CN112017743B - Automatic generation platform and application of disease risk evaluation report - Google Patents
Automatic generation platform and application of disease risk evaluation report Download PDFInfo
- Publication number
- CN112017743B CN112017743B CN202010843194.7A CN202010843194A CN112017743B CN 112017743 B CN112017743 B CN 112017743B CN 202010843194 A CN202010843194 A CN 202010843194A CN 112017743 B CN112017743 B CN 112017743B
- Authority
- CN
- China
- Prior art keywords
- risk
- module
- analysis
- index
- data
- 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
- 201000010099 disease Diseases 0.000 title claims abstract description 36
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 36
- 238000011156 evaluation Methods 0.000 title claims abstract description 34
- 238000010586 diagram Methods 0.000 claims abstract description 47
- 238000004458 analytical method Methods 0.000 claims abstract description 41
- 206010028980 Neoplasm Diseases 0.000 claims abstract description 32
- 201000011510 cancer Diseases 0.000 claims abstract description 32
- 238000003745 diagnosis Methods 0.000 claims abstract description 30
- 230000036210 malignancy Effects 0.000 claims abstract description 26
- 241000510091 Quadrula quadrula Species 0.000 claims abstract description 24
- 230000002159 abnormal effect Effects 0.000 claims abstract description 17
- 238000013058 risk prediction model Methods 0.000 claims abstract description 15
- 238000010606 normalization Methods 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 230000003902 lesion Effects 0.000 claims abstract description 5
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 5
- 238000003759 clinical diagnosis Methods 0.000 claims abstract description 4
- 210000004027 cell Anatomy 0.000 claims description 55
- 244000052769 pathogen Species 0.000 claims description 22
- 230000001717 pathogenic effect Effects 0.000 claims description 20
- 208000015181 infectious disease Diseases 0.000 claims description 15
- 238000012502 risk assessment Methods 0.000 claims description 11
- 230000036541 health Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 9
- 238000001574 biopsy Methods 0.000 claims description 8
- 238000000034 method Methods 0.000 claims description 7
- 238000012512 characterization method Methods 0.000 claims description 6
- 230000035622 drinking Effects 0.000 claims description 6
- 235000006694 eating habits Nutrition 0.000 claims description 6
- 230000000391 smoking effect Effects 0.000 claims description 6
- 108010077544 Chromatin Proteins 0.000 claims description 5
- 210000003483 chromatin Anatomy 0.000 claims description 5
- 239000003086 colorant Substances 0.000 claims description 5
- 230000002380 cytological effect Effects 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 claims description 4
- 230000001788 irregular Effects 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000021164 cell adhesion Effects 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 238000002372 labelling Methods 0.000 claims description 2
- 238000010186 staining Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 210000003855 cell nucleus Anatomy 0.000 description 18
- 230000006872 improvement Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 4
- 210000004881 tumor cell Anatomy 0.000 description 4
- 241000208140 Acer Species 0.000 description 3
- 241000282414 Homo sapiens Species 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000004043 dyeing Methods 0.000 description 3
- 210000004698 lymphocyte Anatomy 0.000 description 3
- 238000003908 quality control method Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 206010030155 Oesophageal carcinoma Diseases 0.000 description 2
- 206010041662 Splinter Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 208000035143 Bacterial infection Diseases 0.000 description 1
- 208000017897 Carcinoma of esophagus Diseases 0.000 description 1
- 208000002699 Digestive System Neoplasms Diseases 0.000 description 1
- 208000000461 Esophageal Neoplasms Diseases 0.000 description 1
- 208000031888 Mycoses Diseases 0.000 description 1
- 230000001093 anti-cancer Effects 0.000 description 1
- 230000001580 bacterial effect Effects 0.000 description 1
- 208000022362 bacterial infectious disease Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 238000001839 endoscopy Methods 0.000 description 1
- 201000004101 esophageal cancer Diseases 0.000 description 1
- 201000005619 esophageal carcinoma Diseases 0.000 description 1
- 210000001035 gastrointestinal tract Anatomy 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000001033 granulometry Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 208000020082 intraepithelial neoplasia Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000004940 nucleus Anatomy 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 238000007430 reference method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 206010041823 squamous cell carcinoma Diseases 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses an automatic generation platform of a disease risk evaluation report, which comprises a sample source database module, a risk prediction model module, an intelligent analysis model diagram module and a risk level analysis and evaluation module; the sample source database module is used for collecting and storing multi-parameter detection data for evaluating the risk of the sample diseases; the risk prediction model module is used for carrying out statistics, quantitative analysis and normalization index output on the data in the sample source database module; the intelligent analysis model diagram module is used for expressing the normalized index of the risk prediction model module by taking the multi-form model diagram as a diagnosis model diagram; and the risk level analysis and evaluation module is used for analyzing the multi-form model diagnosis map, and early warning and evaluating the lesion. The platform also provides application thereof, and the risk level of the malignancy of the diseases is expressed through a risk multi-level early warning color system identifier, a maple leaf model diagram and an abnormal cell analysis diagram, so that reference is provided for clinical diagnosis.
