CN109859834A - A processing method and device for predicting mortality - Google Patents
A processing method and device for predicting mortality Download PDFInfo
- Publication number
- CN109859834A CN109859834A CN201811621489.9A CN201811621489A CN109859834A CN 109859834 A CN109859834 A CN 109859834A CN 201811621489 A CN201811621489 A CN 201811621489A CN 109859834 A CN109859834 A CN 109859834A
- Authority
- CN
- China
- Prior art keywords
- preset model
- patient
- ann
- days
- training
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 16
- 208000002467 Acute-On-Chronic Liver Failure Diseases 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 37
- 239000011734 sodium Substances 0.000 claims abstract description 22
- 208000007386 hepatic encephalopathy Diseases 0.000 claims abstract description 17
- 208000002672 hepatitis B Diseases 0.000 claims abstract description 17
- DGAQECJNVWCQMB-PUAWFVPOSA-M Ilexoside XXIX Chemical compound C[C@@H]1CC[C@@]2(CC[C@@]3(C(=CC[C@H]4[C@]3(CC[C@@H]5[C@@]4(CC[C@@H](C5(C)C)OS(=O)(=O)[O-])C)C)[C@@H]2[C@]1(C)O)C)C(=O)O[C@H]6[C@@H]([C@H]([C@@H]([C@H](O6)CO)O)O)O.[Na+] DGAQECJNVWCQMB-PUAWFVPOSA-M 0.000 claims abstract description 16
- 210000002966 serum Anatomy 0.000 claims abstract description 16
- 229910052708 sodium Inorganic materials 0.000 claims abstract description 16
- 239000000427 antigen Substances 0.000 claims abstract description 15
- 102000036639 antigens Human genes 0.000 claims abstract description 15
- 108091007433 antigens Proteins 0.000 claims abstract description 15
- 102000002260 Alkaline Phosphatase Human genes 0.000 claims abstract description 14
- 108020004774 Alkaline Phosphatase Proteins 0.000 claims abstract description 14
- 238000008050 Total Bilirubin Reagent Methods 0.000 claims abstract description 14
- 230000000694 effects Effects 0.000 claims abstract description 14
- 108010094028 Prothrombin Proteins 0.000 claims abstract description 12
- 102100027378 Prothrombin Human genes 0.000 claims abstract description 12
- 229940039716 prothrombin Drugs 0.000 claims abstract description 12
- 108020004206 Gamma-glutamyltransferase Proteins 0.000 claims abstract description 10
- 102000006640 gamma-Glutamyltransferase Human genes 0.000 claims abstract description 10
- 238000001514 detection method Methods 0.000 claims abstract 9
- 238000012549 training Methods 0.000 claims description 44
- 238000012545 processing Methods 0.000 claims description 14
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims 2
- 108010062466 Enzyme Precursors Proteins 0.000 claims 1
- 102000010911 Enzyme Precursors Human genes 0.000 claims 1
- 230000015271 coagulation Effects 0.000 claims 1
- 238000005345 coagulation Methods 0.000 claims 1
- 241000700721 Hepatitis B virus Species 0.000 abstract 1
- 230000034994 death Effects 0.000 description 52
- 231100000517 death Toxicity 0.000 description 52
- 238000012360 testing method Methods 0.000 description 35
- 230000003466 anti-cipated effect Effects 0.000 description 25
- 208000006454 hepatitis Diseases 0.000 description 23
- 231100000283 hepatitis Toxicity 0.000 description 23
- 241000700605 Viruses Species 0.000 description 22
- 230000008569 process Effects 0.000 description 8
- 210000002569 neuron Anatomy 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 101710142246 External core antigen Proteins 0.000 description 5
- 230000001052 transient effect Effects 0.000 description 5
- 108090000279 Peptidyltransferases Proteins 0.000 description 4
- 238000008087 TBil Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000004393 prognosis Methods 0.000 description 4
- 125000000266 alpha-aminoacyl group Chemical group 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 2
- 206010016654 Fibrosis Diseases 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000007882 cirrhosis Effects 0.000 description 2
- 208000019425 cirrhosis of liver Diseases 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- GZCWLCBFPRFLKL-UHFFFAOYSA-N 1-prop-2-ynoxypropan-2-ol Chemical compound CC(O)COCC#C GZCWLCBFPRFLKL-UHFFFAOYSA-N 0.000 description 1
- 102000013563 Acid Phosphatase Human genes 0.000 description 1
- 108010051457 Acid Phosphatase Proteins 0.000 description 1
- 208000007848 Alcoholism Diseases 0.000 description 1
- 241000208340 Araliaceae Species 0.000 description 1
- 241000894006 Bacteria Species 0.000 description 1
- 206010057573 Chronic hepatic failure Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 208000010334 End Stage Liver Disease Diseases 0.000 description 1
- 102000009123 Fibrin Human genes 0.000 description 1
- 108010073385 Fibrin Proteins 0.000 description 1
- BWGVNKXGVNDBDI-UHFFFAOYSA-N Fibrin monomer Chemical compound CNC(=O)CNC(=O)CN BWGVNKXGVNDBDI-UHFFFAOYSA-N 0.000 description 1
- 208000012671 Gastrointestinal haemorrhages Diseases 0.000 description 1
- 102000001554 Hemoglobins Human genes 0.000 description 1
- 108010054147 Hemoglobins Proteins 0.000 description 1
- 206010019663 Hepatic failure Diseases 0.000 description 1
- 206010019799 Hepatitis viral Diseases 0.000 description 1
- 206010021036 Hyponatraemia Diseases 0.000 description 1
- 241000701076 Macacine alphaherpesvirus 1 Species 0.000 description 1
- 206010053159 Organ failure Diseases 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 125000002252 acyl group Chemical group 0.000 description 1
- 201000007930 alcohol dependence Diseases 0.000 description 1
- 239000003513 alkali Substances 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000001580 bacterial effect Effects 0.000 description 1
- 230000023555 blood coagulation Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 208000011444 chronic liver failure Diseases 0.000 description 1
- 230000004087 circulation Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 229950003499 fibrin Drugs 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 210000000232 gallbladder Anatomy 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 201000011200 hepatorenal syndrome Diseases 0.000 description 1
- 231100000334 hepatotoxic Toxicity 0.000 description 1
- 230000003082 hepatotoxic effect Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 210000003734 kidney Anatomy 0.000 description 1
- 210000004185 liver Anatomy 0.000 description 1
- 208000019423 liver disease Diseases 0.000 description 1
- 208000007903 liver failure Diseases 0.000 description 1
- 231100000835 liver failure Toxicity 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 206010034674 peritonitis Diseases 0.000 description 1
- 230000007420 reactivation Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
本发明实施例提供一种预测死亡率的处理方法及装置，所述方法包括：获取待预测患者的年龄和检测指标参数；所述检测指标参数包括碱性磷酸酶、谷氨酰转肽酶、乙型肝炎E抗原、是否出现肝性脑病、血清钠、凝血酶原活动度和总胆红素；输入所述年龄和所述检测指标参数至预设模型，将所述预设模型的输出结果作为所述待预测患者在预设天数内的死亡率预测结果；所述预设模型是基于ANN构建的。所述装置执行上述方法。本发明实施例提供的预测死亡率的处理方法及装置，通过将基于ANN构建的预设模型的输出结果作为待预测患者在预设天数内的死亡率预测结果，能够提高由乙型病毒性肝炎相关慢加急性肝衰竭导致的患者在预设天数内的死亡率的预测精度。
Embodiments of the present invention provide a processing method and device for predicting mortality. The method includes: acquiring the age and detection index parameters of a patient to be predicted; the detection index parameters include alkaline phosphatase, glutamyl transpeptidase, Hepatitis B E antigen, presence of hepatic encephalopathy, serum sodium, prothrombin activity and total bilirubin; input the age and the detection index parameters into the preset model, and convert the output results of the preset model As the mortality prediction result of the to-be-predicted patient within a preset number of days; the preset model is constructed based on ANN. The apparatus performs the above-described method. The processing method and device for predicting mortality provided by the embodiments of the present invention, by using the output result of the preset model constructed based on the ANN as the mortality prediction result of the patient to be predicted within the preset number of days, can improve the risk caused by hepatitis B virus Prediction accuracy of mortality in prespecified days due to associated acute-on-chronic liver failure.
Description
Technical field
The present embodiments relate to MEDICAL PREDICTION technical fields, and in particular to a kind of processing method and dress for predicting the death rate
It sets.
Background technique
With the growth of the disease incidence of the related acute-on-chronic liver failure (ACLF) of virus B hepatitis (HBV), to by second
The prediction of mortality caused by type virus hepatitis correlation acute-on-chronic liver failure is particularly important.
The prior art predicts the death rate of ACLF patient using end-stage liver disease model M ELD and/or modified form MELD, so
And precision of prediction is not satisfactory.The prior art is more predicted by virus B hepatitis without effective method
The death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
Therefore, drawbacks described above how is avoided, how to improve is caused by virus B hepatitis correlation acute-on-chronic liver failure
The death rate of the patient in preset number of days precision of prediction, becoming need solve the problems, such as.
Summary of the invention
In view of the problems of the existing technology, the embodiment of the present invention provides a kind of processing method and dress for predicting the death rate
It sets.
In a first aspect, the embodiment of the present invention provides a kind of processing method for predicting the death rate, which comprises
Obtain age and the Testing index parameter of patient to be predicted；The Testing index parameter includes alkaline phosphatase, paddy
Whether aminoacyl transpeptidase hepatitis B E antigen, there is hepatic encephalopathy, serum sodium, Prothrombin activity and total bilirubin；
The age and the Testing index parameter are inputted to preset model, using the output result of the preset model as
Anticipated mortality result of the patient to be predicted in preset number of days；The preset model is constructed based on ANN；Wherein,
The anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
Second aspect, the embodiment of the present invention provide a kind of processing unit for predicting the death rate, and described device includes:
Acquiring unit, for obtaining age and the Testing index parameter of patient to be predicted；The Testing index parameter includes
Whether alkaline phosphatase glutamyl transpeptidase, hepatitis B E antigen, there is hepatic encephalopathy, serum sodium, Prothrombin activity
And total bilirubin；
Predicting unit, for inputting the age and the Testing index parameter to preset model, by the preset model
Anticipated mortality result of the output result as the patient to be predicted in preset number of days；The preset model is to be based on
ANN building；Wherein, the anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment, comprising: processor, memory and bus, wherein
The processor and the memory complete mutual communication by the bus；
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to refer to
Order is able to carry out following method:
Obtain age and the Testing index parameter of patient to be predicted；The Testing index parameter includes alkaline phosphatase, paddy
Whether aminoacyl transpeptidase hepatitis B E antigen, there is hepatic encephalopathy, serum sodium, Prothrombin activity and total bilirubin；
The age and the Testing index parameter are inputted to preset model, using the output result of the preset model as
Anticipated mortality result of the patient to be predicted in preset number of days；The preset model is constructed based on ANN；Wherein,
The anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
Fourth aspect, the embodiment of the present invention provide a kind of non-transient computer readable storage medium, comprising:
The non-transient computer readable storage medium stores computer instruction, and the computer instruction makes the computer
Execute following method:
Obtain age and the Testing index parameter of patient to be predicted；The Testing index parameter includes alkaline phosphatase, paddy
Whether aminoacyl transpeptidase hepatitis B E antigen, there is hepatic encephalopathy, serum sodium, Prothrombin activity and total bilirubin；
The age and the Testing index parameter are inputted to preset model, using the output result of the preset model as
Anticipated mortality result of the patient to be predicted in preset number of days；The preset model is constructed based on ANN；Wherein,
The anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
The processing method and processing device of the prediction death rate provided in an embodiment of the present invention, it is default by that will be constructed based on ANN
The output result of model is as anticipated mortality of the patient to be predicted in preset number of days as a result, it is possible to improve by B virus
The precision of prediction of the death rate of the patient in preset number of days caused by hepatitis correlation acute-on-chronic liver failure.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the processing method flow diagram that the embodiment of the present invention predicts the death rate；
Fig. 2 is the processing device structure diagram that the embodiment of the present invention predicts the death rate；
Fig. 3 is electronic equipment entity structure schematic diagram provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Fig. 1 is the processing method flow diagram that the embodiment of the present invention predicts the death rate, as shown in Figure 1, the present invention is implemented
A kind of processing method for prediction death rate that example provides, comprising the following steps:
S101: age and the Testing index parameter of patient to be predicted are obtained；The Testing index parameter includes alkaline phosphatase
Enzyme, glutamyl transpeptidase, hepatitis B E antigen, hepatic encephalopathy, serum sodium, Prothrombin activity and total gallbladder whether occur red
Element.
Specifically, device obtains age and the Testing index parameter of patient to be predicted；The Testing index parameter includes alkali
Acid phosphatase, glutamyl transpeptidase, hepatitis B E antigen, whether occur hepatic encephalopathy, serum sodium, Prothrombin activity and
Total bilirubin.Device can be the equipment etc. for executing this method.Age corresponding A ge, alkaline phosphatase corresponding A LP, glutamy turn
Peptase, which corresponds to GGT, hepatitis B E antigen corresponds to HBeAg, hepatic encephalopathy whether occurs corresponds to HE, when can be admitted to hospital according to patient
Testing result determine whether hepatic encephalopathy；Serum sodium corresponds to NA, Prothrombin activity corresponds to PTA, total bilirubin pair
Answer TBIL.
S102: the age and the Testing index parameter are inputted to preset model, by the output knot of the preset model
Anticipated mortality result of the fruit as the patient to be predicted in preset number of days；The preset model is constructed based on ANN；
Wherein, the anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
Specifically, device inputs the age and the Testing index parameter to preset model, by the preset model
Export anticipated mortality result of the result as the patient to be predicted in preset number of days；The preset model is based on ANN
Building；Wherein, the anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.It is default
Number of days may be set according to actual conditions, and be chosen as 28 days (i.e. from 28 days for predicting to count on the day of the death rate, later) or 90
It.Artificial neural network (ANN) is made of one group of processing unit (neuron) that is highly complex and connecting each other, these processing are single
Member is related to weighting connection, it includes input, output and one or more hidden layers.The 28 days and 90 days death rates include to import
The neuron of data available, including various clinical datas, demography and laboratory data and output layer include exporting not
With the neuron of prediction result.Hidden layer is used to allow to output and input interaction complicated between neuron.
