47 results about "Survival data" patented technology

The subject disclosure presents systems and computer-implemented methods for assessing a risk of cancer recurrence in a patient based on a holistic integration of large amounts of prognostic information for said patient into a single comparative prognostic dataset. A risk classification system may be trained using the large amounts of information from a cohort of training slides from several patients, along with survival data for said patients. For example, a machine-learning-based binary classifier in the risk classification system may be trained using a set of granular image features computed from a plurality of slides corresponding to several cancer patients whose survival information is known and input into the system. The trained classifier may be used to classify image features from one or more test patients into a low-risk or high-risk group.

The subject disclosure presents systems and computer-implemented methods for providing reliable risk stratification for early-stage cancer patients by predicting a recurrence risk of the patient and to categorize the patient into a high or low risk group. A series of slides depicting serial sections of cancerous tissue are automatically analyzed by a digital pathology system, a score for the sections is calculated, and a Cox proportional hazards regression model is used to stratify the patient into a low or high risk group. The Cox proportional hazards regression model may be used to determine a whole-slide scoring algorithm based on training data comprising survival data for a plurality of patients and their respective tissue sections. The coefficients may differ based on different types of image analysis operations applied to either whole-tumor regions or specified regions within a slide.

The invention relates to an intelligent aeronautical data recording instrument, which is characterized in that a main control MCU adopts a dynamic algorithm processing system to analyze and compress the aeronautical flight data which is collected by a flight data collection system and adopts a streaming media dynamic processing system to analyze and filter voice information which is collected by a voice collection system, and the aeronautical flight data and the voice information after being processed are transmitted into an internal memory card and a crash survival data protection system. The present aeronautical data recording instrument is upgraded to three data processing system, and a data backup system is added, so the possibility for searching the aeronautical data recording instrument after the airplane is crashed is further improved, and the service life of the memory hardware is prolonged, the working efficiency is improved; different feasible and reliable technical reports and market analysis reports are supplied for the maintenance of the airplane and the development of the enterprise by comprehensively managing the flight data through a local ground data center; and chip structure is adopted, so the size is greatly reduced, and the memory space is multiplied.

The invention discloses a lung adenocarcinoma individuation prognosis evaluation method based on a polygene expression characteristic spectrum. The method comprises the following steps that a lung adenocarcinoma prognosis risk gene list and weight are acquired; a prognosis evaluation model is established by using a tumor tissue transcriptome and survival data of a patient with lung adenocarcinoma;a risk score of the patient is calculated according to a gene expression spectrum of the tumor tissue of the patient with lung adenocarcinoma; the annual survival probability of the patient is calculated according to the risk score of the patient. According to the method, the annual survival probability of the patient with lung adenocarcinoma and the actual annual survival probability are highlyidentical, the linear correlation R2 is equal to 0.994, and the P value is equal to 2.86E-43. It is proved that the method has high prediction accuracy, and the state is highly identical to the actualsurvival state. Meanwhile, for each patient with lung adenocarcinoma, the method can provide a specific survival probability curve of the patient.

The invention discloses an esophageal squamous carcinoma radical postoperative patient prognosis prediction model construction method and device, and the method comprises the steps: obtaining clinical diagnosis and treatment data and follow-up visit survival data, carrying out multi-factor Cox regression analysis on patient characteristic variables, tumor pathology characteristic variables, treatment condition variables and test index variables according to follow-up visit survival data, carrying out variable screening by utilizing a step-by-step back algorithm and an Akaike information criterion, and carrying out variable screening on the screened candidate variables again to obtain modeling variables; and performing multi-factor Cox regression analysis on modeling variables and interaction items of every two modeling variables to construct a prognosis prediction model of a patient after the esophageal squamous carcinoma radical operation, wherein the prediction variables comprise age, gender, tumor primary position, T stage, lymph node detection number, tumor size, preoperative hemoglobin level and N stage treatment mode interaction items. According to the method, the prediction accuracy can be improved, the optimal benefit group of different treatment schemes is defined, and the prognosis evaluation precision of the esophageal squamous cell carcinoma is realized.

The invention discloses a time sequence deep survival analysis system with active learning. The system comprises a data acquisition module, an active learning module and a time sequence deep survivalanalysis module. The data acquisition module is used for acquiring survival data of a to-be-analyzed object; the active learning module selects part of right deletion data to mark survival time in combination with an active learning method; and the time sequence deep survival analysis module constructs a time sequence deep survival analysis neural network model, and takes the undeleted data and the right deleted data as model inputs to obtain a survival time prediction result of the to-be-analyzed object. According to the method, the right deletion data and the time sequence characteristics inthe survival data can be fully utilized. Compared with a traditional survival analysis model, the system has the advantages that the problem that high-dimensional data is difficult to process is solved, and the problem that the model is poor in performance under the condition that only a small amount of data is not deleted in survival analysis is solved; meanwhile, extraction and utilization of data time dimension features are increased, the application range of the model is expanded, and the expression effect of the model is improved.

The invention discloses a cancer prognosis model construction method combining global weighted LBP (Local Binary Pattern) and texture analysis. The cancer prognosis model construction method comprisesthe following steps: acquiring original preoperative CT (Computed Tomography) image data of a cancer patient, marks of focus parts and survival data; calculating a three-dimensional global weighted LBP for the original CT data, and reconstructing new image data, namely global weighted LBP data; for the obtained global weighted LBP data, using GLSZM texture analysis to extract texture features. Texture analysis characteristics extracted by the method provided by the invention highlight tumor characteristics of a patient, and a prognosis model construction result is more effective; GLSZM is used for representing texture features, the effects in the aspects of texture consistency, rotation invariance and aperiodicity are remarkable, and the method has better performance than a co-occurrencematrix and a travel matrix in the aspect of cell nucleus and CT image texture.

The invention provides a system and a method for predicting tumor mutation load of a patient with advanced lung cancer. The system comprises a data acquisition module for acquiring a first lung tomography image and a real tumor mutation load of an early lung cancer patient, and a second lung tomography image and clinical survival data of an advanced lung cancer patient; a first processing module used for labeling by adopting a preset labeling box to obtain a first lung labeling image and a second lung labeling image; adding each first lung labeling image and the real tumor mutation load into afirst data set; a model training module used for training according to the first data set to obtain a tumor mutation load prediction model; a second processing module used for inputting each second lung labeling image into the tumor mutation load prediction model to obtain a predicted tumor mutation load; and performing survival analysis according to the clinical survival data to obtain a survival analysis curve, and processing to obtain the similarity between the distribution curve for predicting the tumor mutation load and the survival analysis curve. The prediction accuracy is effectivelyimproved; and the cost is saved.

A method for determining the lifespan of an agent utilizing a rule engine and a set of canonical survival rules, in an execution environment comprising collecting a survival rule, asserting a survival data into a working memory and executing the rule engine with the set of survival rules and the working memory.

The invention discloses a prognosis evaluation system for patients with early-stage LUAD and application thereof. According to the invention, disease-free survival data of 632 patients with early-stage LUAD in three independent queues are integrated, and an individual IBRS model for patients with early-stage LUAD is established and verified. The three independent queues include TCGA, GSE31210 and68 cases of frozen tissues. The IBRS model is composed of the following nine genes: DYNC1I2, THOC1, ADAM10, SERPINB6, CCL20, WNT2B, SLC11A1, MAPT and PSEN1. The establishment of the IBRS model for patients with early-stage LUAD is not only beneficial for understanding of the complex interaction mechanism between immune molecules, but also brings great help to the recurrence prediction of the patients with early-stage LUAD and the optimization of a treatment scheme.

The subject disclosure presents systems and computer-implemented methods for providing reliable risk stratification for early-stage cancer patients by predicting a recurrence risk of the patient and to categorize the patient into a high or low risk group. A series of slides depicting serial sections of cancerous tissue are automatically analyzed by a digital pathology system, a score for the sections is calculated, and a Cox proportional hazards regression model is used to stratify the patient into a low or high risk group. The Cox proportional hazards regression model may be used to determine a whole-slide scoring algorithm based on training data comprising survival data for a plurality of patients and their respective tissue sections. The coefficients may differ based on different types of image analysis operations applied to either whole-tumor regions or specified regions within a slide.