Description
Technical Field
The invention belongs to the field of disease risk evaluation, and particularly relates to an automatic generation platform of a disease risk evaluation report.
Background
Cancer is one of the biggest problems of human beings at present, and threatens the life and health of human beings. According to the survey result data of the international cancer research institution, the cancer accounts for 26.9% of the global causes of death in 2012. Statistically, in 2018, the number of new cases reaches 1810 ten thousand and the number of dead cases reaches 960 ten thousand, so that cancer has become a main cause of death and a main obstacle for prolonging the life of human beings in the global scope. Early diagnosis and treatment are the most effective anti-cancer methods, which can greatly improve patient survival and even restore health.
The existing detection report platform mainly has the following problems: the parameters for evaluating the malignancy of the disease are single, no standard diagnosis standard is input for training, and the error diagnosis is easy to judge by relying on artificial subjective experience; the quality control links are weak, medical diagnosis resources are deficient, the distribution of high-quality diagnosis resources is uneven, and the diagnosis quality control work is difficult to systematically develop; the lack of sample integral analysis, insufficient diagnosis basis of an evaluation system and easy missed diagnosis; the detection result has lower consistency level with clinical evaluation, and the reference value provided for clinic is not high.
Disclosure of Invention
In order to solve the problems, a reasonable system and a method are not available, so that the technical problem that a monitoring closed loop cannot be formed is solved, and the invention provides an automatic generation platform for a disease risk evaluation report, which comprises the following specific technical scheme:
the system comprises a sample source database module, a risk prediction model module, an intelligent analysis model diagram module and a risk level analysis and evaluation module;
the sample source database module is used for collecting and storing multi-parameter detection data during disease risk assessment of a sample;
the risk prediction model module is used for carrying out statistics, quantitative analysis and normalization index output on the data in the sample source database module by the internal multiple groups of models;
the intelligent analysis model diagram module is used for expressing the normalized index of the risk prediction model module by taking a multi-form model diagram as a diagnosis model diagram;
the risk level analysis and evaluation module is used for analyzing the multi-form model diagnosis map, and early warning and evaluating the lesions.
As an improvement, the sample source database module selects a reference method of data: the characteristic parameters of disease risk evaluation are selected, at least one group is selected, and the selection criteria are a series of characteristics which reflect the malignancy of the sample in clinical diagnosis.
As an improvement, the risk prediction model module comprises a plurality of groups of models, the number of the model groups corresponds to the number of the characteristic parameters one by one, the modeling adopts the combination of a mathematical modeling principle based on sample data application, manual labeling and deep learning technology, and each group of characteristic parameters is output to quantitatively analyze the malignancy of the sample at multiple angles.
As improvement, when the multiple groups of models are subjected to normalized index processing, a data normalization method is adopted, and characteristic parameters are mapped in a certain range through function transformation, so that the correlation processing among the multiple groups of models is facilitated.
As an improvement, the intelligent analysis model graph module performs characterization and evaluation on the malignancy of the sample through analysis and diagnosis graph modes, wherein the analysis and diagnosis graph modes comprise any one or at least two combinations: maple leaf model diagram, abnormal cell quantitative cell analysis diagram, grading risk early warning system identification and scientific grading unit.
