CN115762764A - HIV negative cryptococcus meningitis treatment outcome prediction model and construction method thereof - Google Patents
HIV negative cryptococcus meningitis treatment outcome prediction model and construction method thereof Download PDFInfo
- Publication number
- CN115762764A CN115762764A CN202211487273.4A CN202211487273A CN115762764A CN 115762764 A CN115762764 A CN 115762764A CN 202211487273 A CN202211487273 A CN 202211487273A CN 115762764 A CN115762764 A CN 115762764A
- Authority
- CN
- China
- Prior art keywords
- cryptococcus
- cerebrospinal fluid
- negative
- data
- variable
- 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
Images
Classifications
-
- 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 provides an HIV negative cryptococcus meningitis treatment outcome prediction model and a construction method thereof, and belongs to the technical field of biology. The treatment outcome prediction model is constructed based on multi-dimensional data of HIV negative cryptococcal meningitis patients, so that an online open platform for accurately predicting the treatment effect of the HIV negative cryptococcal meningitis patients can be established, and clinicians can rapidly obtain the prediction probability of good and bad treatment effect of the patients and the high and low risk stratification of the patients by inputting the current 8 prediction index values of the patients, so that the clinical decision is facilitated, and the personalized treatment scheme is timely provided for the patients.
Description
Technical Field
The invention belongs to the technical field of biology, and particularly relates to an HIV negative cryptococcus meningitis treatment outcome prediction model and a construction method thereof.
Background
Cryptococcal meningitis is a disease of central nervous system infectivity caused by cryptococcus infection, and the incidence rate is very high in patients with significantly low immune function. Cryptococcal meningitis is one of the most common infections in HIV/AIDS patients, with high mortality and poorer prognosis. Research shows that the short-term mortality rate of HIV-negative cryptococcal meningitis patients is higher than that of HIV-positive cryptococcal meningitis patients, and for the HIV-negative part of the specific cryptococcal meningitis patient population, no relevant model is available for early prediction of treatment outcome of patients at present.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a model for predicting the outcome of treatment of HIV-negative cryptococcus meningitis and a construction method thereof. The invention establishes an early risk prediction model for the treatment outcome of HIV-negative cryptococcal meningitis patients, and can be used for searching high-risk people and assisting clinical decision and treatment.
In order to achieve the purpose, the invention adopts the technical scheme that: a construction method of an HIV negative cryptococcus meningitis treatment outcome prediction model comprises the following steps:
s1, collecting clinical data of a patient with HIV-negative cryptococcus meningitis as candidate characteristic data variables;
s2, the treatment effect of 10 weeks after antifungal treatment of the HIV-negative cryptococcal meningitis patient is taken as the treatment outcome of the cryptococcal meningitis patient;
s3, performing data analysis processing on the candidate characteristic data;
s4, screening by using a minimum absolute shrinkage selection operator to obtain an optimal characteristic data variable;
s5, performing model construction on a training set by adopting a machine learning model of a random forest, an extreme gradient lifting tree, logistic regression, gaussian naive Bayes, K-neighborhood and multilayer perceptron;
and S6, screening out an optimal model through internal verification and comprehensive evaluation.
As a preferred embodiment of the construction method of the present invention, the clinical data in step S1 include age, sex, smoking history, drinking history, body mass index, headache, fever, vomiting, stiff neck, visual impairment, hearing impairment, change of consciousness state, intracranial pressure, baseline count of cryptococcus cerebrospinal fluid, mean value of cryptococcus count within 2 weeks of hospitalization, slope of change of cryptococcus count within 2 weeks of hospitalization, cerebrospinal fluid culture state, cerebrospinal fluid leucocytes, cerebrospinal fluid proteins, cerebrospinal fluid sugars, cerebrospinal fluid chlorides, hemoglobin, blood leukocytes, platelets, blood potassium, glutamic pyruvic transaminase, glutamic-oxalacetic transaminase, blood urea nitrogen, creatinine, meningeal reinforcement, encephalic parenchyma involvement, hydrocephalus, immune hypofunction, time from symptom occurrence to hospitalization, number of comorbidities, and number of common clinical symptoms.
As a preferred embodiment of the construction method according to the invention, the outcome of the treatment in step S2 is classified as successful or unsuccessful, wherein unsuccessful is defined as death, disease progression or relapse; success is defined as improvement of clinical symptoms of patients after cerebrospinal fluid culture and negative change after positive staining of ink.
