CN113674820A - Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy through machine learning - Google Patents
Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy through machine learning Download PDFInfo
- Publication number
- CN113674820A CN113674820A CN202111010994.1A CN202111010994A CN113674820A CN 113674820 A CN113674820 A CN 113674820A CN 202111010994 A CN202111010994 A CN 202111010994A CN 113674820 A CN113674820 A CN 113674820A
- Authority
- CN
- China
- Prior art keywords
- data
- cartilage repair
- machine learning
- neural network
- treatment
- 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.)
- Granted
Links
Images
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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/20—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for electronic clinical trials or questionnaires
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- 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
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Abstract
The invention discloses a method for quantitatively predicting cartilage repair rate after a mesenchymal stem cell therapy by machine learning, which comprises the following steps: s1, establishing a database; s2, preprocessing the original data; s3, processing lost data; s4, developing a model; s5, generating MSC treatment guidelines for cartilage repair through machine learning. The invention can accurately predict the treatment result of the patient receiving the MSC cartilage repair treatment according to the condition of the patient and the treatment strategy, thereby providing important reference for a clinician to decide whether and how to carry out MSC treatment on the patient with cartilage damage. Therefore, the invention is very suitable for popularization and application.
Description
Technical Field
The invention relates to the technical field of regenerative medicine, in particular to a method for quantitatively predicting cartilage repair rate after a mesenchymal stem cell therapy by machine learning.
Background
Stem Cell (MSC) therapy is the most promising candidate in regenerative medicine, but inconsistency in the efficacy of MSC treatment has been reported in various clinical trials. One reason for this rationality is the lack of quantitative scientific guidelines for formulating personalized MSC treatment strategies. Since the determination of optimal treatment parameters depends on the individual condition of the patient, it is difficult to quantitatively derive a treatment strategy for optimal treatment outcome of the patient by means of only routine biomedical research.
In recent years, in various biomedical fields, the use of machine learning techniques has made a great scientific progress, but attempts for stem cell therapy are few, mainly because existing machine learning models, especially neural network models, require a large amount of data support, such as the tenserflow of google requires at least 1000 rows or more of data set to run. Whereas clinical trial data is typically small (several to tens of data points), it cannot be processed if the data set is smaller than the number of variables. Furthermore, because of experimental design and data collection issues, entries with incomplete input information (i.e., missing data) may be included, whereas existing machine learning models typically require a complete data set. Therefore, the optimal treatment strategy of the MSC is obtained by predicting the cartilage repair treatment efficacy of the MSC by using the existing machine learning technology, and the requirements on data, time and cost are very high and almost impossible to realize.
Therefore, there is a need to develop a new machine learning model to predict the efficacy of the cartilage repair treatment of MSCs, so as to achieve the optimal treatment strategy of MSCs for different patients with personalized quantitative determination.
Disclosure of Invention
The invention aims to provide a method for quantitatively predicting the cartilage repair rate after a mesenchymal stem cell therapy by machine learning, and aims to obtain the optimal rehabilitation effect expected by a specific patient and a treatment strategy for achieving the target by prediction.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the method for quantitatively predicting the cartilage repair rate after the mesenchymal stem cell therapy by machine learning comprises the following steps:
s1, database establishment: collecting published clinical test reports about MSC cartilage repair therapies, obtaining original data, and classifying the original data into input and output of the data;
s2, preprocessing of the original data: normalizing the output scores using different scoring systems to [0,1], wherein 0 represents the most severe injury or pain and 1 represents a fully healthy tissue; meanwhile, according to the existing common sense basis, some initial values are distributed to missing data in a database to serve as preliminary predicted missing values; for example, when the cartilage of a mouse is artificially damaged using a drill, the depth of damage to the cartilage is calculated as 100% total loss when the depth of damage exceeds the cartilage depth;
s3, missing data processing: the missing value is set to the average of the values present in the dataset, then the following equations are recursively applied until convergence, and finally the result of convergence is returned:
x(n+1)＝γxn+(1-γ)f(xn)
wherein x represents a missing value and n represents an iteration step; f (x)n) Representing a prediction about x obtained from a neural network; gamma denotes a softening parameter, and gamma. epsilon. [0,1]](ii) a The function f remains fixed in each iteration of the loop;
s4, model development:
(1) capturing the functional relation among all variables through an artificial neural network framework;
(2) all x in the database are as close as possible to the function f of the fixed-point equation f (x) x;
(3) hiding nodes to which different activation function constraints are applied;
(4) considering state variables, processing variables and processing results as input and output of the neural network, and iteratively improving the estimation of all missing values by adopting an expectation-maximization algorithm;
(5) establishing a neural network model according to the steps (1) to (4);
s5, generating MSC treatment guidelines for cartilage repair by machine learning:
(1) randomly extracting and replacing a group of fixed data sets to enable the group of data to have a more flexible structure and generate more models;
(2) the generated model is used for training a neural network, and meanwhile, the model is recorded and the predicted performance is evaluated;
(3) obtaining a predicted overall performance based on the performance of all the different models, wherein a prediction with greater uncertainty will guide how to further collect additional data to supplement the database (e.g., stem cells made using cord blood exhibit greater uncertainty in therapeutic efficacy than stem cells made using bone marrow under otherwise identical conditions, and more experimental data will be needed in the future to test the effectiveness of cord blood stem cells); less uncertain predictions will be used to generate a series of quantitative guidelines for personalized MSC treatment for cartilage repair.