As an improvement, the maple leaf model diagram is expressed by setting a maple leaf as a plurality of splits and branches, wherein each group of splits and forms represent a group of characteristic parameters for evaluating risk of diseases, and the integrity degree of the branches and the forms represent the etiology of the diseases, including but not limited to pathogen infection degree and pathogen infection type;
the maple leaf is provided with a plurality of groups of symmetrical or asymmetrical sawteeth, the sawteeth and the leaf have an association relationship, the height of the leaf is halved, the halving points are mapped to two sides of the leaf, the sawteeth position and the width of the bottom edge of the sawteeth are determined, the height of the sawteeth is related to the abnormal cell number of the represented parameters, and the degree of the sawteeth arrangement rule represents the malignancy degree;
the abnormal cell quantitative cell analysis chart is used for reflecting statistical data under characteristic parameters corresponding to a group of splints, expressing the statistical data in the analysis chart, and distributing the statistical data around the maple leaf model chart, wherein each group corresponds to the splints of one maple leaf model chart; wherein the statistics include, but are not limited to, abnormal cell numbers, ratios, index values; the analysis map type includes, but is not limited to, any one of a matrix map, a bar map, a line map, a scatter map, a radar map, a pie map, a polygon, a pointer map, and an irregular thermodynamic diagram;
the grading risk early warning system mark is at least one grade of comprehensive risk index early warning mark by different colors, wherein the highest grade (serious) result has high consistency with the biopsy result;
the scientific scoring unit is provided with a risk factor unit and a scoring item unit, wherein the risk factor unit and the scoring item unit of a testee are used for defining risk factor indexes to be scored, and the scoring item unit is used for defining risk values of different grades of risk factors to perform one-to-one correspondence and expression.
As an improvement, the risk factor index data of the subject's risk factor unit includes, but is not limited to, smoking data, drinking data, eating habits, oral health data, past history, family history, body mass index BMI, commonly corresponding reported values, statistics of risk factors, and changes.
As an improvement, the scoring unit correspondingly defines N levels for each risk factor index, wherein N is more than 1, the malignancy is 0-1.0, and the scoring is carried out through a risk factor scoring RFI model, wherein the risk factor scoring is carried outX i As a dangerous factor category, the coefficient a carries out weight adjustment parameters according to the actual condition of the illness.
As a specific embodiment of the invention, the invention also provides the application of the automatic disease risk evaluation report generation platform in the prevention of the upper gastrointestinal tumor cell screening risk evaluation report.
As an improvement, the characteristic parameters in the sample source database module include, but are not limited to, DNA index DI, depth of stain index SI, chromatin granularity GI, nuclear atypicality HI, pathogen VI, cell aggregation level CI, risk factor score RFI.
The beneficial effects are that: the invention provides an automatic generation platform for a disease risk evaluation report, which has the advantages compared with the prior platform that: quantitatively evaluating the risk level of the disease; the sample components are analyzed in a multi-dimensional mode, a model is established and combined with an artificial intelligence technology, parameter values of the sample are output more accurately, and a diagnosis doctor is assisted in interpreting the results; a good quality control platform is provided, so that the influence of artificial subjective factors on the diagnosis result is reduced, and the probability of misdiagnosis and missed diagnosis is effectively reduced; the method improves the coincidence degree of the cytological screening result and the histological diagnosis result, has higher clinical reference value, and more visual and clear output result, and is convenient for the cytological examination to be widely applied to common people.
Drawings
Fig. 1 is a schematic block diagram of a platform according to the present invention.
FIG. 2 is a schematic diagram of a maple leaf model in a module of the invention.
FIG. 3 is a schematic diagram of a maple leaf model split construction in a module of the invention.
FIG. 4 is a schematic diagram of a pathogen VI infection index calculation unit according to the present invention.
FIG. 5 is a graphical representation of the characterization coefficients of the pathogen VI infection index of the present invention.
FIG. 6 is an overall schematic of a diagnostic model of the present invention.
FIG. 7 is a schematic diagram of 6 parameters and cells in example 1 of the present invention.
FIGS. 8 to 11 are schematic diagrams of a bar graph, a scatter graph, a radar graph, and a pie chart of the DNA index DI in example 1 of the present invention.
FIG. 12 is a schematic diagram of a cell screening risk assessment report generated in accordance with the present invention
Fig. 13 is an image of early esophageal cancer from an endoscopic examination of a sample subject of the invention.
FIG. 14 is an image of the result of a biopsy histopathological examination of a sample subject of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to fig. 1 to 12 in the embodiments of the present invention. It is to be understood that the specific embodiments described herein are merely illustrative of the embodiments of the invention and are not limiting of the embodiments of the invention. For convenience of description, only some, but not all, of the structures related to the embodiments of the present invention are shown in the drawings.
The automatic generation platform of the disease risk evaluation report is shown in figure 1, and is characterized in that a terminal device scans a case sample and carries out data arrangement to obtain basic information of a tester and a health evaluation database; then, multi-factor variable definition and corresponding variable queue data processing are carried out on the tidied data, and classification processing including SVM classifier, establishment of cell identification analysis model and correction processing of cell analysis parameter reference value are carried out; and finally, carrying out characterization and evaluation on the malignancy of the sample by analyzing and diagnosing a graph mode.
