40 results about "Area under the curve" patented technology

The invention discloses a method for establishing a risk model for predicting intrahepatic cholestasis of pregnant women with maternal serum alpha fetoprotein. The method comprises the following steps: (1) dividing pregnant women into a case group and a control group according to the presence or absence of intrahepatic cholestasis in the gestation period, wherein the case group refers to a pregnant woman who is clinically diagnosed with intrahepatic cholestasis in the gestation period, and the control group refers to a pregnant woman who is randomly selected and has normal fetal development inthe same period; (2) detecting AFP levels of serum of two groups of pregnant women by adopting a time-resolved immunofluorescence method, constructing a risk calculation model by combining MOM valuesof the AFP with the prenatal age, and determining an optimal truncation value and an area AUC under the curve according to an ROC curve; (3) during detection, detecting the AFP level of the serum ofthe pregnant woman to be detected by adopting a time-resolved immunofluorescence method; and screening by using a risk calculation model constructed by combining the AFP with the prenatal age, and when the MOM value of the AFP of the serum of the pregnant woman to be detected is combined with the AUC predicted by the prenatal age modeling to exceed a set threshold value, judging that the fetus ofthe pregnant woman to be detected is intrahepatic cholestasis in the gestation period. The method has the beneficial effects that the MOM value of the AFP of the serum of the pregnant woman is combined with the pre-delivery age to screen the intrahepatic cholestasis in the gestation period, so that the method has high sensitivity and specificity and can become a new marker for predicting the intrahepatic cholestasis in the gestation period; the risk calculation model constructed by combining the MOM value of the AFP of the serum of the pregnant woman with the prenatal age has a great clinicalvalue in screening the intrahepatic cholestasis in the gestation period.

The present invention discloses a use of hsa_circ_0069922 as a marker in preparing a product for screening stroke-associated infection (SAI). The product for screening the stroke-related infection comprises a reagent or a drug for detecting an expression level of the hsa_circ_0069922 in peripheral blood mononuclear cells of a patient. The present invention develops the new molecular marker hsa_circ_0069922 for screening diagnosis of the SAI and a new reference for screening the SAI. The marker hsa_circ_0069922 is high in diagnostic value, a ROC curve analysis shows that patients with the SAI can be well distinguished and screened, an area under the curve is 0.832, and the marker hsa_circ_0069922 has good sensitivity and specificity and is expected to become an important index applied to evaluation of clinical diagnosis and treatment of the SAI.

The invention provides a neonatal individualized fluconazole administration amount calculation method and an administration scheme recommendation system, and belongs to the technical field of intelligent medical information processing. The method comprises the following steps: inputting the gestational age of a neonatal patient and the body weight of the patient into a neonatal fluconazole population pharmacokinetic model to obtain initial values V1 and CL1 of individual neonatal pharmacokinetic parameters; adopting a Bayesian maximum posterior probability method to predict the blood concentration of fluconazole at the t moment after administration to the patient according to V1 and CL1; fitting the blood concentration at each time point, calculating the area AUC0-24 under the curve at the time of 0-24 hours, and evaluating the anti-infection curative effect; if the curative effect evaluation is reasonable, adopting the administration amount and the administration interval as an administration scheme; if the curative effect evaluation is not good, optimizing the administration scheme. According to the method, the pharmacokinetic model of the fluconazole is established by taking the Chinese neonates as samples, the pharmacokinetic characteristics of the Chinese neonates are better met, a scientific basis can be provided for clinical fluconazole medication of the Chinese neonates, and the medication safety and the curative effect are improved.

The invention discloses a computer-aided system for predicting N2-stage lymph nodes of preoperative lung cancer patients based on a neural network. The prediction system comprises a first part: a data input part; the second part is a data preprocessing part; the third part is a model construction part; the fourth part is a model training part; and a fifth part: a prediction part: inputting data after preprocessing of CT features and clinical features of a to-be-predicted lung cancer patient, performing prediction by using the trained model, and outputting a prediction result: the to-be-predicted lung cancer patient is in an N2 stage or an N0 / N1 stage. The prediction system can accurately predict whether the lymph node of a T1N0M0 non-small cell lung cancer patient is in an N0 / N1 state or an N2 state, the area under curve (AUC) of a prediction result is up to 0.7847, the sensitivity is up to 89.80%, and the specificity is up to 54.50%. The prediction system plays a very important role in predicting the lifetime of a patient by a doctor, selecting an optimal treatment strategy and performing prognosis evaluation, and has a wide application prospect.

The invention discloses a system developed by utilizing a multiple myeloma diagnosis model based on logistic regression and application of the system. One technical scheme protected by the invention is an application of a system for identifying sex and detecting albumin content and hemoglobin content in preparation of a product for screening or auxiliary screening of multiple myeloma patients. According to the embodiment of the invention, by verifying the constructed diagnosis model, it is found that the area under the ROC curve of the diagnosis model is 0.995, the sensitivity is 0.956 and thespecificity is 0.981, which indicate that the diagnosis result of the diagnosis model has high accuracy and sensitivity. The system developed by the invention is simple and easily available in required indexes, high in sensitivity, capable of noninvasively, efficiently and accurately distinguishing multiple myeloma patients from healthy people, and suitable for popularization and application.