Specifically, in step S1, the input of data includes condition variables and treatment variables; the data counts included treatment results.
Preferably, in step S3, γ is 0.5.
Further, before proceeding to step S5, the accuracy of the model is evaluated by identifying key descriptors.
Still further, in step S5, the neural network is iteratively trained by changing the neural network model weight matrix.
Compared with the prior art, the invention has the following beneficial effects:
(1) in order to focus on the most reliable prediction of treatment outcome and to use a relatively small database of clinical trials for sufficient model training, two unique approaches are provided in the present approach: processing the missing data and calculating the uncertainty of the prediction so that the treatment outcome of a patient receiving MSC cartilage repair treatment can be accurately predicted based on the patient's condition and treatment strategy. This also optimizes the stem cell therapy in conjunction with machine learning techniques well, providing an important reference for the clinician to decide whether and how to treat MSC in patients with cartilage damage.
(2) When missing data processing is carried out, the invention finally returns a convergence result instead of f (x)n) On one hand, the convergence result simultaneously contains data internal information and a data external machine learning feedback result, which is equivalent to jointly predicting by using the correlation inside the data and the result of external neural network learning; on the other hand, γ>0 prevents the predicted fluctuation and divergence (γ is preferably 0.5). Therefore, the invention can utilize all information in the database, and provides guarantee for obtaining more reliable models and improving the prediction quality.
(3) In the invention, after the neural network model is established, the weight matrix is changed to iteratively train the neural network, which is beneficial to minimizing a cost function (also called a loss function) and reducing the final prediction cost.
(4) In the prior art, a common neural network can strictly distinguish a variable as an input or an input, and only one of the variables can be selected, which is not beneficial to further supplementing lost data. When the model is developed, the state variables, the processing variables and the processing results are regarded as the input and the output of the neural network, the expectation maximization algorithm is adopted to improve the estimation of all the missing values in an iterative manner, so that the most important and accurate correlation is found as far as possible, the accuracy of the original (the state variables, the processing variables as the input and the results as the output) model is further improved, and the missing data is further filled in the process. For example, in a common model, state variables and process variables are divided into inputs and the process results become outputs (which do not appear in the input training set). The invention takes the processing result as input, for example, the processing result, the stem cell number as input and the cartilage damage size as output, and then trains a neural network, wherein the correlation and prediction can be used for complementing the lost data in the cartilage damage.
(5) The invention randomly extracts and replaces a group of fixed data sets, so that a group of data has a more flexible structure to generate more models, and the neural network is trained. This is also more statistically relevant and statistically defines the uncertainty of the predicted outcome, thus ensuring the reliability of the quantitative guidelines for personalized MSC treatment. For example, the present invention can generate guidelines for: within 3 months post-surgery, the 60-65 year old patient is expected to receive a dose of 2000 ten thousand bone marrow MSC implants (treatment variable input) with 20-30% of the cartilage area compromised (condition variable input) with a maximum repair rate of 80% (output).
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention.
Figure 2 is a schematic diagram of a patient with possible injuries in accordance with an embodiment of the present invention.
FIG. 3 is a graph showing the number of stem cell injections recommended for a patient at the median injury level in an example of the present invention.
Detailed Description
The present invention is further illustrated by the following examples, which include, but are not limited to, the following examples.
