17 results about "Reference intervals" patented technology

The invention provides a patient vital sign anomaly detection method based on an artificial intelligence technology, and relates to the technical field of data processing, and the method comprises the steps: collecting original sign data of a patient; calculating multi-dimensional characteristic parameters; establishing an individual baseline model, and determining a comprehensive reference interval in the model; the vital sign features monitored in real time are constructed into multi-dimensional feature vectors, the multi-dimensional feature vectors are input into the individual baseline model, and the deviation degree of the real-time multi-dimensional feature vectors in the comprehensive reference interval is calculated; performing a clustering analysis to identify an anomalous aggregation region; performing trend analysis, calculating change direction consistency and continuous change amplitude of the multi-dimensional feature vector, and generating a trend analysis result; calculating an accumulated change index in a continuous time window according to a trend analysis result to obtain a dynamic confidence score; when the dynamic confidence score continuously exceeds an adaptive threshold value, determining that a vital sign abnormal event exists; the autonomy and accuracy of the method for detecting the vital sign abnormity of the patient are improved.

The invention provides a control method and system for interaction between emergency simulation training and a real respirator, and the method comprises the steps: collecting control parameters of the respirator, and carrying out the preprocessing of the control parameters, and obtaining an equipment parameter vector; calculating a physiological response state based on the device parameter vector; the physiological response state comprises alveolar pressure and oxygenation estimation; based on the physiological response state, performing quantitative judgment on the disease stage of the patient to obtain a current disease stage, and executing a corresponding evolution instruction set; a medical specification operation reference table is constructed, a target interval is extracted from the current illness state stage and the evolution instruction set, the matching degree of the target interval and a reference interval is calculated, and the reference interval is extracted from the medical specification operation reference table; determining whether to continue evolution, enter relieving or mark training is completed based on the matching degree; the breathing gas of the patient is simulated and restored through the gas compressor and the gas mixing device and is conveyed to the real breathing machine.

The invention relates to the technical field of physiological signal analysis, in particular to a neuropediatric disease assessment system for real-time physiological signals, which comprises a resting feature extraction module, a mark generation module, an abnormality judgment module, an abnormal fragment division module and a disease fluctuation labeling module. According to the method, by collecting the physiological signals of the child patient in the specific period, extracting the amplitude peak value of the physiological signals in the resting state and constructing the dynamic amplitude feature reference, the fluctuation range of the individual physiological baseline of the child patient can be more accurately captured, and compared with simple statistical feature matching, the method has the advantage that the accuracy is higher. According to the method, the adaptive capacity to individual differences and physiological natural variability is improved, real-time changes of physiological signals are continuously monitored, whether the multi-channel signals continuously deviate from respective reference intervals or not is judged, continuous physiological abnormal events instead of transient accidental artifacts are effectively recognized, then the signal strength in the deviation time periods is deeply analyzed, and the accuracy of the physiological abnormal events is improved. And significant anomalies in intensity can be accurately identified.

The invention discloses a network equipment health degree evaluation method based on multi-modal fusion, and relates to the technical field of new-generation information. According to the scheme, a multi-modal acquisition sequence is constructed based on multi-source operation information, a time sequence incidence matrix is established according to acquisition identification parameters, and a target time difference set is generated; forming a dynamic reference interval based on the time difference set, and performing time interval remapping on each modal sequence; performing time weighted combination processing on the corrected modal data to obtain a feature mapping set with a time association weight; constructing a multi-modal joint feature representation matrix based on the feature mapping set and performing health degree judgment operation; based on the above steps, the method can achieve the time consistency constraint of the multi-modal features in an asynchronous collection and distributed operation environment, reduces the interference of time sequence drift on the health degree evaluation, and improves the stability and reliability of the operation state evaluation of the network equipment.

The present application relates to the technical field of medical data processing, in particular to a method for quickly evaluating thyroid stimulating hormone value, which comprises the following steps: S1, clinical data acquisition; S2, data quality control and feature construction; S3, double-task joint prediction; S4, age self-adaptive reference interval correction; and S5, output of explainable report. The present application compresses the TSH evaluation period from the traditional 2-4 hours to within 5 minutes, reduces the detection cost by about 80%, and greatly improves the accessibility of thyroid function abnormality screening.