The invention discloses a polygenic expression characteristic spectrum based glioblastoma individual prognosis evaluation method. The method comprises the steps of obtaining a prognosis risk gene listand a gene weight of a glioblastoma; building a prognosis evaluation model by utilizing a tumor tissue transcriptome and survival data of a glioblastoma patient; calculating a risk score of the patient based on a gene expression spectrum of tumor tissues of the glioblastoma patient; and calculating the annual survival probability of the patient based on the risk score of the patient. A fact thatannual survival probability of the glioblastoma patient obtained by the method is highly consistent with the actual annual survival rate (the linear correlation R2=0.963, and P value=1.38 E-13) provesthat the method has very high prediction accuracy and is highly consistent with the actual survival state. At the same time, for each tumor patient, the polygenic expression characteristic spectrum based glioblastoma individual prognosis evaluation method can work out the special survival probability curve of the patient.

The invention discloses a personalized prognosis evaluation method for gastric cancer on the basis of amultigene expression characteristic spectrum. The method comprises the following steps that: obtaining a gastric cancer prognosis risk gene list and gene weight; utilizing a gastric cancer patient tumor tissue transcriptome and survival data to construct a prognosis evaluation model; according tothe gene expression spectrum of a gastric cancer patient tumor tissue, calculating the risk score of the patient; and according to the risk score of the patient, calculating the annual survival probability of the patient. The annual survival probability, which is calculated with the method of the invention, of the gastric cancer patient has consistent height with a practical annual survival ratio(linear correlation R2 is equal to 0.980, and a P value is equal to 1.62E-17). The method is proved to have high prediction accuracy which is highly matched with the practical survival state. Meanwhile, for each tumor patient, the method disclosed by the invention can give the specific survival probability curve of the patient.

The invention provides a multi-output gradient lifting tree modeling method for survival risk analysis, which comprises the following steps: firstly, constructing an expression of survival data for establishing a survival prediction model of financial, insurance, medical, traffic or industrial target industries under a model algorithm framework of an optimal gradient lifting tree (XGBoost); then defining and calculating a loss function corresponding to the survival data; then, defining and calculating a first step degree and a second step degree corresponding to the loss function; and finally,inputting the calculated loss function value and the first-order gradient value and the second-order gradient value of the loss function into an XGBoos model algorithm framework at the same time, andperforming automatic training to generate a survival prediction model of the target industry. The modeling method provided by the invention can better represent the relationship between the model covariable and the risk prediction value. The prediction performance and the generalization capability of the model are improved. The prediction performance and the risk distinguishing degree are better,and the application scene is wide.

The invention discloses a colon cancer individuation prognosis evaluation method based on a multigene expression characteristic spectrum. The method comprises the following steps: obtaining a colon cancer prognosis risk gene list and gene weight; using the tumor tissue transcriptome and survival data of colon cancer patients to construct a prognosis evaluation model; calculating the risk scores ofthe patients according to the gene expression profile of tumor tissues of the colon cancer patients; and calculating the annual survival probability of the patients according to the risk scores of the patients. The annual survival probability of the colon cancer patients obtained by the method disclosed by the invention is highly coincident with actual annual survival probability (linear dependence R2 being equal to 0.988, and P value being 7.35E-39). It is proved that the method has extremely high prediction accuracy, and the result is highly coincident with actual survival condition. Moreover, for each tumor patient, the specific survival probability curve of the patient can be given out.

The invention discloses a multi-gene expression characteristic spectrum-based individualized prognosis assessment method for clear cell carcinoma of kidney. The method comprises the following steps of: obtaining a clear cell carcinoma of kidney prognosis risk gene list and a gene weight; constructing a prognosis assessment model by utilizing tumor tissue transcriptomes and survival data of a patient with clear cell carcinoma of kidney; calculating a risk score of the patient according to gene expression profile of tumor tissues of the patient with the clear cell carcinoma of kidney; and calculating an annual survival probability of the patient according to the risk score of the patient. According to the method, the annual survival probabilities of patients with the clear cell carcinoma ofkidney are highly consistent with the practical annual survival rates (linear dependence R2=0.999, P=8.21E-74), so that the method has high prediction correctness and is highly identical with practical survival states; and meanwhile, for each tumor patient, the method is capable of giving s peculiar survival probability curve of the patient.

The invention discloses an individual cervical cancer prognosis evaluation method based on a poly-gene expression characteristic spectrum. The method comprises the following steps: acquiring a cervical cancer prognosis risk gene table and a gene weight; establishing a prognosis evaluation model by using cervical cancer patient tumor tissue transcriptome and survival data; according to a gene expression spectrum of tumor tissue of a patient suffering from cervical cancer, calculating a risk score of the patient; according to the risk score of the patient, calculating a survival probability of the patient each year. By adopting the method disclosed by the invention, the survivals probability of the patient suffering from cervical cancer each year is highly accordant with an actual survival rate each year (linear relationship R2=0.988, a P value=5.04E-38). Tests show that the method has very high prediction accuracy, and is highly coincided with an actual survival state. Meanwhile, for each tumor patient, a unique survival probability of the patient can be drawn by using the method.

A system and method for retaining routes in a control plane learned by an inter-domain routing protocol in the event of a connectivity failure between routers. Routers are classified as either route reflectors or originators. A connectivity failure between two routers is determined. A determination is then made whether to propagate a withdrawal of learned routes based on whether the connectivity failure occurred between a route reflector and an originator, two originators, or two route reflectors. A withdrawal of learned routes is propagated to neighboring routers if the connectivity failure occurred between two originators, or between a route reflector and an originator that is inaccessible via an intra-domain routing protocol. No withdrawal of learned routes is propagated if the connectivity failure occurred between two route reflectors, or between a route reflector and an originator that is accessible via an intra-domain routing protocol.

The subject disclosure presents systems and computer-implemented methods for assessing a risk of cancer recurrence in a patient based on a holistic integration of large amounts of prognostic information for said patient into a single comparative prognostic dataset. A risk classification system may be trained using the large amounts of information from a cohort of training slides from several patients, along with survival data for said patients. For example, a machine-learning-based binary classifier in the risk classification system may be trained using a set of granular image features computed from a plurality of slides corresponding to several cancer patients whose survival information is known and input into the system. The trained classifier may be used to classify image features from one or more test patients into a low-risk or high-risk group.

Belonging to the field of tumor treatment, the invention in particular relates to proteins for breast cancer risk assessment and application thereof. The inventor of the invention detects the expression levels of ATG5 (autophagy-related gene 5) and FIP200 (family kinase-interacting protein 200) in samples to identify samples with low expression of ATG5 and FIP200, and also can further predict thesamples as the high risk and poor prognosis type. In addition, the expression data and survival data of 200 breast cancer patient samples confirm that characterization of the expression of ATG5 and FIP200 can effectively predict high risk and poor prognosis samples, therefore ATG5 and FIP200 have high industrial utilization value in preparation or screening of drugs or kits for diagnosis of breastcancer, or screening of drugs for treatment of breast cancer.

An innovative nonlinear hybrid dynamic model of survival data analysis implemented in a systematic and unified way that is independent on any particular form of survival distributions functions or data sets. The introduction of one or more intervention processes provides a measure of influence for new tools, procedures and approaches continuous-time states of a time-to-event dynamic process.

The invention discloses a data collection method and system, and relates to the field of computer network technology. Via dynamical monitoring, survival data processing centers and collection probes are dynamically searched to implement corresponding collection when a data processing center is disconnected from a collection probe, and the problem that the whole data collection task cannot be carried out correctly due to stop of the collection probe or data processing center for certain reason is solved. Via load balancing, multiple collection probes can be used to implement collection when multiple devices are collected in a batch manner, and compared with a single collection probe in the prior art, the efficiency can be improved. Compared with a single-collection-probe mechanism in the prior art, collection is prevented from stop caused by that the collection probe is unavailable due to the network and the like.

The invention discloses a breast cancer prediction method based on penalty COX regression, which comprises the following steps of: processing follow-up data into survival data for later use, taking all prediction factors after data preprocessing as input variables of a model, sampling through a bootstrap method to obtain T self-service sample sets, a penalty COX regression model is independently constructed on the basis of different self-service sample sets to serve as a base predictor of integrated learning, after the base predictors are constructed, a simple average method is used for combining the T base predictors, and finally an integrated penalty COX regression model is formed to serve as an integrated predictor for breast cancer incidence prediction. According to the breast cancer prediction method based on penalty COX regression, a unique structure of a Bagging integrated framework and a penalty regression model is adopted, and the relationship between different dimension factors and female breast cancer onset risks in China is favorably discussed, so that doctors are assisted to give suggestions for preventing breast cancer onset, the variance of an estimator can be reduced, and the prediction accuracy is improved. The instability of estimation of a single classifier is avoided, and the prediction performance is improved.

The invention discloses a triple-negative breast cancer prognosis prediction model construction method, which comprises the following steps: realizing collection of original gene expression data and corresponding clinical survival data of a triple-negative breast cancer sample, and obtaining a gene expression matrix after realizing standardization processing of the gene expression data; obtaining an independent cancer formation specific gene of the triple-negative breast cancer, and obtaining the expression quantity of the gene; screening out parameters for constructing a prognosis prediction model from the obtained independent cancer formation specific genes and obtaining corresponding regression coefficients, wherein the parameters are multiple gene types; and based on the screened genes, calculating a risk score according to the expression quantity and the corresponding regression coefficient of the genes to obtain a triple-negative breast cancer prognosis prediction model. The triple-negative breast cancer prognosis prediction model constructed by the invention realizes risk stratification of prognosis of triple-negative breast cancer patients, significantly separates high and low risk patients, further can predict clinical results of triple-negative breast cancer and guide individualized treatment, and has high clinical application value.

The invention discloses a data analysis method and device, and the method comprises the steps: respectively carrying out the fitting of testee survival data corresponding to a first test object and testee survival data corresponding to a second test object, and obtaining a test group survival function and a control group survival function; based on preset clinical test parameters, a test group survival function and a control group survival function, determining survival data of the testee during middle-stage analysis and final-stage analysis; counting the survival probability of the test group relative to the control group in the middle-stage analysis and the survival probability of the test group relative to the control group in the final-stage analysis by using a preset statistical method according to the survival data in the middle-stage analysis and the survival data in the final-stage analysis; and judging whether the survival probability in the middle-stage analysis and the survival probability in the last-stage analysis meet preset conditions or not. Therefore, by carrying out the simulated clinical test based on the real test data, the reliability of the application of the preset method for mid-term analysis under the condition of the preset clinical test parameters can be accurately evaluated.

The invention discloses a double-gene combination for identifying a gastric cancer high-risk immune overdrive subgroup. The double-gene combination is prepared by the following steps: acquiring a gastric cancer immune checkpoint related double-gene combination by utilizing a gastric cancer patient tumor tissue transcriptome and survival data in a public database; detecting expression of double genes in a queue of gastric cancer patients to be detected; identifying respective most significant survival grouping points of the double genes according to the expression of the double genes and the survival data of the queue of the gastric cancer patients to be detected; and according to respective most significant survival grouping points of the double genes, dividing high-risk immune over-driven subgroups of the queue of the gastric cancer patients to be detected. The patient of the high-risk immune over-driven subgroups, identified according to the invention not only has obvious poor prognosis, but also has relatively high pre-stored immunocompetence and immune checkpoint abundance. For the queue of the gastric cancer patients, a patient group suitable for immunotherapy can be identified.

The invention discloses a method for evaluating the individualized prognosis of glioblastoma based on a multi-gene expression characteristic profile, comprising the following steps: obtaining a list of prognostic risk genes and gene weights of glioblastoma; using the tumor of a patient with glioblastoma The tissue transcriptome and survival data were used to construct a prognostic evaluation model; the patient's risk score was calculated according to the gene expression profile of the tumor tissue of the glioblastoma patient; the patient's annual survival probability was calculated according to the patient's risk score. The annual survival probability of glioblastoma patients obtained by the method of the present invention is highly consistent with the actual annual survival rate (linear correlation R 2 = 0.963, P-value = 1.38E-13). It is confirmed that the method has high prediction accuracy, which is highly consistent with the actual survival status. At the same time, for each tumor patient, the present invention can provide the patient-specific survival probability curve.

The invention provides a multi-output gradient boosting tree modeling method for survival risk analysis. The method includes: firstly, under the model algorithm framework of the optimal gradient boosting tree (XGBoost), construct a , the expression of the survival data of the survival prediction model of the transportation or industrial target industry; then define and calculate the loss function corresponding to the survival data; then define and calculate the first-order gradient and second-order gradient corresponding to the loss function; finally calculate The obtained loss function value and the first-order gradient and second-order gradient value of the loss function are simultaneously input into the XGBoos model algorithm framework, and automatically trained to generate the survival prediction model of the target industry. The modeling method of the invention can better represent the relationship between model covariates and risk prediction values; improve the predictive performance and generalization ability of the model; have better predictive performance and risk discrimination; and be applicable to a wide range of scenarios.