Examples
The invention provides a method for quantitatively predicting cartilage repair rate after a mesenchymal stem cell therapy by machine learning, which aims to develop a machine learning model by training a database generated from published clinical test reports and then predict the curative effect of repairing cartilage by MSC transplantation by utilizing the correlation among attributes and the correlation among the attributes. The invention mainly comprises the procedures of database establishment, preprocessing of original data, lost data processing, model development, model accuracy evaluation and generation of MSC treatment guidelines for cartilage repair.
The above-described process will be described in detail below, as shown in FIG. 1.
Firstly, establishing a database
Published clinical trial reports on MSC cartilage repair therapies are collected (e.g., published reports from PubMed and international association for cartilage repair (ICRS) databases, including traumatic cartilage defects, osteoarthritis, etc.) and raw data is obtained and classified as input and output of data. The input of data includes condition variables such as age, sex, BMI, severity of cartilage damage (pre-treatment defect diameter and depth), etc., as well as treatment variables such as cell dose, source, delivery method, etc. The output of the data will be the treatment outcome, measured by different cartilage repair scores, pain level scores (e.g., ICRS score system, visual analog score of pain (VAS)), etc.
Secondly, preprocessing the original data
The output scores using different scoring systems (e.g., the international association for cartilage repair (ICRS) scoring system, the O' driscll score, the Pineda score, the Mankin score, the international association for osteoarthritis research (OARSI) scoring system, the international committee for knee literature (IKDC) score, the pain visual simulation score (VAS), the knee injury and osteoarthritis outcome score (KOOS), the west ampere major and the university of macystel osteoarthritis index (WOMAC) and the lysholm scoring system) were normalized to [0,1] for cross-comparison, where 0 represents the most severe injury or pain and 1 represents a completely healthy tissue. Meanwhile, according to the existing common sense basis, some initial values are assigned to missing data in the database as preliminary predicted missing values, and these initial values will be used as starting points for iterative machine learning to be performed later. For example, not all reports, particularly for Osteoarthritis (OA) trials, represent cartilage damage in terms of defect diameter of area and depth, which would result in "loss" of input data, and then preliminary guessing of these values is based on underlying assumptions (defect size scales linearly with OA levels).
Third, lost data processing
It is noted that due to experimental design or data collection problems, databases may contain entries with incomplete input information, and such missing data is more likely if the results of studies with acceptable differences in design and purpose are assembled to form a database. For example, osteochondral defect studies use the area of defects as a key data commonly used to assess the severity of injury in the databases of the present invention [14,17 ]. However, this information is not always provided in osteoarthritis studies [25,48] due to the difficulty of accurately measuring defect regions [25,48] with complex geometries, which can lead to a lack of data in the entries.
The design idea of the invention is that potential correlation may exist between different attributes, and missing information can be filled up by neural network extraction; typical neural networks require that each attribute be an input or output of the network, and that all inputs must be provided to compute a valid output; in contrast, in the present invention, the neural network takes as input the known treatment conditions and treatment outcome (if known) and then outputs a prediction of the unknown treatment conditions and treatment outcome; finally, the neural network is iteratively applied to loop predict unknown treatment conditions and treatment outcomes.
Thus, for any unknown attribute, the missing value is first set to the average of the values present in the dataset. By estimating all values of the neural network, then recursively applying the following equations until convergence, and finally returning the result of convergence:
x(n+1)＝γxn+(1-γ)f(xn)
wherein x represents a missing value and n represents an iteration step; f (x)n) Representing a prediction about x obtained from a neural network; gamma denotes a softening parameter, and gamma. epsilon. [0,1]]Typically set to 0.5; the function f remains fixed in each iteration of the loop.
Fourth, model development
After preliminary prediction and processing are carried out on lost data, according to the design thought, firstly, a functional relation among all variables is captured through an artificial neural network framework, then all x in a database are as close as possible to a fixed point equation f (x) which is a function f of x, and different activation function constraints are applied to hide nodes.
The state variables, process variables and process results are then considered as inputs and outputs to the neural network, and an expectation-maximization algorithm is employed to iteratively refine the estimates of all missing values.
And finally, establishing a neural network model according to the processing.
Fifthly, evaluating the accuracy of the model
The accuracy of the model is evaluated by identifying the key descriptors, and the method specifically comprises the following steps: principal Component Analysis (PCA) will be performed to identify linear combinations of key descriptors that can be used to confer model accuracy using a minimum number of input parameters.
Random cross-validation on each training data set would then be used to select the hyper-parameters, including the number of hidden nodes per output, the iteration period, and the number of training cycles (without reference to the corresponding test data set).
Accordingly, the performance of the neural network model will be based on the independent validation data and compared to other machine learning methods, such as Random Forest (RF), set matrix factorization (CMF), and multi-target Deep Neural Network (DNN) models, to further evaluate the accuracy of the neural network model.
Sixth, generating MSC therapeutic guidelines for cartilage repair
The present invention uses machine learning to calculate the uncertainty of the prediction while taking into account the uncertainty of the experiment and the uncertainty of the model itself.
The method comprises the following implementation steps:
(1) randomly extracting and replacing a group of fixed data sets to enable the group of data to have a more flexible structure and generate more models;
(2) the generated model is used for training a neural network, and meanwhile, the model is recorded and the predicted performance is evaluated; in this step, the activation function, the trained weight matrix and the robustness of the hyper-parameter are tested and benchmark tested, and the neural network is iteratively trained by changing the weight matrix thereof, so as to minimize the cost function;
(3) obtaining the predicted overall performance according to the performance of all different models, wherein the prediction with greater uncertainty will guide how to further collect additional data to supplement the database; less uncertain predictions will be used to generate a series of quantitative guidelines for personalized MSC treatment for cartilage repair.
A series of quantitative guidelines for personalized MSC treatment for cartilage repair include: the expected optimal therapeutic outcome for a particular patient, the treatment strategy to achieve the optimal outcome, the point in time after treatment for optimal healing and minimal pain development.
For example, the degree of joint damage of the patient is known (the damage depth percentage, see x-axis in fig. 2, and the damage area percentage, see y-axis in fig. 2). The model will optimize the optimal treatment protocol (including stem cell type, number of stem cell injections, etc.) in the background and give the expected results after treatment (see z-axis in fig. 2). The therapeutic effect is normalized to the [0,1] interval, with 1 indicating the best recovery state known to date and 0 indicating no effect. It can be seen that in areas of deep injury (e.g. injury depths greater than 60% and injury areas greater than 80%), stem cells do not provide effective recovery for the patient, in which case the patient should consider choosing an alternative to stem cells, saving time and money. Fig. 2 contains all possible damage scenarios.
For a single patient, the lesion status is known and determined, and the model is optimized based on the information of the particular patient. Fig. 3 is the recommended number of stem cell injections for a patient at the median lesion (fig. 3 corresponds to a point on the 3D contour plot), and it can be seen that the prognosis works best when two thousand seven million (27 millions) stem cells are injected. Recommended doses for general patients in a set of clinical control trials for different injected cell numbers (25mil,50mil,75mil,150mil) [ references: Gupta PK, Chullikana A, Renganam, Shetty N, Pandey V, Agrewal V, et al. efficiency and safety of adult human bone marrow-derivative, cut, poolated, allogenic genetic structural cells (Stempel): clinical and clinical trial in clinical trial of the knee joint. 18(1):301 ], in which the effect of injecting 25mil was found to exceed that of the other groups, but the information of the patients in the experimental group could not be obtained due to the ethical requirements of the experiment. The median damage value adopted has certain representativeness, and the trend of the model decreasing the efficacy in higher dose is also in line with the experimental result. It is worth noting that the model data does not contain any such experimental data and can be considered as a test set in machine learning.
Furthermore, by analyzing the severity of cartilage damage and determining damage thresholds, e.g., exceeding these thresholds, in the optimal cartilage repair protocol for a particular patient, the present invention predicts that the outcome will be a post-MSC treatment failure, and other treatment strategies should be considered to avoid unnecessary loss of time and cost.
It can be seen that the present invention is a pioneering study of MSC treatment efficacy based on machine learning patient and treatment specific information, and the success of this study will establish quantitative guidelines for clinicians to provide reference for future formulation of personalized cell therapy strategies. The invention breaks through the limitation of the prior art well, realizes innovation, accords with the trend of scientific and technological development, has obvious technical progress and has prominent substantive characteristics and remarkable progress compared with the prior art.
The above-mentioned embodiment is only one of the preferred embodiments of the present invention, and should not be used to limit the scope of the present invention, and all the technical problems solved by the present invention should be consistent with the present invention, if they are not substantially modified or retouched in the spirit and concept of the present invention.