As a specific embodiment of the invention, cells are acquired clinically through a cytological sampling method, pretreatment is completed through standardized procedures such as slice preparation and the like, cell images are digitized, and diagnostic specialists mark a large number of positive cells, normal cells, lymphocytes and pathogen infection areas by using an image marking tool, wherein the cases are negative or positive through endoscopic examination and biopsy gold standard diagnosis. The same group of data sets are marked and confirmed by not lower than 3-bit expert, a computer segmentation algorithm is applied to segment marked cell areas, and an algorithm model is used for training in combination with deep learning.
After receiving the complete sample data, the platform performs segmentation classification on the data, calls each parameter model and a risk factor scoring system to perform data processing, and outputs a series of parameter values such as DI, SI, GI, HI, VI, CI and risk factor scores after standardized normalization processing. The intelligent analysis model diagram module expresses through a risk multi-level early warning color system identifier, a maple leaf model diagram and an abnormal cell analysis diagram; and carrying out comprehensive risk index early warning identification through different colors.
The diagnosis expert marks a large number of positive cells, normal cells, lymphocytes and pathogen infection areas by using an image marking tool, wherein the negative or positive of the case is confirmed by endoscopic examination and biopsy gold standard diagnosis. The same group of data sets are marked and confirmed by a plurality of experts, a computer segmentation algorithm is applied to segment marked cell areas, a mathematical modeling principle is applied to model based on sample data, and the model is trained by combining deep learning to form a risk prediction model.
Expressing a risk multi-level early warning color system identifier, a maple leaf model diagram and an abnormal cell analysis diagram; the risk multi-level early warning color system identification is carried out by carrying out at least one level of comprehensive risk index early warning identification through different colors; the maple leaf model diagram is expressed by setting a maple leaf as a plurality of splits and branches, wherein each group of splits and forms represent a group of characteristic parameters for evaluating risks of diseases, and the integrity degree and the forms of the branches represent the causes of the diseases, such as pathogen infection degree and pathogen infection type; the maple leaf is provided with a plurality of groups of symmetrical or asymmetrical sawteeth, the sawteeth and the leaf have an association relationship, the height of the leaf is halved, the halving points are mapped to two sides of the leaf, the sawteeth position and the width of the bottom edge of the sawteeth are determined, the height of the sawteeth is related to the abnormal cell number of the represented parameters, and the degree of the sawteeth arrangement rule represents the malignancy degree;
the abnormal cell analysis chart is used for reflecting statistical data under a group of characteristic parameters corresponding to the splinters, including but not limited to abnormal cell quantity, proportion and index value, and the analysis chart is set to include but not limited to any one of a matrix chart, a column chart, a line chart, a scatter chart, a radar chart, a pie chart, a polygon, a pointer chart and an irregular thermodynamic chart;
the risk level analysis and evaluation module is used for carrying out disease risk evaluation analysis in a scientific scoring unit and maple leaf model diagram diagnosis mode; the scientific scoring unit is configured as a subject risk factor and a scoring item, wherein the subject risk factor comprises, but is not limited to, smoking data, drinking data, eating habits, oral health data, past history, family history, body Mass Index (BMI), usual places, corresponding report values, statistics of risk factors and change conditions; the scoring item is set to define different grade risk values through different score segments, and the higher the score, the greater the grade and the higher the malignancy.
Specifically, N classes, N > 1, malignancy 0-1.0, risk factor scoring (RFI) model are defined for each index for subject smoking data, drinking data, eating habits, oral health data, past history, family history, body Mass Index (BMI), and report value statistics for frequent correspondence:
RFI = sum of malignancy per index grade/number of indices.
As a specific embodiment of the invention, the automatic generation platform of the disease risk evaluation report is applied to the screening risk evaluation report of the tumor cells of the upper digestive tract.
Specifically, the characteristic parameters in the sample source database module include, but are not limited to, a nuclear area index DI, a staining depth index SI, a chromatin granularity GI, a nuclear atypicality HI, a pathogen VI, a cell aggregation level CI value, and a risk factor score RFI.
DNA Index (DI): the DNAIOD value of the tested cells/that of the normal cells is relatively stable, and the DNA content and morphology of the lymphocytes are used as the reference of the normal cells. All numbers added together by the "DNA index" divided by the number are calculated to describe the proliferation of nuclei into cells, the greater the malignancy. After calculating the mean value of the DNA, calculating the square sum of the difference between the index and the mean value of each cell again, dividing the square sum by the total number of cells of the sample, and finally obtaining the distribution variance of the cell DNA, wherein the distribution variance is used for describing the distribution and the difference of the cell nuclei DI, and the larger the distribution variance is, the higher the malignancy is.
Dyeing depth index (SI): calculating the average value of all the numerical values of the 'cell nucleus average gray scale', sequencing all the numerical values from large to small, screening the cell nucleus average gray scale with the DNA index of the same group of data as a standard, calculating the average value, and dividing the average value of the cell nucleus average gray scale with the DNA index of more than 2.5 by the average value of all the cell nucleus average gray scales to obtain a final result, wherein the final result is used for describing the cancer cell nucleus dyeing depth, and the higher the malignancy is.
Calculating the average value of all the numerical values of the cell nucleus area, sorting all the numerical values from large to small, screening the cell nucleus area with the DNA index of more than 2.5 by taking the DNA index of the same group of data as a standard, calculating the average value, and dividing the average value of the cell nucleus area with the DNA index of more than 2.5 by the average value of all the cell nucleus areas to obtain a final result which is used for describing that the larger the cancerous cell nucleus area is, the larger the malignancy is.
Chromatin Granulometry (GI): calculating the average value of all the numerical values of the 'cell nucleus variances', sequencing all the numerical values from large to small, screening the cell nucleus variances with the DNA indexes larger than 2.5 by taking the DNA indexes of the same group of data as the standard, calculating the average value, and dividing the average value of the cell nucleus variances with the DNA indexes larger than 2.5 by the average value of all the cell nucleus variances to obtain a final result which is used for describing deeper texture of the cancerous cells, and the larger the malignancy the higher.
Nuclear atypical degree (HI): and (3) circularly calculating the numerical value of each group of 'cell nucleus perimeter' divided by the numerical value of 'cell nucleus area', then carrying out root opening on the obtained value to be indicated as a, dividing the perimeter by a, and finally obtaining the value which is used for reflecting the morphological variation of the cell nucleus, wherein the greater the malignancy degree is, the higher the malignancy degree is.
Pathogen (VI): the integrity and morphology of the stems characterize the etiology of the disease, such as the extent of pathogen infection and the type of pathogen infection. No pathogen infects the stem and has no mark, and there is pathogen infects, through the morphological characterization different kinds of pathogens. Such as red, yellow, white, black dots, which characterize the number and extent of infections with a bacterial population; identifying a pathogen; identifying fungus infection; identifying a bacterial infection. The VI index calculation unit is used for identifying infected cells and counting pathogen species n and the infection index, the infection index calculation unit is shown in fig. 5, and the characterization form is shown in fig. 6.
Degree of cell aggregation (CI): the cell nest group is considered to be a distribution form of tumor cells, and cancer and precancerous cells are mutually combined due to adhesion molecules or other special reasons and are not easy to be damaged by external force, so that the adhesion index has important significance as a parameter for prompting the tumor cells.
Risk factor score (RFI): smoking data, drinking data, eating habits, oral health data, corresponding report values, statistics of risk factors and change conditions; the scoring item is set to define different grade risk values through different score segments, and the higher the score, the greater the grade and the higher the malignancy.X i As the dangerous factor category, the coefficient a is weighted according to the actual condition of the illnessReadjusting.
The invention is based on sample big data, applies artificial intelligence technology, quantifies diagnosis indexes, improves rechecking confirmation report efficiency, and creates a disease risk evaluation report system. The system is used for counting various characteristic parameters based on hundreds of millions of sample image information and establishing a hierarchical risk prediction model. And the multiple parameter algorithm models quantitatively analyze each sample, output a normalized index, establish an intelligent analysis model diagram, intuitively display risk levels, improve diagnosis efficiency and accuracy and provide a reference basis with higher value for clinical diagnosis and treatment schemes.
Example 1
Sample source database module: and (3) taking the acquired cell sample for inspection, performing laboratory full-flow treatment, forming a digitized cell image on the physical sample on the glass slide, and collecting and storing cell information by using the platform.
Risk prediction model module: the multiple groups of models used in the interior are used for carrying out statistics, quantitative analysis and normalization index output on the data in the sample source database module, and the following 6 parameters and cell diagrams are output, as shown in fig. 7:
and an intelligent analysis model diagram module: the normalization index for the risk prediction model module is expressed in a multi-form model diagram as a diagnosis model diagram.
The risk multi-stage early warning system marks: the color is divided into five stages, and the color is replaced by different colors, wherein the stage 1 is green, the stage 2 is blue, the stage 3 is yellow, the stage 4 is orange, and the stage 5 is red. The higher the level, the higher the risk.
In the maple leaf model diagram, taking 5 maple leaves as an example, as shown in the figure, each group of splits of the maple leaves respectively represent a nuclear area index DI, a dyeing depth index SI, a chromatin granularity GI, a nuclear abnormal degree HI and a cell adhesion degree CI value, and maple leaves branch and stem represent a pathogen VI.
Specifically, a maple leaf model diagram is constructed:
1) Reading a maple leaf map as shown in fig. 2, and storing picture contour information (coordinates and positions);
2) The method comprises the steps of respectively fixing the bottom ends of the five directions of the blades, analyzing the data statistics condition of each index without change, obtaining the interval distribution condition of each row of data (each index), respectively adjusting the shapes of the blades in the 5 directions, and setting the single blade as an isosceles triangle;
3) The straight line between the isosceles triangles A and Y is divided into five equal parts, and marked as 0, 2, 4, 6 and 8 marking positions respectively;
4) The parallel lines with the two sides of the triangle waist outwards are used as auxiliary lines and marked as S;
5) Respectively extending the equal points of 0, 2, 4, 6 and 8 to the S line at 45 degrees outwards, marking the connection point as C, and marking the intersection point of the isosceles triangle waist as B;
6) As shown in the figure, the center position from the point B to the point D of the next-stage mark is marked as E;
7) Connecting B, C, E points to form an outwardly protruding triangle;
the splinter and leaf stem are shown in figures 3-5.
The abnormal cell analysis chart is used for reflecting statistical data under the characteristic parameters corresponding to a group of splits, and the abnormal cell number, proportion and index value are set as a column chart, a scatter chart, a radar chart and a pie chart, as shown in figures 8-11.
The risk level analysis and evaluation module is provided with a scientific scoring unit which is used for setting risk factors and scoring items of a subject, wherein the risk factors of the subject comprise, but are not limited to, smoking data, drinking data, eating habits, oral health data, past history, family history, body quality index (BMI), frequent places, corresponding report values, statistics of the risk factors and change conditions; the scoring item is set to define different grade risk values through different score segments, and the higher the score, the greater the grade and the higher the malignancy.
Based on the automatic disease risk evaluation report generation platform, a cell screening risk evaluation report is generated, and as shown in fig. 12, the report comprises, but is not limited to, sample description content, cell number, sample satisfaction, quantitative analysis results, scatter diagrams, histograms, 10 times, 20 times and 40 times of cell images under a limiting mirror, single-cell images, multicellular images, DI, SI, GI, HI, VI, CI and other series of parameter detection values, normal reference values, diagnostic expert subjective diagnosis results and comprehensive suggestions. The report shows the highest risk, the risk level is the fifth level, and the doctor makes diagnostic advice: intraepithelial high grade lesions, seen as suspicious cancer cells, endoscopic and biopsy.
The sample subjects were confirmed by a magnifying endoscope to diagnose early esophageal carcinoma, and the diagnosis was M2-M3, as shown in FIG. 13, and the biopsy was taken under the endoscope to detect the tissue pathology, namely medium differentiation squamous cell carcinoma, and peripheral lesion high-level intraepithelial neoplasia, as shown in FIG. 14. In summary, the highest level of risk is highly consistent with cytology, endoscopy and biopsy results from the disease risk assessment report generated by the present invention.
The above examples illustrate only a few embodiments of the invention, which are described in detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (8)
1. An automatic generation platform of a disease risk evaluation report is characterized in that: the system comprises a sample source database module, a risk prediction model module, an intelligent analysis model diagram module and a risk level analysis and evaluation module;
the sample source database module is used for collecting and storing multi-parameter detection data during disease risk assessment of a sample;
the risk prediction model module is used for carrying out statistics, quantitative analysis and normalization index output on the data in the sample source database module by the internal multiple groups of models;
the intelligent analysis model diagram module is used for expressing the normalized index of the risk prediction model module by taking a multi-form model diagram as a diagnosis model diagram;
the risk level analysis and evaluation module is used for analyzing the multi-form model diagnosis map, and early warning and evaluating the lesions;
the intelligent analysis model diagram module is used for carrying out characterization and evaluation on the malignancy of a sample through analysis and diagnosis diagram modes, wherein the analysis and diagnosis diagram modes comprise any one or at least two combinations: maple leaf model diagram, abnormal cell quantitative cell analysis diagram, grading risk early warning system mark and scientific scoring unit;
the maple leaf model diagram is expressed by setting a maple leaf as a plurality of splits and branches, wherein each group of splits and forms represent a group of characteristic parameters for evaluating risks of diseases, and the integrity degree and the forms of the branches represent the causes of the diseases, including but not limited to pathogen infection degree and pathogen infection type; the maple leaf is provided with a plurality of groups of symmetrical or asymmetrical sawteeth, the sawteeth and the leaf have an association relation, the height of the leaf is halved, the halving points are mapped to two sides of the leaf, the sawteeth positions and the width of the bottom edge of the sawteeth are determined, the height of the sawteeth is related to the number of abnormal cells of the represented parameters, and the degree of the sawteeth arrangement rule represents the malignancy;
the abnormal cell quantitative cell analysis chart is used for reflecting statistical data under characteristic parameters corresponding to a group of splints, expressing the statistical data in the analysis chart, and distributing the statistical data around the maple leaf model chart, wherein each group corresponds to the splints of one maple leaf model chart; wherein the statistics include, but are not limited to, abnormal cell numbers, ratios, index values; the analysis map type includes, but is not limited to, any one of a matrix map, a bar map, a line map, a scatter map, a radar map, a pie map, a polygon, a pointer map, and an irregular thermodynamic diagram;
the grading risk early warning system marks at least one grade of comprehensive risk index early warning mark through different colors; wherein the highest-level results indicated are highly consistent with biopsy results;
the scientific scoring unit is provided with a risk factor unit and a scoring item unit, wherein the risk factor unit and the scoring item unit of a testee are used for defining risk factor indexes to be scored, and the scoring item unit is used for defining risk values of different grades of risk factors to perform one-to-one correspondence and expression.
2. The automated disease risk assessment report generation platform of claim 1, wherein: the sample source database module selects a datum method of data: the characteristic parameters of disease risk evaluation are selected, at least one group is selected, and the selection criteria are a series of characteristics which reflect the malignancy of the sample in clinical diagnosis.
3. The automated disease risk assessment report generation platform of claim 1, wherein: the risk prediction model module comprises a plurality of groups of models, the number of the model groups corresponds to the characteristic parameters one by one, the modeling adopts the combination of a mathematical modeling principle, manual labeling and a deep learning technology based on sample data, and each group of characteristic parameters is output to quantitatively analyze the malignancy of the sample at multiple angles.
4. The automated disease risk assessment report generating platform of claim 3, wherein: when the multiple groups of models are subjected to normalized index processing, a data normalization method is adopted, and characteristic parameters are mapped in a certain range through function transformation, so that the correlation processing among the multiple groups of models is facilitated.
5. The automated disease risk assessment report generation platform of claim 1, wherein: the risk factor index data of the subject's risk factor unit includes, but is not limited to, smoking data, drinking data, eating habits, oral health data, past history, family history, body mass index BMI, frequently corresponding report values, statistics of risk factors, and variations.
6. The automated disease risk assessment report generation platform of claim 1, wherein: the scoring unit correspondingly defines N levels for each risk factor index, wherein N is more than 1, the malignancy is 0-1.0, and scoring is carried out through a risk factor scoring RFI model, wherein the risk factor scoring is carried outX i As a dangerous factor category, the coefficient a carries out weight adjustment parameters according to the actual condition of the illness.
7. Use of the automated disease risk assessment report generating platform according to any one of claims 1-6 for cytological screening risk assessment reports.
8. The use according to claim 7, characterized in that: characteristic parameters in the evaluation report include, but are not limited to, single cell images, multicellular images, multiplex cell images under a microscope, DNA index DI, staining depth index SI, chromatin granularity GI, karyotype HI, pathogen VI, cell adhesion level CI, and risk factor score RFI.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010843194.7A CN112017743B (en) | 2020-08-20 | 2020-08-20 | Automatic generation platform and application of disease risk evaluation report |
| PCT/CN2020/110450 WO2022036673A1 (en) | 2020-08-20 | 2020-08-21 | Automatic disease risk evaluation and prediction report generation platform and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010843194.7A CN112017743B (en) | 2020-08-20 | 2020-08-20 | Automatic generation platform and application of disease risk evaluation report |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112017743A CN112017743A (en) | 2020-12-01 |
| CN112017743B true CN112017743B (en) | 2024-02-20 |
Family
ID=73505284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010843194.7A Active CN112017743B (en) | 2020-08-20 | 2020-08-20 | Automatic generation platform and application of disease risk evaluation report |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112017743B (en) |
| WO (1) | WO2022036673A1 (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113095376A (en) * | 2021-03-24 | 2021-07-09 | 四川大学 | Oral cavity epithelial abnormal proliferation distinguishing and grading equipment and system based on deep learning |
| CN116137185B (en) * | 2023-04-20 | 2023-06-27 | 南京引光医药科技有限公司 | Clinical examination report generation system and method |
| CN116703917B (en) * | 2023-08-07 | 2024-01-26 | 广州盛安医学检验有限公司 | Female genital tract cell pathology intelligent analysis system |
| CN117198527B (en) * | 2023-08-24 | 2024-02-23 | 北京大学人民医院 | Risk assessment system and method for transplanted related hematopoietic stem cells |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108122612A (en) * | 2017-12-20 | 2018-06-05 | 姜涵予 | The foundation of database, various dimensions health risk grade determine method and device |
| CN111192687A (en) * | 2018-11-14 | 2020-05-22 | 复旦大学附属儿科医院 | Line graph prediction model for advanced appendicitis and application thereof |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6066826B2 (en) * | 2013-05-17 | 2017-01-25 | 株式会社日立製作所 | Analysis system and health business support method |
| CN106874663A (en) * | 2017-01-26 | 2017-06-20 | 中电科软件信息服务有限公司 | Cardiovascular and cerebrovascular disease Risk Forecast Method and system |
| CN111009322B (en) * | 2019-10-21 | 2022-08-02 | 四川大学华西医院 | Perioperative risk assessment and clinical decision intelligent auxiliary system |
-
2020
- 2020-08-20 CN CN202010843194.7A patent/CN112017743B/en active Active
- 2020-08-21 WO PCT/CN2020/110450 patent/WO2022036673A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108122612A (en) * | 2017-12-20 | 2018-06-05 | 姜涵予 | The foundation of database, various dimensions health risk grade determine method and device |
| CN111192687A (en) * | 2018-11-14 | 2020-05-22 | 复旦大学附属儿科医院 | Line graph prediction model for advanced appendicitis and application thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022036673A1 (en) | 2022-02-24 |
| CN112017743A (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112017743B (en) | Automatic generation platform and application of disease risk evaluation report | |
| CN103745217B (en) | Automatic analysis method of tongue color and coating color in traditional Chinese medicine based on image retrieval | |
| CN101756696B (en) | Multiphoton skin lens image automatic analytical system and method for diagnosing malignant melanoma by using same system | |
| Bevilacqua et al. | A novel approach to evaluate blood parameters using computer vision techniques | |
| CN112088394A (en) | Computerized classification of biological tissue | |
| CN109670510A (en) | A kind of gastroscopic biopsy pathological data screening system and method based on deep learning | |
| WO2014131013A1 (en) | Cell-based tissue analysis | |
| Pan et al. | Mitosis detection techniques in H&E stained breast cancer pathological images: A comprehensive review | |
| CN101167101A (en) | Automated image analysis | |
| CN116188423B (en) | Super-pixel sparse and unmixed detection method based on pathological section hyperspectral image | |
| CN111738997A (en) | Method for calculating new coronary pneumonia lesion area ratio based on deep learning | |
| CN110148126A (en) | Blood leucocyte dividing method based on color component combination and contour fitting | |
| AU2020103474A4 (en) | A Method for Otolith Measurement and Fish Population Identification Based on Polar Angle Coordinates | |
| CN104573668B (en) | Facial color and luster adaptability automatic identifying method based on spectral reflectivity | |
| CN110766665A (en) | Tongue picture data analysis method based on strong supervision algorithm and deep learning network | |
| CN103735247A (en) | Palm color analysis method for health state identification | |
| CN109711306B (en) | Method and equipment for obtaining facial features based on deep convolutional neural network | |
| Saxena et al. | Study of Computerized Segmentation & Classification Techniques: An Application to Histopathological Imagery | |
| Pham et al. | Visualization techniques for tongue analysis in traditional Chinese medicine | |
| Zhao et al. | A backlight and deep learning based method for calculating the number of seeds per silique | |
| Ibaraki et al. | Image analysis for plants: basic procedures and techniques | |
| CN113255718B (en) | Cervical cell auxiliary diagnosis method based on deep learning cascade network method | |
| Pellegrino et al. | Development of Anemia Cells Recognition System Using Raspberry Pi | |
| Daghrir et al. | Selection of statistic textural features for skin disease characterization toward melanoma detection | |
| CN113178228A (en) | Cell analysis method based on nuclear DNA analysis, computer device, and 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |