35 results about "Prognostic model" patented technology

The invention discloses an immunity-related biomarker and an application of the biomarker to prognosis diagnosis of head and neck squamous cell carcinoma. The biomarker comprises four immune-related genes PVR, TNFRSF12A, IL-21R and SOCS1, and the biomarker and Path_N jointly form a head and neck squamous cell carcinoma prediction prognosis model. A group of new high-sensitivity and specific immunity-related molecular markers are developed for HNSCC prediction prognosis diagnosis to improve the outcome of a patient, which meets the current urgent demands. A model detection method is simple andeasy to operate, helps to better identify HNSCC patients, and can be used as a guidance basis for HNSCC immunotherapy decisions.

The invention belongs to the technical field of bioengineering and tumor markers, and relates to a prognosis model for the total survival rate of a bladder cancer patient, in particular to a prognosis model capable of predicting the total survival rate of the bladder cancer patient based on 12 DNA repair related genes. According to the invention, a prognosis model with 12 DRRGs is established, and bladder cancer patients are divided into high-risk and low-risk groups. The risk score of the bladder cancer patient in the training queue is significantly related to the OS (Plt, 0.001, HR = 6.3 [4.1, 9.8]). ROC curve analysis shows that AUC is 0.763 and 0.735 and 0.735 and 0.735 respectively in follow-up visits in one year, three years and five years. The prediction performance has been verified in the test set.

The invention relates to a hepatocellular carcinoma alternative splicing event prognosis model and a construction method and an application thereof. The prognosis alternative splicing variation of patients with hepatocellular carcinoma is systematically evaluated, a hepatocellular carcinoma alternative splicing prognosis model based on 26 alternative splicing events is constructed, the independentprognostic ability is verified further in combination with clinical pathological variables, and the key alternative splicing events related to the overall survival of the patients with hepatocellularcarcinoma are obtained. The research results show that the candidate events included in the final alternative splicing prognosis model are related to regulation of hepatocellular carcinoma, and the related key events can serve as potential biomarkers or therapeutic targets of hepatocellular carcinoma and are of clinical significance for deciphering the potential mechanism of hepatocellular carcinoma in tumor formation and pathogenesis.

The present invention describes a method of using a molecular prognostic signature, to predict overall survival in patients with metastatic clear cell renal cell carcinoma. The present invention also describes a method of selecting therapy and a process for patient risk stratification based on the molecular prognostic signature analysis.

The invention discloses a construction method and an application method of a lung cancer prognosis model, electronic equipment and a storage medium. The construction method comprises the following steps: acquiring peripheral blood leukocyte transcription spectrum expression data of a plurality of lung cancer patients and a plurality of reference people; screening candidate genes based on the transcriptional profile expression data; constructing a risk scoring model based on the candidate genes; wherein a lung cancer prognosis model comprises a risk scoring model; screening candidate genes based on the transcriptional profile expression data, including: performing similarity analysis on the transcriptional profile expression data to determine a plurality of lung cancer subgroups; respectively carrying out differential expression gene analysis on the plurality of lung cancer subgroups and a plurality of reference people, and determining respective differential expression genes of the plurality of lung cancer subgroups; comparing the expression states of the differentially expressed genes of the plurality of lung cancer subgroups, and selecting the lung cancer subgroup with more immunosuppressive genes in a high expression state from the plurality of lung cancer subgroups as a candidate lung cancer subgroup; and screening candidate genes from the differentially expressed genes of the candidate lung cancer subgroups.

The invention discloses a construction method of a lipid metabolism related endometrial cancer (UCEC) prognosis model. According to the construction method, differential genes are screened out layer by layer by using transcriptome data of UCEC in a public database TCGA to construct a prognosis model of lipid metabolism related UCEC, and it is verified that the prognosis model has good predictive capacity for endometrial cancer patients. The prognosis model can be used for guiding clinical endometrial cancer molecular typing and individualized treatment strategies, and products related to endometrial cancer diagnosis, treatment and prognosis can be further developed according to the core gene in the prognosis model.

The invention discloses a construction method of a metabolism-related endometrial cancer (UCEC) prognosis model. According to the construction method, differential genes are screened out layer by layer by using transcriptome data of UCEC in a public database TCGA to construct a prognosis model of metabolism-related UCEC, and it is verified that the prognosis model has good predictive capacity for endometrial cancer patients. The prognosis model can be used for guiding clinical endometrial cancer molecular typing and individualized treatment strategies, and products related to endometrial cancer diagnosis, treatment and prognosis can be further developed according to the core gene in the prognosis model.

The invention provides an application of a combined STAT signal channel related gene in a colorectal cancer prognosis model. The combined STAT signal channel related gene is CAV1, EPO, IL13, LEP and NEUROD1. The establishment method of the colorectal cancer prognosis model comprises the following steps: (1) collecting and sorting data; (2) screening differentially expressed STAT signal pathway related genes; (3) constructing a prognosis model of the STAT signal pathway related genes; and (4) constructing a column diagram. The prognosis model has the advantage of high accuracy, can clinically provide a new method for disease diagnosis and prognosis for colorectal cancer patients, reveals the prediction effect of the prognosis model on tumor microenvironment, and provides important information for the development of targeted therapy.