Claims (6)
1. The method for quantitatively predicting the cartilage repair rate after the mesenchymal stem cell therapy by machine learning is characterized by comprising the following steps of:
s1, database establishment: collecting published clinical test reports about MSC cartilage repair therapies, obtaining original data, and classifying the original data into input and output of the data;
s2, preprocessing of the original data: normalizing the output scores using different scoring systems to [0,1], wherein 0 represents the most severe injury or pain and 1 represents a fully healthy tissue; meanwhile, according to the existing common sense basis, some initial values are distributed to missing data in a database to serve as preliminary predicted missing values;
s3, missing data processing: the missing value is set to the average of the values present in the dataset, then the following equations are recursively applied until convergence, and finally the result of convergence is returned:
x(n+1)＝γxn+(1-γ)f(xn)
wherein x represents a missing value and n represents an iteration step; f (x)n) Representing a prediction about x obtained from a neural network; gamma denotes a softening parameter, and gamma. epsilon. [0,1]](ii) a The function f remains fixed in each iteration of the loop;
s4, model development:
(1) capturing the functional relation among all variables through an artificial neural network framework;
(2) all x in the database are as close as possible to the function f of the fixed-point equation f (x) x;
(3) hiding nodes to which different activation function constraints are applied;
(4) considering state variables, processing variables and processing results as input and output of the neural network, and iteratively improving the estimation of all missing values by adopting an expectation-maximization algorithm;
(5) establishing a neural network model according to the steps (1) to (4);
s5, generating MSC treatment guidelines for cartilage repair by machine learning:
(1) randomly extracting and replacing a group of fixed data sets to enable the group of data to have a more flexible structure and generate more models;
(2) the generated model is used for training a neural network, and meanwhile, the model is recorded and the predicted performance is evaluated;
(3) obtaining the predicted overall performance according to the performance of all different models, wherein the prediction with greater uncertainty will guide how to further collect additional data to supplement the database; less uncertain predictions will be used to generate a series of quantitative guidelines for personalized MSC treatment for cartilage repair.
2. The method for machine learning quantitative prediction of cartilage repair rate after mesenchymal stem cell therapy according to claim 1, wherein in the step S1, the input of data comprises condition variables and treatment variables.
3. The method for machine learning quantitative prediction of cartilage repair rate after mesenchymal stem cell therapy according to claim 2, wherein in the step S1, the data is counted to include the treatment result.
4. The method for machine learning to quantitatively predict cartilage repair rate after mesenchymal stem cell therapy according to claim 3, wherein in step S3 γ is 0.5.
5. The method for machine learning to quantitatively predict cartilage repair rate after mesenchymal stem cell therapy according to claim 4, wherein before proceeding to step S5, the accuracy of the model is evaluated by identifying key descriptors.
6. The method for machine learning to quantitatively predict cartilage repair rate after mesenchymal stem cell therapy according to claim 5, wherein in the step S5, the neural network is iteratively trained by changing the weight matrix of the neural network model.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2020109105184 | 2020-09-02 | ||
| CN202010910518.4A CN112053751A (en) | 2020-09-02 | 2020-09-02 | Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy through machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113674820A true CN113674820A (en) | 2021-11-19 |
| CN113674820B CN113674820B (en) | 2023-10-17 |
Family
ID=73606775
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010910518.4A Pending CN112053751A (en) | 2020-09-02 | 2020-09-02 | Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy through machine learning |
| CN202111010994.1A Active CN113674820B (en) | 2020-09-02 | 2021-08-31 | Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy by machine learning |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010910518.4A Pending CN112053751A (en) | 2020-09-02 | 2020-09-02 | Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy through machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN112053751A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090010896A1 (en) * | 2007-07-05 | 2009-01-08 | Centeno Christopher J | Methods and compositions for optimized expansion and implantation of mesenchymal stem cells |
