129results about How to "Avoid missed diagnosis" patented technology

The invention discloses a real-time auxiliary system for a controllable capsule endoscope operation on the basis of deep learning, and an operation method. The system comprises at least one client side and a service side, wherein at least one client side is connected with a capsule endoscope, is used for obtaining a capsule endoscope image collected by a current capsule endoscope and uploading thecapsule endoscope image to the service side through a network, and is also used for receiving and displaying an analysis result fed back from the display service side; the service side carries out capsule endoscope image processing according to the capsule endoscope image sent from the client side, judges positions and position characteristics corresponding to the capsule endoscope image in realtime, and feeds back the analysis result to the client side; and the service side comprises a sample database, a convolutional neural network model and a web service module. By use of the system, theimage collected by the controllable capsule endoscope is subjected to blind area monitoring and cancer focus identification and is displayed on the client side, an operation physician is assisted in checking the controllable capsule endoscope, detection accuracy and effectiveness is improved, and a missed diagnosis probability of occurrence is lowered.

The present invention provides a method, device, equipment for the segmentation of a lesion in a biological image and a storage medium. The method comprises a step of acquiring a target biological image, a step of performing coarse segmentation processing on the target biological image and obtaining a coarse segmentation mask after the rough segmentation processing, wherein the coarse segmentationmask includes information of candidate lesions in the target biological image, a step of identifying a non-real lesion from the candidate lesions, correcting the rough segmentation mask based on a recognition result such that the information of an identified non-real lesion is not included in the coarse segmentation mask and a target segmentation mask obtained after the correction is used as a lesion segmentation mask corresponding to the target biological image. According to the method, the device, the equipment and the storage medium, the lesion can be automatically positioned from the target biological image, the mode is labor-saving, the time consumption of the positioning of the lesion is reduced, misdiagnosis and missed diagnosis caused by the manual positioning of the lesion are avoided, the positioned lesion can also assist the doctor to carry out fast and accurate analysis, and the diagnostic efficiency and diagnostic accuracy of doctors are improved.

The invention discloses an endoscopic image lesion detection method based on based on fusion of global and local features. According to the method, supervised learning and multi-instance learning are combined, two levels of lesion detection mechanisms are established, the rate of missing detection is farthest reduced under the condition that the lesion detection accuracy and reducing labor intensity of doctors are ensured at the same time, and missed diagnosis or error diagnosis due to that a lesion image is mistakenly identified to a non-lesion image is avoided. According to the endoscopic image lesion detection method based on based on fusion of global and local features, only whether a trained image has lesion or not needs to be labeled and a specific lesion region does not need to be labeled when local features are extracted for training a classifier due to introduction of the multi-instance learning, thus the method is more convenient and easy in practical application than conventional method.

The invention provides a method for locating and identifying a lesion of a medical image, a device, an equipment and a storage medium. The method comprises the following steps of obtaining a target image, wherein the target image is a medical image to be localized and identified; preprocessing the target image to obtain a preprocessed image; inputting the pre-processed images into the pre-established lesion location recognition model, and obtaining the corresponding lesion location indication map of the target image and the lesion classification in the target image; obtaining the lesion location recognition model by training medical images with the lesion classification labeled; The lesion location recognition model based on that present application can automatically detect a lesion from amedical image, the location of the lesion is given, which not only saves manpower and reduces the time consumed in identifying and locating the lesion, but also avoids the misdiagnosis, missed diagnosis, locating and identifying lesion caused by manual locating and locating the lesion, which can also assist doctors to carry out rapid and accurate analysis, and improves the diagnosis efficiency and accuracy of doctors.

The invention relates to a medical diagnostic reagent, in particular to a quick detection reagent for the early diagnosis of acute myocardial infarction and a test strip. The invention provides a double-index united detection reagent containing a specific antibody resisting a human heart-type fatty acid binding protein H-FABP and a human cardiac troponin cTnI. The invention also provides a colloidal gold labeling immunological chromatographic test strip containing the specific antibody resisting the H-FABP and the cTnI, which is used for quickly detecting the acute myocardial infarction. The double-index united detection reagent can carry out early diagnosis on a patient having the acute myocardial infarction, can also prevent the missed diagnosis of a patient having long-time chronic myocardial infarction with mild symptoms and better solves the influence caused by a difference existing in the detection time. The colloidal gold labeling immunological chromatographic test strip provides a quick, convenient, cheap and practical detection tool for the early diagnosis of the acute myocardial infarction, is hopefully used for hospitals at all levels and can also be used for the self monitoring of the patient.

The invention relates to a fault diagnose method for variable load machine. The fault of the variable load machine is not easy to diagnose automatically, because the vibration value is compared with the alarm threshold set in advance in the traditional diagnose method, but the load is positively correlated to the vibration value; as a result, misinformation and missing report occurs if the traditional method is applied to the variable load machine. The method of the fault diagnose method is that the vibration signal of important parts of the machine and the power signal of the machine are collected simultaneously; then the vibration signal and the power signal are disposed with statistical regression analysis in order to elicit regression equation; the alarm threshold corresponding to the instant power of the machine is worked out according to the equation and the instant power of the machine; then the vibration and the alarm threshold are compared; alarm is given if the vibration surpasses the threshold; the alarm threshold varies with the power the machine. The fault diagnose method for variable load machine is mainly applied to the development and design of the diagnosis system of variable machines (such as the variable flow rate pump used in industry, variable power motor, and wind power generator).

The invention relates to the field of electric power system analysis and control, and particularly discloses an automatic evaluation platform and an automatic evaluation method for power grid equipment status. The automatic evaluation method includes: extracting monitoring data in real time on line; retrieving data which changes; finding out an equipment classification which power grid equipment which changes belongs to according to the data which changes; invoking an evaluation model for evaluating; and grading evaluation results, and proposing an overhaul proposal. The automatic evaluation platform and the automatic evaluation method for the power grid equipment status can comprehensively diagnose equipment insulation status according to various testing information and equipment operation status, provides decision support for fault locating and equipment overhauls, and can efficiently and accurately evaluate the power grid equipment status in real time.

The invention relates to a tube cavity wriggling shooting detector. The tube cavity wriggling shooting detector is characterized by comprising a front air bag, a middle air bag and a rear air bag, the middle air bag is arranged between the front air bag and the rear air bag, the front air bag and the rear air bag play a role in fixation with a tube cavity wall after being inflated, the middle air bag plays a telescopic wriggling walking role, the front air bag, the middle air bag and the rear air bag are correspondingly provided with a front air bag air nozzle, a middle air bag air nozzle and a rear air bag air nozzle respectively, a control circuit is arranged on a fixing support in the middle air bag, and the control circuit is provided with a three-cavity air pump and controller. The three-cavity air pump and controller is composed of a three-cavity air pump and a control circuit unit. The three-cavity air pump inflates and deflates the front air bag, the middle air bag and the rear air bag through the air nozzles connected with the air bags, the stepping state and the step of a wriggling shooting device are determined according to the pressure value of the middle air bag, meanwhile, and the grabbing strength of the air bag grabbing tube wall is determined according to the pressure value of the front air bag and the rear air bag to ensure walking tread stability.

The invention discloses a method for detecting catecholamine and metabolites by using liquid chromatography-tandem mass spectrometry technology. The method can be used for one-time detection of catecholamine and metabolites thereof with high sensitivity and specificity. The method can be used for detecting MN and NMN (indices which are firstly recommended in an association guide), E and NE (effects for auxiliary diagnosis) as well as DA and 3-MT, and the method can be used for carrying out diagnosis of PPGL of various types with very high sensitivity and specificity, wherein the sensitivity and the specificity respectively reach 100% and 97.2%.

The invention discloses a turbo generator set vibration fault diagnosis method based on forward reasoning. After a diagnosed turbo generator set starts, a parameter detection device is employed to carry out real-time detection on related work parameters in a start and operation process of the diagnosed turbo generator set and synchronously sends the detection information to a vibration fault diagnosis device for automatic vibration fault diagnosis, vibration faults of the diagnosed turbo generator set are diagnosed level by level by the vibration fault diagnosis device according to the detection result transmitted by the parameter detection device in combination with the active power value of the diagnosed turbo generator set, and the fault diagnosis process comprises steps of 1, shaft vibration swinging value diagnosis; 2, start process vibration fault diagnosis; 3, zero load operation vibration fault diagnosis; and 4, loaded operation vibration fault diagnosis. The method is advantaged in that the steps are simple, and the method is reasonable in design, is convenient to realize, has good use effects, can conveniently and rapidly accomplish the online steam turbine vibration fault diagnosis process and can further realize accurate and reliable diagnosis results.

The invention relates to a fault diagnosis method of a lithium ion battery sensor of an electric vehicle based on an observer, and belongs to the technical field of battery management. The method comprises the following steps: determining parameters of a lithium ion battery, and establishing an electrothermal coupling dynamic model of the lithium ion battery of the electric vehicle; performing anopen circuit voltage test and an HPPC experiment on the measured battery to obtain characteristic parameters of the battery under different ambient temperatures; establishing a relationship between battery OCV and SoC, and identifying the parameters in the electrothermal coupling dynamic model by using a recursive least square method with a forgetting factor to obtain a quantitative relationship among the battery parameters, the ambient temperatures and the battery SoC; and importing current, a voltage and a temperature sensor measured value into a fault diagnosis and separation algorithm of the lithium ion battery based on the observer, estimating a quantity of state through an extended Kalman filter algorithm to generate a residual, and performing residual evaluation by using a CUSUM test method, and finally achieving fault diagnosis and separation of the lithium ion battery sensor according to combined response conditions of different residuals.