As a preferred embodiment of the construction method of the present invention, the data analysis processing of step S3 includes deleting Gao Queshi variable, deleting high correlation variable, data padding, data transformation, and data normalization.
As a preferred embodiment of the construction method of the present invention, the deletion Gao Queshi variable is a variable with a deletion rate of more than 30% in the deleted data set; the deletion high correlation variable is a variable with a deletion variable correlation coefficient larger than 0.8; the data filling is a continuous variable using a mean filling method, and a classification variable using a population filling method; the data transformation performs Yeo-Johnson normality transformation on all continuous variables; the data is normalized by centering and normalizing all continuous variables.
As a preferred embodiment of the construction method of the present invention, the optimal characteristic data variables obtained by screening in step S4 include baseline cryptococcus cerebrospinal fluid count, change in consciousness state, cerebrospinal fluid leukocytes, change slope of cryptococcus count within 2 weeks of hospital admission, hearing impairment, intracranial pressure, cerebrospinal fluid chloride, and glutamic-oxaloacetic transaminase.
As a preferred embodiment of the construction method of the present invention, the internal verification in step S6 includes verifying an optimal hyper-parameter and an optimal padding mode of each model.
As a preferred embodiment of the construction method of the present invention, the comprehensive evaluation of step S6 includes AUC, sensitivity, specificity, accuracy, positive predictive value, negative predictive value and F1 value.
The invention also provides an HIV negative cryptococcus meningitis treatment outcome prediction model constructed by the construction method.
The invention also provides application of the HIV-negative cryptococcus meningitis treatment outcome prediction model in evaluation of treatment effect of HIV-negative cryptococcus meningitis patients.
The invention has the beneficial effects that: the invention provides an HIV negative cryptococcus meningitis treatment outcome prediction model and a construction method thereof, the invention is a treatment outcome prediction model constructed based on multi-dimensional data of HIV negative cryptococcus meningitis patients, so that an online open platform for accurately predicting the treatment effect of HIV negative cryptococcus meningitis patients can be established, and clinicians can rapidly obtain the prediction probability of good and bad treatment effect of patients and the high and low risk stratification of the patients by inputting the current 8 prediction index values of the patients, thereby facilitating the auxiliary clinical decision and providing a personalized treatment scheme for the patients in time.
Drawings
FIG. 1 is a graph of correlation coefficients between variables;
FIG. 2 is a LASSO feature screening process;
FIG. 3 is a graph of feature importance ranking based on permatation;
FIG. 4 is a predictive model web calculator interface.
Detailed Description
In order to more concisely and clearly demonstrate technical solutions, objects and advantages of the present invention, the following detailed description of the technical solutions of the present invention is provided with reference to specific embodiments and accompanying drawings.
Example 1
The embodiment provides a method for constructing an HIV-negative cryptococcal meningitis treatment outcome prediction model, which comprises the following specific steps:
1. clinical data were collected from patients diagnosed with HIV-negative cryptococcal meningitis as a training set, with the following 36 clinical features: 1) Demographic characteristics: age, gender, smoking history, drinking history, body Mass Index (BMI); 2) The clinical common symptoms are: including headache, fever, vomiting, neck rigidity, visual disturbance, hearing disturbance and change of consciousness state; 3) Laboratory examination: intracranial pressure, baseline cerebrospinal fluid cryptococcus count, average cryptococcus count within 2 weeks of hospitalization, change slope of cryptococcus count within 2 weeks of hospitalization, cerebrospinal fluid culture state, cerebrospinal fluid leukocytes, cerebrospinal fluid proteins, cerebrospinal fluid sugars, cerebrospinal fluid chlorides, hemoglobin, blood leukocytes, platelets, blood potassium, glutamic-pyruvic transaminase, glutamic-oxalacetic transaminase, blood urea nitrogen, creatinine; 4) Imaging examination: strengthening meninges, involvement of brain parenchyma, hydrocephalus; 5) And others: whether the immune function is low (such as liver disease, long-term use of corticosteroids, autoimmune diseases, type II diabetes, chronic kidney diseases or organ transplantation and the like), the time from the appearance of symptoms to the clinic, the number of common diseases and the number of clinical common symptoms. The outcome to be predicted is a categorical variable indicating that the outcome of treatment for cryptococcal meningitis patients was assessed 10 weeks after the start of antifungal therapy, as successful or unsuccessful. Unsuccessful definition is death, disease progression (increased disturbance of consciousness, fever, headache or other clinical manifestations, and persistent positive cerebrospinal fluid culture status) or relapse (cerebrospinal fluid culture status turns negative and then positive again, infection symptoms and signs disappear and then reappear); success is defined as the improvement of clinical symptoms of patients after cerebrospinal fluid culture and negative change after positive inkblot staining.
2. Deleting variables with variable deletion rate more than 30% in the data set: BMI, glutamic-pyruvic transaminase.
3. Deleting variables with variable correlation coefficients r larger than 0.8: correlation coefficient between variables as shown in fig. 1, the correlation coefficient r =0.87 between baseline cryptococcus cerebrospinal fluid count and mean cryptococcus count within 2 weeks of admission, thus deleting mean cryptococcus count within 2 weeks of admission.
4. Data preprocessing:
(1) According to the type of clinical variables, the missing data is processed by different filling methods, a mean value filling method is used for continuous variables (age, BMI, time from symptom occurrence to visit, cerebrospinal fluid leucocyte, cerebrospinal fluid protein, cerebrospinal fluid sugar, cerebrospinal fluid chloride, hemoglobin, leucocyte, platelet, blood potassium, glutamic pyruvic transaminase, glutamic oxalacetic transaminase, blood urea nitrogen, creatinine, baseline cerebrospinal fluid cryptococcus count, cryptococcus count mean value within 2 weeks of hospital admission and cryptococcus count change slope within 2 weeks of hospital admission), and a mode filling method is used for classification variables (sex, smoking history, drinking history, headache, fever, vomiting, neck rigidity, vision disorder, hearing disorder, consciousness state change, cerebrospinal fluid culture state, intracranial pressure, meningeal strengthening, brain parenchyma involvement, hydrocephalus, low immunity, number of comorbidities and number of clinical common symptoms). The coding of the categorical variables is detailed in table 1.
TABLE 1 Specification Table of encoding rules for categorical variables
a, low immunity: liver disease, chronic use of corticosteroids, autoimmune disease, type 2 diabetes, chronic kidney disease, or organ transplantation.
b, co-morbidity: including tumor, hypertension, liver disease, coronary heart disease, cerebrovascular disease, long-term use of corticosteroid, autoimmune disease, type 2 diabetes, and chronic kidney disease.
c clinically common symptoms: including headache, fever, vomiting, stiff cervical spine, visual disturbance, hearing disturbance, mental state change.
(2) And carrying out Yeo-Johnson normal transformation on all continuous variables, wherein the operation formula is as follows:
(3) Centralizing and normalizing all continuous variables;
(4) All feature variables with zero variance are deleted.
5. And (3) feature screening: in the feature screening process, all candidate predictive variables are screened by using LASSO regression, an optimal regularization parameter lambda is selected based on 10-fold cross validation, a regression loss function is set as a mean square error, and the optimal lambda of a regression model is obtained by taking the simplest model which can be obtained under the condition of sacrificing one-time standard deviation as an evaluation index. The LASSO feature selection process is detailed in FIG. 2. As shown in fig. 3, the variables corresponding to the finally screened 8 non-sparse coefficients from high to low are respectively: baseline cryptococcus cerebrospinal fluid count, change in state of consciousness, cerebrospinal fluid leukocytes, slope of change in cryptococcus count within 2 weeks of admission, hearing impairment, intracranial pressure, cerebrospinal fluid chloride, glutamic oxaloacetic transaminase. Higher weight values demonstrate greater importance of the variable.
6. Constructing a model: the 8 screened prediction indexes are modeled by using a sklern library of python software, and the method comprises the following specific steps:
(1) Random sampling 8 is performed on the training set without deletion padding: 2, dividing the ratio into a development set and an internal test set; and simultaneously, carrying out model construction on the training set by 6 machine learning models such as RF, XGboost, logistic Regression, gaussianNB, KNN and MLP. The optimal hyper-parameters of each algorithm are searched based on grid search, and the range of the grid search hyper-parameters is set and shown in table 2. And on the basis of the over-parameter range set by the grid search, 3 data filling modes such as KNN filling, mean filling, median filling and the like are respectively added.
TABLE 2 grid search out-of-parameter settings for each machine learning algorithm
(2) And 5-fold cross validation is used, and AUC is used as a main evaluation index to select the optimal hyper-parameter of each model. The optimal hyper-parameters and filling modes of each algorithm are shown in table 3.
TABLE 3 optimal hyper-parameters and fill-in modes for each algorithm
(3) And carrying out internal verification on each algorithm based on the optimal hyper-parameter and the optimal filling mode. The prediction performance was comprehensively evaluated using a plurality of indices such as AUC, sensitivity (Sensitivity), specificity (Specificity), accuracy (Accuracy), positive Predictive Value (PPV), negative Predictive Value (NPV), and F1 Value (F1 Score). In the confusion matrix composed of the gold standard outcome label and the classifier prediction label, samples are classified into four types, namely True Positive (TP), false Positive (FP), false Negative (FN) and True Negative (TN).
The calculation formula of each evaluation index is as follows:
sensitivity = Recall (Recall, R) = TP/(TP + FN) × 100%
Specificity = TN/(FP + TN). Times.100%
Accuracy = (TP + TN)/(TP + FP + TN + FN). Times.100%
PPV = Precision (Precision, P) = TP/(TP + FP) × 100%
NPV＝TN/(FN+TN)×100％
F1 Score＝(2×P×R)/(P+R)
The predicted performance of each algorithm on the internal validation set is detailed in table 4.
TABLE 4 predictive Performance of algorithms on an internal validation set
* Note: the threshold is chosen based on criteria that balance sensitivity and specificity.
And finally selecting a Logistic Regression algorithm as an optimal algorithm by comprehensively evaluating and comparing various evaluation indexes of the algorithms, and carrying out subsequent external verification and model popularization and application on the model.
Example 2
A webpage calculator is developed by adopting the HIV-negative cryptococcus meningitis treatment outcome prediction model constructed in the embodiment 1 to predict the HIV-negative cryptococcus meningitis treatment outcome (the publicly accessible website is http://8.134.96.171 8507 /), and the specific operation method is shown in FIG. 4: and respectively inputting the 8 numerical values of the interface, wherein the first 3 indexes are input in a pull-down menu click mode, and the last 5 indexes are input in a manual mode. Clicking on "Prediction Start!after all inputs! By running the prediction model of the invention, the prediction probability of poor prognosis of the HIV-negative cryptococcal meningitis patient can be obtained (for example, the probability of poor prognosis of the patient in the interface is 75.595%).
Finally, it should be noted that the above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, many variations and modifications are possible without departing from the inventive concept, which falls within the scope of the invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A construction method of an HIV negative cryptococcus meningitis treatment outcome prediction model is characterized by comprising the following steps:
s1, collecting clinical data of a patient with HIV negative cryptococcus meningitis as candidate characteristic data variables;
s2, the treatment effect of 10 weeks after antifungal treatment of the HIV-negative cryptococcal meningitis patient is taken as the treatment outcome of the cryptococcal meningitis patient;
s3, performing data analysis processing on the candidate characteristic data;
s4, screening by using a minimum absolute shrinkage selection operator to obtain an optimal characteristic data variable;
s5, performing model construction on a training set by adopting a machine learning model of a random forest, an extreme gradient lifting tree, logistic regression, gaussian naive Bayes, K-neighborhood and multilayer perceptron;
and S6, screening out an optimal model through internal verification and comprehensive evaluation.
2. The method according to claim 1, wherein the clinical data in step S1 include age, sex, smoking history, drinking history, body mass index, headache, fever, vomiting, stiff neck, visual impairment, hearing impairment, change in consciousness state, intracranial pressure, baseline cerebrospinal fluid cryptococcus count, mean cryptococcus count within 2 weeks of hospitalization, slope of change in cryptococcus count within 2 weeks of hospitalization, cerebrospinal fluid culture status, cerebrospinal fluid leukocytes, cerebrospinal fluid proteins, cerebrospinal fluid sugar, cerebrospinal fluid chloride, hemoglobin, leukocytes, platelets, potassium blood, glutamic pyruvic transaminase, glutamic-oxalacetic transaminase, blood urea nitrogen, creatinine, meningeal reinforcement, brain parenchyma involvement, hydrocephalus, immune function deficiency, symptom occurrence-to-diagnosis time, number of comorbidities, and number of common clinical symptoms.
3. The method of constructing according to claim 1, wherein the outcome of treatment in step S2 is classified as successful or unsuccessful, wherein unsuccessful is defined as death, disease progression or recurrence; success is defined as improvement of clinical symptoms of patients after cerebrospinal fluid culture and negative change after positive staining of ink.
4. The building method according to claim 1, wherein the data analysis processing of step S3 includes deleting Gao Queshi variables, deleting high correlation variables, data padding, data transformation, and data normalization.
5. The construction method according to claim 4, wherein the deletion Gao Queshi variable is a variable with deletion rate of more than 30% in a deletion dataset; deleting the high-correlation variable, wherein the deletion variable correlation coefficient is a variable with a deletion variable correlation coefficient larger than 0.8; the data filling is a continuous variable using a mean filling method, and a classification variable using a mode filling method; the data transformation carries out Yeo-Johnson normality transformation on all continuous variables; the data normalization is to centralize and normalize all continuous variables.
6. The constructing method of claim 1, wherein the optimal characteristic data variables obtained by screening in step S4 include baseline cryptococcus cerebrospinal fluid count, change in consciousness state, cerebrospinal fluid leukocytes, change slope of cryptococcus count within 2 weeks of hospital admission, hearing impairment, intracranial pressure, cerebrospinal fluid chloride, glutamic-oxaloacetic transaminase.
7. The building method according to claim 1, wherein the internal verification of step S6 includes verifying an optimal hyper-parameter and an optimal padding mode of each model.
8. The method of constructing according to claim 1, wherein the comprehensive evaluation of step S6 includes AUC, sensitivity, specificity, accuracy, positive predictive value, negative predictive value, and F1 value.
9. An HIV-negative cryptococcus meningitis treatment outcome prediction model constructed by the construction method of any one of claims 2 to 8.
10. Use of the HIV-negative cryptococcal meningitis treatment outcome prediction model of claim 9 in assessing the efficacy of a treatment on a patient with HIV-negative cryptococcal meningitis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211487273.4A CN115762764A (en) | 2022-11-25 | 2022-11-25 | HIV negative cryptococcus meningitis treatment outcome prediction model and construction method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211487273.4A CN115762764A (en) | 2022-11-25 | 2022-11-25 | HIV negative cryptococcus meningitis treatment outcome prediction model and construction method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115762764A true CN115762764A (en) | 2023-03-07 |
Family
ID=85338831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211487273.4A Pending CN115762764A (en) | 2022-11-25 | 2022-11-25 | HIV negative cryptococcus meningitis treatment outcome prediction model and construction method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115762764A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117238522A (en) * | 2023-11-08 | 2023-12-15 | 查理高特(青岛)健康科技有限公司 | Febuxostat curative effect prediction system, febuxostat curative effect prediction equipment and febuxostat curative effect prediction medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103492590A (en) * | 2011-02-22 | 2014-01-01 | 卡里斯生命科学卢森堡控股有限责任公司 | Circulating biomarkers |
| US20150211070A1 (en) * | 2011-09-22 | 2015-07-30 | Immu-Metrix, Llc | Compositions and methods for analyzing heterogeneous samples |
| CN106102781A (en) * | 2013-08-29 | 2016-11-09 | 英联邦高等教育系统天普大学 | Methods and compositions for rna-guided treatment of hiv infection |
| CN112309576A (en) * | 2020-09-22 | 2021-02-02 | 江南大学 | Colorectal cancer survival period prediction method based on deep learning CT (computed tomography) image omics |
| CN113707327A (en) * | 2021-08-25 | 2021-11-26 | 景元明 | Multi-tumor marker tumor diagnosis model based on medical big data |
| CN114188014A (en) * | 2021-09-28 | 2022-03-15 | 中国医学科学院阜外医院 | Patient hospital unhealthy prognosis prediction model construction method, system and application |
| CN115029410A (en) * | 2021-03-04 | 2022-09-09 | 中国科学院微生物研究所 | Application of enterococcus faecium as marker for predicting clinical outcome and prognosis of respiratory severe patients |
| CN115295151A (en) * | 2022-09-01 | 2022-11-04 | 中山大学附属第三医院 | Sepsis prediction system, prediction model construction method, system and kit |
-
2022
- 2022-11-25 CN CN202211487273.4A patent/CN115762764A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103492590A (en) * | 2011-02-22 | 2014-01-01 | 卡里斯生命科学卢森堡控股有限责任公司 | Circulating biomarkers |
| US20150211070A1 (en) * | 2011-09-22 | 2015-07-30 | Immu-Metrix, Llc | Compositions and methods for analyzing heterogeneous samples |
| CN106102781A (en) * | 2013-08-29 | 2016-11-09 | 英联邦高等教育系统天普大学 | Methods and compositions for rna-guided treatment of hiv infection |
| CN112309576A (en) * | 2020-09-22 | 2021-02-02 | 江南大学 | Colorectal cancer survival period prediction method based on deep learning CT (computed tomography) image omics |
| CN115029410A (en) * | 2021-03-04 | 2022-09-09 | 中国科学院微生物研究所 | Application of enterococcus faecium as marker for predicting clinical outcome and prognosis of respiratory severe patients |
| CN113707327A (en) * | 2021-08-25 | 2021-11-26 | 景元明 | Multi-tumor marker tumor diagnosis model based on medical big data |
| CN114188014A (en) * | 2021-09-28 | 2022-03-15 | 中国医学科学院阜外医院 | Patient hospital unhealthy prognosis prediction model construction method, system and application |
| CN115295151A (en) * | 2022-09-01 | 2022-11-04 | 中山大学附属第三医院 | Sepsis prediction system, prediction model construction method, system and kit |
Non-Patent Citations (1)
| Title |
|---|
| 黄薇 等: "艾滋病合并隐球菌脑膜炎的诊断及治疗", 中国真菌学杂志, vol. 16, no. 2, pages 131 - 134 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117238522A (en) * | 2023-11-08 | 2023-12-15 | 查理高特(青岛)健康科技有限公司 | Febuxostat curative effect prediction system, febuxostat curative effect prediction equipment and febuxostat curative effect prediction medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Javan et al. | An intelligent warning model for early prediction of cardiac arrest in sepsis patients | |
| Halasz et al. | A machine learning approach for mortality prediction in COVID-19 pneumonia: development and evaluation of the Piacenza score | |
| Ding et al. | Mortality prediction for ICU patients combining just-in-time learning and extreme learning machine | |
| CN115762764A (en) | HIV negative cryptococcus meningitis treatment outcome prediction model and construction method thereof | |
| Qezelbash-Chamak et al. | A survey of machine learning in kidney disease diagnosis | |
| CN116682557A (en) | Chronic complications early risk early warning method based on small sample deep learning | |
| Mitra et al. | Sepsis prediction and vital signs ranking in intensive care unit patients | |
| Nagaraj et al. | Deep learning and risk score classification of mild cognitive impairment and Alzheimer’s Disease | |
| Reddy et al. | Diabetes correlated renal fault prediction through deep learning | |
| Moreno-Sánchez | Data-driven early diagnosis of chronic kidney disease: development and evaluation of an explainable AI model | |
| Yang et al. | Machine learning-assisted ensemble analysis for the prediction of Acute Pancreatitis with Acute kidney Injury | |
| CN113643756A (en) | Protein interaction site prediction method based on deep learning | |
| CN112967803A (en) | Early mortality prediction method and system for emergency patients based on integrated model | |
| Alhmiedat et al. | The Investigation of Employing Supervised Machine Learning Models to Predict Type 2 Diabetes Among Adults. | |
| Suneetha et al. | Fine tuning bert based approach for cardiovascular disease diagnosis | |
| Sachdeva et al. | An Organized Method for Heart Failure Classification | |
| Gholamzadeh et al. | Comparison of different machine learning algorithms to classify patients suspected of having sepsis infection in the intensive care unit | |
| Haouassi et al. | An efficient classification rule generation for coronary artery disease diagnosis using a novel discrete equilibrium optimizer algorithm | |
| Ghias et al. | Using Machine Learning Algorithms to predict sepsis and its stages in ICU patients | |
| Shanmugarajeshwari et al. | Intelligent Decision Support for Identifying Chronic Kidney Disease Stages: Machine Learning Algorithms | |
| Ajay et al. | An Aquila-optimized SVM classifier for Diabetes prediction | |
| CN116403714B (en) | Cerebral apoplexy END risk prediction model building method and device, END risk prediction system, electronic equipment and medium | |
| Saroja et al. | Hybrid missing data imputation and novel weight convolution neural network classifier for chronic kidney disease diagnosis | |
| Dash et al. | Enhancing Disease Diagnosis: Statistical Analysis of Haematological Parameters in Sickle Cell Patients, Integrating Predictive Analytics | |
| Komalavalli et al. | An Effective Heart Disease Prediction Using Machine Learning |
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 |