| US20090157083A1 (en) * | 2007-12-18 | 2009-06-18 | Ilwhan Park | System and method for manufacturing arthroplasty jigs |
| US20100189313A1 (en) * | 2007-04-17 | 2010-07-29 | Prokoski Francine J | System and method for using three dimensional infrared imaging to identify individuals |
| CN108064380A (en) * | 2014-10-24 | 2018-05-22 | 皇家飞利浦有限公司 | Use the prediction of the medical prognosis and therapeutic response of various kinds of cell signal transduction path activity |
| CN108766506A (en) * | 2018-08-31 | 2018-11-06 | 安徽国科生物科技有限公司 | A kind of multipotential stem cell diagnosis and therapy system |
-
2020
- 2020-09-02 CN CN202010910518.4A patent/CN112053751A/en active Pending
-
2021
- 2021-08-31 CN CN202111010994.1A patent/CN113674820B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100189313A1 (en) * | 2007-04-17 | 2010-07-29 | Prokoski Francine J | System and method for using three dimensional infrared imaging to identify individuals |
| US20090010896A1 (en) * | 2007-07-05 | 2009-01-08 | Centeno Christopher J | Methods and compositions for optimized expansion and implantation of mesenchymal stem cells |
| US20090157083A1 (en) * | 2007-12-18 | 2009-06-18 | Ilwhan Park | System and method for manufacturing arthroplasty jigs |
| CN108064380A (en) * | 2014-10-24 | 2018-05-22 | 皇家飞利浦有限公司 | Use the prediction of the medical prognosis and therapeutic response of various kinds of cell signal transduction path activity |
| CN108766506A (en) * | 2018-08-31 | 2018-11-06 | 安徽国科生物科技有限公司 | A kind of multipotential stem cell diagnosis and therapy system |
Non-Patent Citations (1)
| Title |
|---|
| 张季永;崔维顶;: "不同间充质干细胞来源的外泌体治疗骨关节炎的研究进展", 南京医科大学学报(自然科学版), no. 05 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112053751A (en) | 2020-12-08 |
| CN113674820B (en) | 2023-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Johnson et al. | Patient specific predictions in the intensive care unit using a Bayesian ensemble | |
| CN110841205B (en) | Accurate dose verification device for tumor patients | |
| WO2020220545A1 (en) | Long short-term memory model-based disease prediction method and apparatus, and computer device | |
| Ospina et al. | Random forests to predict rectal toxicity following prostate cancer radiation therapy | |
| US20160253473A1 (en) | Integrated virtual patient framework | |
| CN103116707A (en) | Heart disease intelligent diagnostic method based on case reasoning | |
| Craven et al. | Efficient argumentation for medical decision-making | |
| CN108647707A (en) | Probabilistic neural network creation method, method for diagnosing faults and device, storage medium | |
| Khovanova et al. | Neural networks for analysis of trabecular bone in osteoarthritis | |
| CN108962382A (en) | A kind of layering important feature selection method based on breast cancer clinic high dimensional data | |
| Zanna et al. | Bias reducing multitask learning on mental health prediction | |
| WO2009148803A2 (en) | Methods and systems for integrated health systems | |
| CN113674820B (en) | Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy by machine learning | |
| Halliwell et al. | Linguistic probabilities: theory and application | |
| CN106570325A (en) | Partial-least-squares-based abnormal detection method of mammary gland cell | |
| CN113707247B (en) | Method for quantitatively predicting cartilage repair rate after mesenchymal stem cell therapy based on machine learning of blockchain | |
| US20210174970A1 (en) | Integrated virtual patient framework | |
| CN113517044A (en) | Clinical data processing method and system for evaluating citicoline based on pharmacokinetics | |
| CN113469203A (en) | Method, electronic device and computer program product for evaluating operation results | |
| CN112635024A (en) | Automatic planning and designing system for radiotherapy and construction method thereof | |
| CN111243697A (en) | Method and system for judging target object data based on neural network | |
| Blobel et al. | A Modeling Framework for Decision Support in Periprosthetic Joint Infection Treatment | |
| Lin et al. | A Survey of Heart Disease Risk Computer Intelligent Prediction Model | |
| Sharifi et al. | A novel technique based on principal component analysis and multi-layer perceptron with genetic algorithm optimization for diagnosis of lung cancer | |
| Obeidat et al. | An MRI Based Algorithm for Detecting Multiple Sclerosis |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220207 Address after: 200082 room 308-02, building 1, No. 135, Yixian Road, Yangpu District, Shanghai Applicant after: Shanghai Shenyu Racing Technology Co.,Ltd. Address before: No.99, west section of Hupan Road, Xinglong Street, Tianfu New District, Chengdu, Sichuan 610000 China (Sichuan) pilot Free Trade Zone Applicant before: Chengdu Yushi Information Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |