross-validation and ROC curve, determines the accuracy and reliability of the model, and optimizes according to the evaluation result, and the optimization method is that the learning efficiency of the model is changed and the model is subjected to countermeasure training.
In this embodiment, the real-time monitoring 5 is to apply the trained model to real-time blood sample quality monitoring. When a new blood sample image is input, the model can automatically judge whether the blood sample contains the lipidemia or the hemolysis, and give corresponding results and suggestions.
In this embodiment, the feedback method of the feedback mechanism 6 is: judging the blood sample, sending an alarm to remind a inspector if the blood sample does not meet the requirements, displaying the reasons of the blood sample, and prompting a corresponding processing method, wherein the alarm mode comprises the steps of displaying an alarm, a lamplight alarm and a sound alarm on a visual screen.
When in use, firstly, a large number of blood samples marked by a professional doctor or an experimenter are collected: lipidemia and hemolysis, and then generating a number of sample corresponding images by a camera, as well as images of normal blood samples. These images need to be annotated by a professional doctor or experimenter to indicate whether there is any blood fat or hemolysis in the images, and the computer performs preprocessing on the collected images, including image size unification, normalization processing, and the like. In addition, a data enhancement method is adopted: the method comprises the steps of rotating, translating, zooming and the like, increasing the diversity of data, improving the generalization capability of a model, training a marked image by a computer through a convolutional neural network CNN deep learning model, training a large amount of marked data and unmarked data by adopting a supervised learning method in the training process, evaluating the model by the computer after the training is finished, evaluating the model by adopting cross verification, ROC curve and other evaluation methods, determining the accuracy and reliability of the model, adjusting and optimizing the model according to the evaluation result, such as changing a network structure, adjusting a learning rate and the like, applying the trained model to real-time blood sample quality monitoring, automatically judging whether the blood sample has fat or hemolysis by the model when a new blood sample image is input, giving corresponding results and suggestions, and sending an alarm by a computer monitoring system in time if the quality of the blood sample does not meet the requirements, and simultaneously, automatically adjusting model parameters according to the monitoring results to improve the accuracy and reliability of the model.
Embodiments of the invention it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations may be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. A blood sample lipidemia and hemolysis quality monitoring method based on a deep learning method comprises the following specific steps of data collection and labeling (1), pretreatment and enhancement (2), model selection and training (3), model evaluation and optimization (4), real-time monitoring (5) and feedback mechanism (6): the data collection and labeling (1) is used for collecting blood sample pictures and labeling the pictures, the preprocessing and enhancement (2) is used for processing the collected pictures and enhancing the diversity of the pictures, the model selection and training (3) is used for training the labeled data, the model evaluation and optimization (4) is used for detecting the model, the reliability of the model is guaranteed, the real-time monitoring (5) is used for detecting the blood samples and giving corresponding suggestions, and the feedback mechanism (6) is used for feeding back the problems in the detection process.
2. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 1, wherein the method comprises the following steps: the data collection and labeling (1) blood sample picture acquisition method is camera shooting, and the shooting method is as follows: the shooting light is adjusted to be in dark, normal and bright modes for shooting, each group of photos is respectively shot in front of, behind, left of, right of and top view, the camera is one to one distance away from the blood sample, and each photo under the shooting light needs to shoot the photo of blood just extracted, one hour after extraction and three hours after extraction.
3. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 2, wherein the method comprises the following steps: the marking method of the data collection and marking (1) comprises the following steps: firstly, a blood sample is required to be identified by a large number of specialized doctors and experimenters, the blood sample is marked, and then the corresponding photographed picture is marked to indicate whether the blood sample in the image has the lipid blood or the hemolysis.
4. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 1, wherein the method comprises the following steps: the image processing method of the preprocessing and enhancement (2) comprises the following steps: firstly, preprocessing an image, carrying out graying treatment on the image, taking the brightness of three components in a color image as the gray values of the three gray images, selecting one gray image according to application requirements, then carrying out weighted average on the three components according to different weights according to importance and other indexes, and carrying out weighted average on the three components of RGB according to the following formula because the sensitivity of human eyes to green is highest and the sensitivity to blue is lowest, so that a more reasonable gray image can be obtained, wherein f (i, j) =0.30R (i, j) +0.59G (i, j) +0.11B (i, j), and fk (i, j) (k=1, 2, 3) is the gray value of the converted gray image at (i, j).
5. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 4, wherein the method comprises the following steps: the image enhancement method in the preprocessing and enhancement (2) comprises the following steps: the acquired images are processed through geometric transformations such as translation, transposition, mirroring, rotation, scaling and the like, so as to correct systematic errors of an image acquisition system and random errors of instrument positions, and a gray interpolation algorithm is also needed, because the pixels of the output images are possibly mapped onto non-integer coordinates of the input images according to the transformation relation, and then the processed images of all views are synthesized to form the 3D image.
6. The method for monitoring the quality of blood sample lipemia and hemolysis based on the deep learning method according to claim 5, wherein the method comprises the following steps: the training method of the model selection and training (3) comprises the following steps: training the marked image by using a computer through a Convolutional Neural Network (CNN) deep learning model, wherein in the training process, a supervised learning method is adopted, and training is performed by using a large amount of marked data and unmarked data, wherein the supervised learning method is used for performing data learning according to the marking of the marked data, the unmarked data is trained by comparing the unmarked data with marked data, if the unmarked data is not marked, the marking is added, and if the marked data is judged to be accurate, the data is judged to be accurate manually.
7. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 1, wherein the method comprises the following steps: the apple method of the model evaluation and optimization (4) is that a computer evaluates the model, adopts evaluation methods such as cross validation, ROC curves and the like, determines the accuracy and reliability of the model, and optimizes according to the evaluation result, wherein the optimization method is that the learning efficiency of the model is changed and the model is subjected to countermeasure training.
8. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 7, wherein the method comprises the following steps: the real-time monitoring (5) monitoring method is to apply the trained model to real-time blood sample quality monitoring. When a new blood sample image is input, the model can automatically judge whether the blood sample contains the lipidemia or the hemolysis, and give corresponding results and suggestions.
9. The method for monitoring the quality of blood sample lipidemia and hemolysis based on the deep learning method according to claim 7, wherein the method comprises the following steps: the feedback method of the feedback mechanism (6) comprises the following steps: judging the blood sample, and sending out an alarm to remind a inspector if the blood sample does not meet the requirements, displaying the reasons of the blood sample, and prompting a corresponding processing method, wherein the alarm mode comprises the steps of displaying an alarm, a lamplight alarm and a sound alarm on a visual screen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311568523.1A CN117522834A (en) | 2023-11-23 | 2023-11-23 | Blood sample lipidemia and hemolysis quality monitoring method based on deep learning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311568523.1A CN117522834A (en) | 2023-11-23 | 2023-11-23 | Blood sample lipidemia and hemolysis quality monitoring method based on deep learning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117522834A true CN117522834A (en) | 2024-02-06 |
Family
ID=89741672
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311568523.1A Pending CN117522834A (en) | 2023-11-23 | 2023-11-23 | Blood sample lipidemia and hemolysis quality monitoring method based on deep learning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117522834A (en) |
-
2023
- 2023-11-23 CN CN202311568523.1A patent/CN117522834A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11842556B2 (en) | Image analysis method, apparatus, program, and learned deep learning algorithm | |
| CN111474701B (en) | Pathological microscopic image real-time acquisition and analysis system, method, device and medium | |
| DK2973397T3 (en) | Tissue-object-based machine learning system for automated assessment of digital whole-slide glass | |
| JP2023065352A (en) | Image analysis method, image analysis device, program, method for manufacturing learned deep learning algorithm, and learned deep learning algorithm | |
| KR102155381B1 (en) | Method, apparatus and software program for cervical cancer decision using image analysis of artificial intelligence based technology | |
| CN112967285B (en) | Chloasma image recognition method, system and device based on deep learning | |
| EP2774067A2 (en) | Generate percentage of positive cells for biomarkers by normalizing and autothresholding the image intensity produced by immunohistochemistry technique | |
| WO2021256514A1 (en) | Living cell analysis device, living cell analysis system, living cell analysis program, and living cell analysis method | |
| CN106706643B (en) | A kind of liver cancer comparison slice detection method | |
| CN107543788A (en) | A kind of urine erythrocyte abnormal rate detection method and system | |
| CN109886346A (en) | A kind of cardiac muscle MRI image categorizing system | |
| CN111656393A (en) | Histological image analysis | |
| CN111462082A (en) | Focus picture recognition device, method and equipment and readable storage medium | |
| EP1360481A2 (en) | System for automated and remote histological analysis and new drug assessment | |
| CN117152152B (en) | Production management system and method for detection kit | |
| WO2017145172A1 (en) | System and method for extraction and analysis of samples under a microscope | |
| CN109711306B (en) | Method and equipment for obtaining facial features based on deep convolutional neural network | |
| CN112690815A (en) | System and method for assisting in diagnosing lesion grade based on lung image report | |
| CN109003659A (en) | Stomach Helicobacter pylori infects pathological diagnosis and supports system and method | |
| CN117522834A (en) | Blood sample lipidemia and hemolysis quality monitoring method based on deep learning method | |
| US20210174147A1 (en) | Operating method of image processing apparatus, image processing apparatus, and computer-readable recording medium | |
| CN115035086A (en) | Intelligent tuberculosis skin test screening and analyzing method and device based on deep learning | |
| JP7358411B2 (en) | Methods, analytical devices, and in vitro diagnostic systems for determining characteristics of sample containers in in vitro diagnostic systems | |
| De Leon et al. | Detection of Sickle Cell Anemia in Blood Smear using YOLOv3 | |
| CN114518362A (en) | Sperm quality analysis device, system, method and readable storage medium |
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 |