The invention discloses a new method for lung adenocarcinoma biomarker screening, prognosis model construction and biological verification. The method comprises the following steps: 1, carrying out data preprocessing on an original gene expression matrix; 2, carrying out biomarker screening on the data obtained after preprocessing; 3, constructing a prognosis model by using the lung adenocarcinoma prognosis marker subjected to the multi-step feature selection; 4, carrying out signal analysis on the selected marker after modeling. The invention has the beneficial effects that 45 gene markers are verified by using a linear support vector machine model, AUC and ACC which are poorly verified by five folds are used as model evaluation indexes, and the final result is that AUC is equal to 0.98, and ACC is equal to 0.92. The result is obviously superior to that of most traditional gene marker selection methods, the marker related to the lung cancer can be found out, the lifetime of a lung cancer patient can be accurately predicted, meanwhile, a channel related to lung adenocarcinoma occurrence can be explored, and the occurrence mechanism of lung adenocarcinoma can be explored.

The invention discloses a method for establishing a prognosis model for treating advanced liver cancer patients by HAIC, which comprises the following steps: (1) acquiring sample data, and screening the sample data; (2) carrying out single-factor analysis and multi-factor analysis by adopting a Cox proportional danger model, and selecting factors influencing prognosis from the sample data; (3) establishing a column graph prediction model for predicting prognosis according to the sample data and the factors influencing prognosis; and (4) verifying the column graph prediction model by adopting c-index and a correction curve. The model established by the method can evaluate the hepatic artery perfusion chemotherapy prognosis model of the local advanced liver cancer patient, and the model is displayed in the form of a column diagram, so that the model is intuitive and simple, and is convenient for clinical doctors and patients to operate and carry out individualized prediction on the patients.

The invention provides a prognostic scoring model of mixed cell type liver cancer and a construction method thereof, and the construction method comprises the following steps: collecting clinical pathological indexes related to survival of a mixed cell type liver cancer patient according to clinical cases, including albumin, gamma-glutamyltranspeptidase, carcino-embryonic antigen, carbohydrate antigen 19-9, tumor size, tumor quantity, microvessels and lymph nodes; analyzing and screening indexes related to prognosis of the mixed cell type liver cancer patient by adopting a single factor; further incorporating the selected indexes into multi-factor analysis, and obtaining independent prognosis factors related to patient survival; assigning each index and establishing a scoring table for prognosis prediction according to the risk ratio of multi-factor analysis; the AUC and C-index of the prognostic scoring model of the mixed cell type liver cancer reach 0.72 and are superior to those of other existing prognostic models, so that the prognostic scoring model of the mixed cell type liver cancer is high in prediction efficiency and can better guide clinical decisions and further improve prognosis of patients.

The safety and integrity of an embedded computer system is monitored using an independent safety monitoring module in communication with the main controller module via a serial connection to a safety monitoring module proxy in the main controller module. The main controller module is monitored through the use of alive-telegram exchanges and computational challenges. The safety monitoring module also receives temperature information and supply voltage information about the main controller module. The monitored information may be evaluated using a prognostic model constructed using a simulation of failure modes off line.

The invention relates to a method and equipment for predicting postoperative biochemical relapse of a prostate cancer patient. The method comprises the following steps: acquiring clinical pathological data of the prostate cancer patient, wherein the clinical pathology data at least comprise existence of IDC-P, age, NCCN standard prostate cancer risk classification and clinical T stage, and the clinical pathology data has certain correlation with the probability of biochemical recurrence of the prostate cancer patient. According to the clinical pathology data of the prostate cancer patient, based on a pre-trained prognosis model, a prediction probability of biochemical recurrence of the prostate cancer patient within a preset time is obtained. Prediction is performed through the prognosis model, compared with physicians, the prediction efficiency is higher, the prediction accuracy of the prognosis model can be continuously improved along with increase of sample data, and the prediction accuracy is higher compared with the physicians.

The invention provides a construction method and an application method of a liver cancer prognosis model and electronic equipment. The construction method comprises the following steps: acquiring transcriptional profile expression data of a plurality of liver cancer patients and a plurality of reference persons; screening candidate genes based on the transcriptional profile expression data; and based on the candidate genes, constructing a risk scoring model. The liver cancer prognosis model comprises a risk scoring model.

A system and method for predicting mild cognitive impairment (MCI) related diagnosis and prognosis utilizing deep learning. More specifically, the system and method produce predictions of MCI conversions to Alzheimer's/dementia and prognosis related thereof. Using available medical imaging and non-imaging data a diagnosis and prognosis model is a deep learned model trained using transfer learning. An MCI-DAP server may then receive a request from a clinician to process predictions related to a target patient's diagnosis or prognosis. The target patient's medical data is retrieved and used to create a model for the target patient. Then details of the target patient's model and the diagnosis and prognosis model are compared, a prediction is generated, and the prediction is returned to the clinician. As new medical data becomes available it is fed into the respective model to improve accuracy and update predictions.

The invention discloses a primary central nervous system lymphoma prognosis model. The prognosis model comprises related prognosis indexes, risk groups corresponding to different scores of the prognosis indexes and prognosis states corresponding to the risk groups. The prognosis indexes comprise at least two of the ratio of lactic dehydrogenase to lymphocyte, age and KPS score. According to the invention, the ratio of lactic dehydrogenase to lymphocyte, age and KPS score are integrated as prognosis evaluation indexes of primary central nervous system lymphoma for the first time, and a novel prognosis evaluation model for patients with primary central nervous system lymphoma is established.

The invention discloses a prognostic scoring method after intravascular treatment of aneurysm subarachnoid hemorrhage (aSAH), which comprises the following steps of: taking admission data of aSAH patients receiving an intravascular treatment method as a sample, taking one part of the admission data as a training set for training a prognostic model, and taking the other part of the admission data as a test set for comparing with a traditional method to explain the effectiveness and feasibility of the proposed model. A scoring system provided by the invention is designed on the basis of aSAH patient data of intravascular treatment; a data set has the advantages of large sample size and uniform patient treatment methods compared with the prior art; meanwhile, compared with a traditional scoring method, the algorithm can better perform prognosis scoring on an aSAH patient group; the prognostic scoring system involved in the invention is also suitable for aSAH mild symptom patient groups; and the prognostic scoring system provided by the invention not only can predict the prognosis condition of a patient in one year after the operation, but also can analyze the influence degree of each baseline factor on the prognosis.

The invention discloses a prognosis model for gram-negative bacillus blood flow infection of a tumor patient. The prognosis model comprises a plurality of optimal prognosis factors, evaluation scores corresponding to the optimal prognosis factors in different states and a survival probability corresponding to a total score obtained by adding the evaluation scores, wherein the prognosis factors mainly comprise platelet counting, lymphocyte counting, ICU (Intensive Care Unit), pulmonary infection and shock; wherein the optimal prognosis factors in different states include different counts of platelets before death of a gram-negative bacillus blood flow infected tumor patient, different counts of lymphocytes before death of the gram-negative bacillus blood flow infected tumor patient, different counts of the platelets before death of the gram-negative bacillus blood flow infected tumor patient, whether a gram-negative bacillus bloodstream infected tumor patient enters the ICU prior to death, whether a gram-negative bacillus bloodstream infected tumor patient has the shock prior to death, and whether a tumor patient has the pulmonary infection prior to being infected with gram-negative bacillus bloodstream infection. By means of the method, the death risk of the patient can be quickly, timely and accurately predicted.

The invention provides application of ferroptosis related genes in prognosis evaluation of diffuse large B-cell lymphoma, and belongs to the technical field of disease prognosis and molecular biology. The diffuse large B-cell lymphoma prognosis model is constructed based on the ferroptosis related gene, so that the prognosis of the diffuse large B-cell lymphoma is predicted and evaluated, tests prove that the prognosis model is remarkably related to the OS of a DLBCL patient, the OS of the high-risk-value DLBCL patient is obviously shorter than that of a low-risk-value patient, and the death possibility is higher; therefore, the model has a good prediction effect, can be used for prognosis prediction of diffuse large B-cell lymphoma and auxiliary clinical decision making, and has a good practical application value.

The invention provides an application of combined pyroptosis related genes to an esophageal adenocarcinoma prognosis model. The combined pyroptosis related genes are CASP1, GSDMB, IL1B, PYCARD and ZBP1. An establishment method of the esophageal adenocarcinoma prognosis model comprises the following steps: step 1, collecting and sorting data; step 2, screening differentially expressed pyroptosis related genes; step 3, reanalyzing the differentially expressed pyroptosis related genes; step 4, establishing a risk index evaluation model; and step 5, constructing an alignment diagram. The prognosis model provided by the invention has the advantage of high accuracy, and can provide a new method for disease diagnosis and prognosis for esophageal adenocarcinoma patients clinically.

A system and method for predicting mild cognitive impairment (MCI) related diagnosis and prognosis utilizing deep learning. More specifically, the system and method produce predictions of MCI conversions to Alzheimer's/dementia and prognosis related thereof. Using available medical imaging and non-imaging data a diagnosis and prognosis model is a deep learned model trained using transfer learning. An MCI-DAP server may then receive a request from a clinician to process predictions related to a target patient's diagnosis or prognosis. The target patient's medical data is retrieved and used to create a model for the target patient. Then details of the target patient's model and the diagnosis and prognosis model are compared, a prediction is generated, and the prediction is returned to the clinician. As new medical data becomes available it is fed into the respective model to improve accuracy and update predictions.

The invention relates to a liver cancer patient survival rate prognosis model. According to the invention, three genes are selected, and three gene-based tags (including BZW2, KIAA0907 and PSME3) are determined and constructed through an LASSO Cox regression model. The model provided by the invention has a potential clinical value in individualized treatment of liver cancer.