This disclosure provides a method and electronic device for diagnosing bearing faults. The diagnostic method includes: acquiring historical vibration signals of the bearing in three channel directions to form a benchmark database; extracting several key influence features associated with bearing faults from each historical vibration signal in the benchmark database, and determining a reference interval corresponding to each key influence feature; acquiring actual vibration signals of the target bearing in the three channel directions; extracting several target features corresponding to the key influence features from the actual vibration signals; calculating the comprehensive deviation of the actual vibration signals based on the target features and the reference intervals; and determining the bearing fault feature vector and target fault characteristics in response to the comprehensive deviation exceeding a preset threshold. This disclosure ensures the accuracy of bearing diagnosis while meeting the real-time monitoring needs of unlabeled, multi-condition industrial applications, and can dynamically adapt to changes in operating conditions without supervision, thereby significantly improving the accuracy of bearing fault identification.

The invention discloses a medical report interpretation method and device based on multi-modal dynamic knowledge fusion, and the method comprises the steps: recognizing a key region of a target medical report, and extracting a target physiological value index and a target text description index; inputting the baseline feature data into a physiological index interval prediction model to obtain a physiological index reference interval; calculating the physiological value index deviation degree according to the target physiological value index and the physiological index reference interval, and determining the abnormal degree of the physiological value index; key descriptors are extracted, and disease entities, standard text description and disease severity are inquired in the multi-dimensional medical knowledge graph; calculating the semantic similarity between the target text description index and the standard text description, and determining the abnormal degree of the text description index in combination with the disease severity; and determining a comprehensive anomaly score according to the physiological value index anomaly degree and the text description index anomaly degree. The method improves the accuracy and comprehensiveness of medical report interpretation, and can be widely applied to the technical field of artificial intelligence.

The invention relates to an auxiliary diagnosis system and device for neurodegenerative diseases and a storage medium. Based on a single molecule immunoassay technology, the relevance between the ratio of a P-tau217 concentration value and an A beta 42 concentration value and the neurodegenerative disease is deeply analyzed, a reference interval is designed, and an efficient screening tool is further developed; and high-specificity, high-sensitivity and high-accuracy auxiliary screening of neurodegenerative diseases such as Alzheimer's disease, subcortical ischemic vascular dementia and frontotemporal dementia is realized.

The present application relates to the field of image analysis, and particularly relates to an artificial intelligence identification method for urodynamic examination images. The present application obtains urodynamic contrast images and monitoring point displacement information, calculates a velocity time domain curve based on the displacement information, and captures a steady state time period. The present application positions a key contrast frame through time sequence alignment, selects adjacent continuous frames to construct a reference frame group through a sliding window, identifies a reference frame group organization contour to generate a superimposed thermal map, determines a superimposed main contour through frequency threshold screening and contour closure processing, calculates the interval width perpendicular to the contour based on displacement extreme values, constructs a strip-shaped dynamic reference interval along the main contour, and performs local enhancement on the images in the tissue contour reference interval. The present application specifically considers dynamic interference caused by physiological movement, improves the key area development quality, improves the image processing efficiency, and helps doctors to improve the diagnosis speed and efficiency.

The invention is suitable for the technical field of artificial intelligence and document intelligent analysis, and provides an AI-based bid-invitation purchasing document cross difference analysis method and system, and the method comprises the steps: judging whether the current reverse retrieval triggering frequency of an auditor is located in a reference interval, and maintaining the current operation if the current reverse retrieval triggering frequency is located in the reference interval; if not, intervention is executed according to the reference interval, so that the reviewer keeps the reverse retrieval triggering frequency matched with the reference interval in the review of the target to-be-checked item. Compared with the prior art, the method disclosed by the invention not only reveals the internal association between the behavior parameters and the auditing performance, but also constructs an auditing mechanism capable of closed-loop learning and intelligent regulation and control, and provides a new technical path and theoretical basis for AI auxiliary document auditing.

The invention discloses a septicopyemia early warning method, device and system and a storage medium. The septicopyemia early warning method comprises the following steps: acquiring patient observation data of a target patient under continuous time steps; acquiring a hidden state vector of the target patient under the current time step and a corresponding individualized baseline reference interval; generating an anti-fact observation result in each anti-fact disease cause scene; calculating a disease cause attribution vector and an infection specific residual error of the target patient under the current time step; calculating a septicopyemia event window risk value of the target patient in a preset time window; and outputting risk prompt information of the target patient in the current time step. By adopting the technical scheme of the invention, the problem that the existing early warning method is difficult to distinguish the causes of diseases among patients with multiple basic diseases is solved.

The invention, which relates to the technical field of data analysis, discloses a risk early warning method for a motor production process and a related device, the method comprising: acquiring real-time operation parameters and real-time environmental parameters of production equipment in the motor production process to perform production efficiency influence analysis; based on the real-time operation parameters and the real-time microstructure data of the insulating layer of the motor, a behavior response boundary database is used for carrying out production precision influence analysis; determining abnormal information based on the real-time microstructure data of the insulating layer by using the microstructure feature reference interval; determining production quality risk information based on the abnormal information, the production precision influence data and the production efficiency influence data; and generating early warning information and target intervention measure information based on the production quality risk information, and monitoring execution effect information of the target intervention measure information to update an intervention strategy library. According to the invention, the problems of risk identification lagging and inaccurate early warning in the prior art are effectively solved, and accurate risk early warning and active intervention in the production process are realized.

The invention relates to a human body tumble early warning method based on a millimeter wave radar, and the method comprises the steps: collecting a human body echo signal, obtaining the three-dimensional point cloud data on a time sequence after processing, and generating a human body motion track; calculating multi-dimensional track characteristics of each human body motion track, wherein the track characteristics represent physiological function states related to the falling risk from different dimensions; counting the number of times that each track feature exceeds a set reference interval in a set time period, and normalizing to obtain the abnormal frequency of each track feature; determining a weight factor of each trajectory feature, and further calculating to obtain a weighted sum of abnormal frequencies of all trajectory features as a fall risk score; and performing graded early warning according to the falling risk score. According to the method, the limitation of binary judgment of the tumble event in the prior art is broken through, a quantitative scoring system fusing feature importance and abnormal behavior frequency is constructed, and conversion from passive response to active prevention is realized.

The application discloses a medical test report image text detection method based on SFE-DBnet, and the process is as follows: after obtaining the image of a medical test report and inputting the SFE-DBnet network in a computer, a prediction candidate box is obtained to mark the text in the test report; the SFE-DBnet network is constructed based on a DBnet network and comprises a backbone network of ResNext50 combined with a CBAM (CBAM) and an aggregation residual block to extract features, then multi-scale feature extraction and fusion are carried out through an FPEM_FFM module, and then horizontal direction spatial sequence feature extraction is carried out through a BLSTM network to obtain a feature map. The SFE-DBnet network can effectively detect and identify the text such as serial numbers, item names, English names, measured values, marks, units, reference intervals and test methods in the medical test report, and the detection capacity is improved.

The invention relates to a method for adjusting the bounds of a diagnostic interval for calculating the lateral inclination angle of a two-wheeled motor vehicle (1), implemented by an electronic control unit (11), comprising the steps of: - in a first diagnostic phase, calculating (E3) a diagnostic value on the basis of predefined measurements of the inclination sensor (111); - storing (E5) the diagnostic value in the memory zone (112); - in the second phase of updating the bounds of the diagnostic interval (E9), checking that the motor (10) of the vehicle (1) is stopped; - comparing (E12) the measured inclination angle with the reference inclination interval; - if the measured inclination angle is within the reference interval of inclination when stopped, modifying the bound of the at least one diagnostic interval corresponding to the sign of the diagnostic value calculated during the first phase.

The application provides a kind of infantile electrocardiosignal intelligent analysis method and system based on conductive fabric, it is related to electrocardiosignal processing technical field, including: through flexible conductive fabric, multi-channel electrocardio data is collected, space correlation matrix is constructed to select optimal reference channel to obtain complete electrocardio data.Waveform reconstruction and feature extraction are carried out on complete electrocardio data, and physiological state feature vector is constructed.By querying heart rate baseline reference interval, the feature vector is calibrated and corrected, the heart rhythm type is determined by combining the interval stability and autonomic nervous state, and the analysis result is generated.The application realizes the stable collection and intelligent analysis of infantile electrocardiosignal, improves the analysis accuracy and reliability.