Construct the prediction model and need first to acquire data, specifically may is that the demography of patient, Laboratory Variables,
(HBV reactivation, bacterium infection, alcoholism, hemorrhage of digestive tract, hepatotoxic medication, operation), and complication (hyponatremia, it is spontaneous
Property bacterial peritonitis, hepatic encephalopathy, hepatorenal syndrome), the Organ Failures such as liver, kidney, brain, blood coagulation, circulation, breathing be
It is obtained from patient medical record or hospital database when HBV-ACLF is diagnosed and during being hospitalized.4 relevant to clinical prognosis are commented
Subsystem, including CLIF-ACLF, MELD, MELD-Na and CTP scoring, calculate in baseline.28 days and 90 days after registration
The death rate is obtained by the case history of patient or with directly contacting for household.
It should be understood that being needed pre- to being constructed based on ANN in advance before using the preset model prediction death rate
If model is trained.It is possible to further be trained using back-propagation algorithm to based on the preset model that ANN is constructed.
The advantages of artificial neural network includes self study, adaptive and reasoning process.ANN can learn from example, by changing mind
The weight connected in member connects each input with corresponding output.When applicable, input will be from first layer neuron
Each layer of neuron is traveled to, until generating output.Then adaptive process is carried out.The value of output and the value of desired output carry out
Compare.If had differences between the two values, an error signal will be generated, therefore, it is possible to use backpropagation (BP)
Algorithm changes the weight connected between neuron, to reduce the global error of network.In the training process, ANN and desired output
Error between value reduces, until reaching minimum value (i.e. network convergence).ANN can according to the knowledge accumulated in training process,
Output is generated from new input data, as reasoning process.It therefore, being capable of Accurate Prediction data by training ANN.It can lead to
The error calculated between generated and desired output valve is crossed, is calculated by learning process of the BP to ANN.Adjustment nerve
The weight connected between member, to reduce the global error of network.If the sum of square error reaches compared with cross-validation data set
Minimum then terminates training.Finally, each patient is provided 28 days and 90 days anticipated mortality results.
Further, this method can also include: that sample data is respectively divided into training data and verify data；Complete
The pairs of training data after training, the training data for completing training is verified using the verify data.With 684 patients
Data as sample data for, can be using the data of 423 patients (accounting 61.8%) therein as training data；It can
Using by the data of 261 patients (accounting 38.2%) therein as verify data.
Can also be for statistical analysis to sample data, all statistical analysis can use SPSS software (19.0 editions
This, IBM, Armonk, NY) it carries out.Assess whether sample data is normal distribution using Kolmogorov-Smirnov inspection.
28 days of follow-up and 90 days, patient was divided into Survivor and died.Continuous variable, with average value ± standard deviation or middle position
Several and quartile range.For classified variable, data are indicated with frequency or percentage.Determine that (input becomes clinical or biochemical parameter
Amount) there is the variable of statistical difference or important clinical feature as input layer with choosing after the relationship of prognosis (output variable),
Artificial neural network is constructed, to predict the ACLF patient 28-90 days death rates.Ratio calculated ratio (OR) and corresponding 95% confidence
Section (95%CI).It is analyzed using ROC curve, the estimated performance of artificial neural network in training and verifying queue is commented
Estimate, respectively using area (AUR) under ROC curve, compares the pre- of ANN and meld points-scoring system sequence using Masmaticl method
Survey performance.Think that difference is statistically significant in P < 0.05.
As a result
The baseline characteristic of patient
During this investigation it turned out, review and summarization analyzes 684 patients for being diagnosed as ACLF, average age is 43.9 ±
11.7 years old, masculinity proportion 85.1%.175 patients dead (25.6%) in follow-up in 28 days；In follow-up in 90 days,
251 deaths (36.7%).Wherein the score of CTP, CLIF-ACLF and MELD-Na are 11 (10,12) respectively in research,
38.6 ± 8.1,22.9 (20.0,26.5), and 20.0 (18.1,28.0).
ANN structure
According to 28 days and follow-up in 90 days as a result, patient is divided into existence group and died.PTA (OR=0.908,95%CI:
0.891-0.926, P < 0.001), TBil (OR=1.003,95%CI:1.002-1.004, P < 0.001), age (OR=
1.037,95%CI:01.022-1.053, P < 0.001), Na (OR=0.923,95%CI:0.892-0.955, P < 0.001),
ALP (OR=0.995,95%CI:0.991-0.999, P=0.009), GGT) (OR=0.984,95%CI:0.975-0.993,
P < 0.001), HE (OR=5.623,95%CI:2.358-10.891, P < 0.001), HBeAg positive rate (OR=0.616,95%
CI:0.429-0.884, P=0.009) with 28 days death rates there is significant correlation.All these variables are included in building ANN
In.
In addition, PTA (OR=0.922,95%CI:0.907-0.937, P < 0.001), TBil (OR=1.003,95%CI:
1.003-1.004, P < 0.001), age (OR=1.040,95%CI:1.026-1.054, P < 0.001), NA (OR=1.379,
95%CI:1.113-1.708, P=0.003), ALP (OR=0.997,95%CI:0.993-1.000, P=0.034), GGT
(OR=0.997,0.984,95%CI:0.994-0.999, P=0.005), HE (OR=2.235,95%CI:1.883-
4.837, P < 0.001) and HBeAg positive rate (OR=1.005,95%CI:1.003-1.006, P < 0.001) all with death in 90 days
The significant correlation of rate.Equally, these variables are also contained in building ANN.
Evaluation of the ANN to 28 days/90 days anticipated mortality accuracys of ACLF
For 28 days death rates in training queue, the precision of prediction of ANN (AUR 0.948,95%CI 0.925-0.970)
It is significantly higher than meld points-scoring system, including MELD (AUR0.777,95%CI:0.735-0.820, P < 0.001), MELD-na
(AUR 0.758,95%CI:0.711-0.805, P < 0.001), in verifying queue, the AUR of 28 days death rates of ANN is
0.748 (95%CI:0.673-0.822), still better than MELD (AUR 0.619,95%CI:0.536-0.701, P=0.0099),
MELD-Na (AUR 0.720,95%CI:0.642-0.799, P=0.424), CTP (AUR 0.713,95%CI:0.634-
0.792, P=0.303), CLIF-ACLF (AUR 0.696,95%CI:0.615-0.777, P=0.2004).
The precision of prediction of training 90 days death rate ANN of queue is AUR 0.913 (95%CI 0.887-0.938), significantly high
In MELD (AUR 0.765,95%CI:0.726-0.805, P < 0.001), MELD-Na (AUR0.775,95%CI:0.733-
0.817, P < 0.001), (AUR 0.712,95%CI:0.669-0.755, P < 0.001) CTP, CLIF-ACLF (AUR0.818,
95%CI:0.782-0.854, P < 0.001), respectively in the verification, the AUR of ANN be 0.754 (95%CI:
0.697-0.812), still better than MELD (AUR 0.626,95%CI:0.560-0.691, P=0.0193), MELD-Na (AUR
0.669,95%CI:0.604-0.733, P=0.1330), CTP (AUR 0.656,95%CI:0.591-0.720, P=
, and CLIF-ACLF (AUR 0.632,95%CI:0.565-0.698, P=0.264) 0.0763)
It discusses
MELD and other points-scoring systems based on MELD are mainly used for Decompensational cirrhosis patient.However, due to ACLF
It is had differences (such as death rate) with cirrhosis in many aspects, MELD etc. is not enough to predict ACLF's based on the points-scoring system of MELD
The death rate.In addition, compared with single factor test, with it is multifactor combine can more effectively predict the prognosis of ACLF.
Artificial neural network can be used to simulate complicated non-linear biosystem.The risk factors of ACLF have first been determined.
In this research, discovery is between 28~90 days dead groups and existence group, PTA, age, TBil, Na, ALP, GGT, HE, HBeAg sun
There are significant differences for property rate.Studies have shown that PTA is related to the death rate of HBV-ACLF.Age, TBil and serum sodium are also HBV-
The predictive factors of ACLF prognosis.In addition, low hemoglobin concentration is according to independent related to the development of ACLF.In addition, HBeAg positive quilt
It confirms related to more serious hepatopathy.
Then, we establish the model based on ANN, to predict HBV-ACLF 28-90 days in individual patient
The death rate.The model is trained and constructs in large quantities of HBV-ACLF patients (n=423), then in another independent team
(n=261) is verified in column.ROC analysis shows, compared with the points-scoring system based on meld, ANN model to 28,90 death
The precision of prediction of rate is higher, including MELD, MELD-na, CTP, CLIF-ACLF.This may be due to complicated, neural network
Multidimensional and nonlinear characteristic.
This research establishes prediction model, and the cross validation in different queues.It may from not concentric data
It will form an independent prognostic model.However, artificial neural network is related to a reasoning process, drawn with reducing by new data set
The mistake risen.As a result, final table provides accurate 28 days and 90 days mortality risks, higher Score on Prediction HBV-
The higher mortality risk of ACLF.
28 days HBV-ACLF and 90 days are predicted based on the model of ANN in conclusion the embodiment of the present invention constructs one
The death rate, it show compared to traditional points-scoring system based on meld superiority.
The processing method of the prediction death rate provided in an embodiment of the present invention, passes through the preset model that will be constructed based on ANN
Result is exported as anticipated mortality of the patient to be predicted in preset number of days as a result, it is possible to improve by virus B hepatitis phase
Close the precision of prediction of the death rate of the patient in preset number of days caused by acute-on-chronic liver failure.
On the basis of the above embodiments, the method also includes:
It is trained in advance to based on the preset model that ANN is constructed.
Specifically, device is trained to based on the preset model that ANN is constructed in advance.It can refer to above-described embodiment, no longer
It repeats.
The processing method of the prediction death rate provided in an embodiment of the present invention, is further able to improve by virus B hepatitis
The precision of prediction of the death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
It is on the basis of the above embodiments, described to be trained in advance to based on the preset model that ANN is constructed, comprising:
It is trained using back-propagation algorithm to based on the preset model that ANN is constructed.
Specifically, device is trained using back-propagation algorithm to based on the preset model that ANN is constructed.It can refer to above-mentioned
Embodiment repeats no more.
The processing method of the prediction death rate provided in an embodiment of the present invention, is further able to improve by virus B hepatitis
The precision of prediction of the death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
On the basis of the above embodiments, the method also includes:
Sample data is respectively divided into training data and verify data.
Specifically, sample data is respectively divided into training data and verify data by device.It can refer to above-described embodiment, no
It repeats again.
It completes to the training data to verify the training number for completing training using the verify data after training
According to.
Specifically, device is completed to the training data to verify using the verify data after training and complete instruction
Experienced training data.It can refer to above-described embodiment, repeat no more.
The processing method of the prediction death rate provided in an embodiment of the present invention, is further able to improve by virus B hepatitis
The precision of prediction of the death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
Fig. 2 is the processing device structure diagram that the embodiment of the present invention predicts the death rate, as shown in Fig. 2, the present invention is implemented
Example provides a kind of processing unit for predicting the death rate, including acquiring unit 201 and predicting unit 202, in which:
Acquiring unit 201 is used to obtain age and the Testing index parameter of patient to be predicted；The Testing index parameter packet
It includes alkaline phosphatase, glutamyl transpeptidase, hepatitis B E antigen, hepatic encephalopathy, serum sodium, factor activity whether occur
Degree and total bilirubin；Predicting unit 202 is used to input the age and the Testing index parameter to preset model, will be described pre-
If anticipated mortality result of the output result of model as the patient to be predicted in preset number of days；The preset model is
Based on ANN building；Wherein, the anticipated mortality by virus B hepatitis correlation acute-on-chronic liver failure the result is that caused
's.
Specifically, acquiring unit 201 is used to obtain age and the Testing index parameter of patient to be predicted；The Testing index
Parameter includes alkaline phosphatase, glutamyl transpeptidase, hepatitis B E antigen, hepatic encephalopathy, serum sodium, fibrin ferment whether occurs
Former mobility and total bilirubin；Predicting unit 202 is incited somebody to action for inputting the age and the Testing index parameter to preset model
Anticipated mortality result of the output result of the preset model as the patient to be predicted in preset number of days；It is described default
Model is constructed based on ANN；Wherein, the anticipated mortality is the result is that by the related slow extra urgaent dispatch liver failure of virus B hepatitis
Caused by exhausting.
The processing unit of the prediction death rate provided in an embodiment of the present invention, passes through the preset model that will be constructed based on ANN
Result is exported as anticipated mortality of the patient to be predicted in preset number of days as a result, it is possible to improve by virus B hepatitis phase
Close the precision of prediction of the death rate of the patient in preset number of days caused by acute-on-chronic liver failure.
On the basis of the above embodiments, described device is specifically used for: carrying out in advance to the preset model constructed based on ANN
Training.
Specifically, described device is specifically used for: being trained in advance to based on the preset model that ANN is constructed.
The processing unit of the prediction death rate provided in an embodiment of the present invention, is further able to improve by virus B hepatitis
The precision of prediction of the death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
On the basis of the above embodiments, described device is specifically used for: using back-propagation algorithm to being constructed based on ANN
Preset model is trained.
Specifically, described device is specifically used for: being instructed using back-propagation algorithm to based on the preset model that ANN is constructed
Practice.
The processing unit of the prediction death rate provided in an embodiment of the present invention, is further able to improve by virus B hepatitis
The precision of prediction of the death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
On the basis of the above embodiments, described device is specifically used for: by sample data be respectively divided into training data and
Verify data；It completes to the training data to verify the training number for completing training using the verify data after training
According to.
Specifically, described device is specifically used for: sample data is respectively divided into training data and verify data；It completes
After training to the training data, the training data for completing training is verified using the verify data.
The processing unit of the prediction death rate provided in an embodiment of the present invention, is further able to improve by virus B hepatitis
The precision of prediction of the death rate of the patient caused by related acute-on-chronic liver failure in preset number of days.
The processing unit of the prediction death rate provided in an embodiment of the present invention specifically can be used for executing above-mentioned each method and implement
The process flow of example, details are not described herein for function, is referred to the detailed description of above method embodiment.
Fig. 3 is electronic equipment entity structure schematic diagram provided in an embodiment of the present invention, as shown in figure 3, the electronic equipment
It include: processor (processor) 301, memory (memory) 302 and bus 303；
Wherein, the processor 301, memory 302 complete mutual communication by bus 303；
The processor 301 is used to call the program instruction in the memory 302, to execute above-mentioned each method embodiment
Provided method, for example, obtain age and the Testing index parameter of patient to be predicted；The Testing index parameter includes
Whether alkaline phosphatase glutamyl transpeptidase, hepatitis B E antigen, there is hepatic encephalopathy, serum sodium, Prothrombin activity
And total bilirubin；The age and the Testing index parameter are inputted to preset model, by the output result of the preset model
As anticipated mortality result of the patient to be predicted in preset number of days；The preset model is constructed based on ANN；Its
In, the anticipated mortality is the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
The present embodiment discloses a kind of computer program product, and the computer program product includes being stored in non-transient calculating
Computer program on machine readable storage medium storing program for executing, the computer program include program instruction, when described program instruction is calculated
When machine executes, computer is able to carry out method provided by above-mentioned each method embodiment, for example, obtains patient's to be predicted
Age and Testing index parameter；The Testing index parameter include alkaline phosphatase, glutamyl transpeptidase, hepatitis B E antigen,
Whether hepatic encephalopathy, serum sodium, Prothrombin activity and total bilirubin are occurred；Input the age and Testing index ginseng
Number is pre- as the death rate of the patient to be predicted in preset number of days using the output result of the preset model to preset model
Survey result；The preset model is constructed based on ANN；Wherein, the anticipated mortality is the result is that by virus B hepatitis
Caused by related acute-on-chronic liver failure.
The present embodiment provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage medium
Computer instruction is stored, the computer instruction makes the computer execute method provided by above-mentioned each method embodiment, example
It such as include: age and the Testing index parameter for obtaining patient to be predicted；The Testing index parameter includes alkaline phosphatase, paddy ammonia
Whether acyl transpeptidase hepatitis B E antigen, there is hepatic encephalopathy, serum sodium, Prothrombin activity and total bilirubin；Input
The age and the Testing index parameter are to preset model, using the output result of the preset model as the trouble to be predicted
Anticipated mortality result of the person in preset number of days；The preset model is constructed based on ANN；Wherein, the death rate is pre-
It surveys the result is that as caused by virus B hepatitis correlation acute-on-chronic liver failure.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
The various media that can store program code such as disk.
The embodiments such as electronic equipment described above are only schematical, wherein it is described as illustrated by the separation member
Unit may or may not be physically separated, and component shown as a unit may or may not be object
Manage unit, it can it is in one place, or may be distributed over multiple network units.It can select according to the actual needs
Some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying wound
In the case where the labour for the property made, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above various embodiments is only to illustrate the technical solution of the embodiment of the present invention, rather than it is right
It is limited；Although the embodiment of the present invention is described in detail referring to foregoing embodiments, the ordinary skill of this field
Personnel are it is understood that it is still possible to modify the technical solutions described in the foregoing embodiments, or to part
Or all technical features are equivalently replaced；And these are modified or replaceed, it does not separate the essence of the corresponding technical solution
The range of various embodiments of the present invention technical solution.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811621489.9A CN109859834A (en) | 2018-12-28 | 2018-12-28 | A processing method and device for predicting mortality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811621489.9A CN109859834A (en) | 2018-12-28 | 2018-12-28 | A processing method and device for predicting mortality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109859834A true CN109859834A (en) | 2019-06-07 |
Family
ID=66892892
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811621489.9A Pending CN109859834A (en) | 2018-12-28 | 2018-12-28 | A processing method and device for predicting mortality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859834A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111627559A (en) * | 2020-06-17 | 2020-09-04 | 北京大学第三医院（北京大学第三临床医学院） | System for predicting patient mortality risk |
| CN113077888A (en) * | 2021-03-30 | 2021-07-06 | 福州宜星大数据产业投资有限公司 | Automatic MELD scoring method, terminal and storage medium |
| RU2761729C1 (en) * | 2021-04-26 | 2021-12-13 | Вячеслав Леонидович Коробка | Method for assessing the risk of death in liver cirrhosis |
| CN114795114A (en) * | 2022-03-29 | 2022-07-29 | 电子科技大学 | A predictive method for late-onset encephalopathy in carbon monoxide poisoning based on multimodal learning |
| CN118888152A (en) * | 2024-07-09 | 2024-11-01 | 重庆医科大学附属第二医院 | A LATI prognostic model for patients with liver failure and its construction method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102472756A (en) * | 2009-07-31 | 2012-05-23 | 百奥科瑞茨生命科学公司 | Method for predicting the likelihood of an onset of an inflammation associated organ failure |
| CN103038772A (en) * | 2010-03-15 | 2013-04-10 | 新加坡保健服务集团有限公司 | Method of predicting the survivability of a patient |
| CN106153923A (en) * | 2016-06-21 | 2016-11-23 | 北京大学第医院 | Prediction glutamate pyruvate transaminase is less than the system of the Patients with Chronic Hepatitis B inflammation degree of two times of upper limits of normal |
-
2018
- 2018-12-28 CN CN201811621489.9A patent/CN109859834A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102472756A (en) * | 2009-07-31 | 2012-05-23 | 百奥科瑞茨生命科学公司 | Method for predicting the likelihood of an onset of an inflammation associated organ failure |
| CN103038772A (en) * | 2010-03-15 | 2013-04-10 | 新加坡保健服务集团有限公司 | Method of predicting the survivability of a patient |
| CN106153923A (en) * | 2016-06-21 | 2016-11-23 | 北京大学第医院 | Prediction glutamate pyruvate transaminase is less than the system of the Patients with Chronic Hepatitis B inflammation degree of two times of upper limits of normal |
Non-Patent Citations (2)
| Title |
|---|
| M.-H. ZHEN: ""A model to predict 3-month mortality risk of acute-on-chronic hepatitis B liver failure using artificial neural network"", 《JOURNAL OF VIRAL HEPATITIS》 * |
| 刘磊: ""不同评分系统对乙肝相关慢加急性肝衰竭患者短期预后评估的比较"", 《实用医学杂志》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111627559A (en) * | 2020-06-17 | 2020-09-04 | 北京大学第三医院（北京大学第三临床医学院） | System for predicting patient mortality risk |
| CN111627559B (en) * | 2020-06-17 | 2023-08-29 | 北京大学第三医院（北京大学第三临床医学院） | System for predicting patient mortality risk |
| CN113077888A (en) * | 2021-03-30 | 2021-07-06 | 福州宜星大数据产业投资有限公司 | Automatic MELD scoring method, terminal and storage medium |
| RU2761729C1 (en) * | 2021-04-26 | 2021-12-13 | Вячеслав Леонидович Коробка | Method for assessing the risk of death in liver cirrhosis |
| CN114795114A (en) * | 2022-03-29 | 2022-07-29 | 电子科技大学 | A predictive method for late-onset encephalopathy in carbon monoxide poisoning based on multimodal learning |
| CN118888152A (en) * | 2024-07-09 | 2024-11-01 | 重庆医科大学附属第二医院 | A LATI prognostic model for patients with liver failure and its construction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109859834A (en) | A processing method and device for predicting mortality | |
| JP6463572B1 (en) | Support system to estimate the internal state of the target system | |
| WO2022111385A1 (en) | Graph neural network-based clinical omics data processing method and apparatus, device, and medium | |
| Ikemura et al. | Using automated machine learning to predict the mortality of patients with COVID-19: prediction model development study | |
| Jerlin Rubini et al. | Efficient classification of chronic kidney disease by using multi‐kernel support vector machine and fruit fly optimization algorithm | |
| Nahar et al. | A comparative analysis of the ensemble method for liver disease prediction | |
| JP6916107B2 (en) | Bayesian Causal Network Model for Health Examination and Treatment Based on Patient Data | |
| JP6931897B2 (en) | A method for generating a prediction result for predicting the occurrence of a fatal symptom of a subject at an early stage, and a device using the prediction result. | |
| Masood et al. | Diversity-inducing policy gradient: Using maximum mean discrepancy to find a set of diverse policies | |
| CN112786204A (en) | Machine learning diabetes onset risk prediction method and application | |
| CN112786203A (en) | Machine learning diabetic retinopathy morbidity risk prediction method and application | |
| CN116403714B (en) | Cerebral apoplexy END risk prediction model building method and device, END risk prediction system, electronic equipment and medium | |
| Day et al. | Inflammation and disease: modelling and modulation of the inflammatory response to alleviate critical illness | |
| Foteinou et al. | Translational potential of systems‐based models of inflammation | |
| Zhou et al. | Integration of artificial intelligence and multi-omics in kidney diseases | |
| CN113990502A (en) | A prediction system of ICU heart failure based on heterogeneous graph neural network | |
| Mohammadi et al. | Learning to identify patients at risk of uncontrolled hypertension using electronic health records data | |
| Vyas et al. | Chronic kidney disease prediction using robust approach in machine learning | |
| US20220013218A1 (en) | Patient Treatment Resource Utilization Predictor | |
| Dong et al. | Readmission prediction of diabetic patients based on AdaBoost-RandomForest mixed model | |
| Holmes | Knowledge discovery in biomedical data: theory and methods | |
| KR102461582B1 (en) | Computer program for providing diagnostic information corresponding to examination information using artificial intelligence | |
| CN115547489A (en) | Method for predicting death risk of maintenance hemodialysis patient | |
| Wang et al. | Prediction of target range of intact parathyroid hormone in hemodialysis patients with artificial neural network | |
| Banujan et al. | Boosting ensemble machine learning approach for covid-19 death prediction |
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 |