The present invention discloses a female reproductive tract microecology evaluation and detection system. The system includes a full-automatic intelligent stainer, a vaginitis automatic detection workstation, a morphology full-automatic microscopic detection platform, a computer, an input device and an output device. The full-automatic intelligent stainer is used for automatic staining on a sample of vaginal secretions; the vaginitis automatic detection workstation is used for biochemical detection on preformed enzymes and metabolites in the vaginal secretions to obtain the functional assessment result data; the morphology full-automatic microscopic detection platform is connected to the computer and used for the detection of a smear formed after staining by the full- automatic intelligent stainer, and the morphology full-automatic microscopic detection platform detects and analyzes the visible components in the vaginal secretions to obtain morphological pattern data; the computer is used for controlling the full-automatic intelligent stainer, the vaginitis automatic detection workstation, the morphology full-automatic microscopic detection platform, the input device and the output device; the computer is equipped with application software and pathological knowledge base; and the input device is connected to the computer.

The embodiment of the invention provides an image quality assessment method and device and a computer readable storage medium. The method comprises the steps that an eye fundus image for a target object is acquired; the fundus image is input into a first preset model to obtain at least one optic disc area candidate position information in the fundus image output by the first preset model and a confidence coefficient corresponding to each optic disc area candidate position information; based on the candidate position information of the at least one optic disk area and the corresponding confidence coefficient, a target confidence coefficient of the optic disk area is determined; and an evaluation result of the fundus image is determined for the target object based on the target confidence coefficient of the optic disc area. The problems that the medical image analysis scene requirement cannot be met and the accuracy of an output result cannot be ensured are solved.

Disclosed are a magnetic resonance imaging system protecting method and device, and an automatic shutoff testing method. The protecting method includes a), providing expected safety parameters; b), after the magnetic resonance imaging system starts scanning corresponding to initial sequence information, collecting actual sequence information in the process of scanning; c), according to the actual sequence information, calculating actual safety parameters; d), judging whether the actual safety parameters do not exceed the expected safety parameters or not; e), if yes, continuing to execute scanning corresponding to an initial sequence; f), if not, stopping scanning. The device comprises an input module used for inputting the expected safety parameters, a collection module used for collecting the actual sequence information, a calculation module used for calculating the actual safety parameters according to the actual sequence information, a judgment module used for judging whether all of the actual safety parameters do not exceed the expected safety parameters or not, and the control module is used for controlling to continue to execute scanning if yes and to stop scanning if not. By the protecting method and the protecting device, safety of a scanning object in the actual process of scanning can be effectively guaranteed.

The invention discloses a bogie fault diagnosis method integrated with multiple sensors. The method comprises the following steps: synchronously acquiring vibration impact signals of the multiple sensors installed at fault-prone positions of a bogie; calculating a phase difference of the vibration impact signals of each sensor relative to the vibration impact signals of a certain sensor; according to the phase differences, positioning fault positions; according to the phase differences, performing phase shifting on the vibration impact signals of each sensor through a phase shift technology and carrying out signal integration; and calculating a maximum value of integrated signals and carrying out quantified diagnosis on faults. The fault diagnosis method provided by the invention can improve fault position locating and grade-alarm precision and sensor redundancy and can realize diagnosis of the faults of non-rotatable parts of the bogie.

The invention provides a method and system for processing a medical image. The method comprises the following steps that the image is read and displayed, the image is zoomed and rotated, negative film effect processing is conducted on the image, the image is cut, the grey level of the image is transformed, the contrast ratio of the image is increased, the image is displayed in the mode of a histogram, the histogram of the image is equalized, noise of the image is eliminated, the edges of the image are detected, the image is smoothed, and the image is sharpened. The system comprises a central processing unit, an image collecting module, an image processing module, a display module and a storage module, wherein the image collecting module, the image processing module, the display module and the storage module are respectively connected with the central processing unit. According to the method and system for processing the medical image, the medical image is read in and displayed, so that a series of processing is conducted on the medical image, the image processing technologies are fully used for effectively improving the image quality, diagnosis and treatment which are relative to the state of an illness and conducted by a doctor are facilitated, and the phenomena of missed diagnosis and misdiagnosis are avoided.

A process of preparing single-colon antibody resisting surface antigen of hepatitis B virus mutant from CGMCCHBSP2. The single-colon antibody can detect 14 types hepatitis B virus containing mutant and wild strain and replace HBsAb in fluorogence diagnostic kit. It can avoid mistake induced by normal agent and be used widely.

The invention discloses a micro calcification point automatic detection method based on an ultrasonic breast tumor image, and the method comprises the steps: segmenting an original image of the ultrasonic breast tumor image to obtain a breast tumor interest region; carrying out weak corrosion and strong corrosion distinguishing on the segmented breast tumor interest region; carrying out multi-scale superpixel segmentation on an ultrasonic breast tumor weak-corrosion interest region, and fusing the texture segmentation result of each scale to obtain a first suspected calcification point with a compact edge; performing single-scale superpixel segmentation on the original image of the ultrasonic breast tumor image, combining the original image of the ultrasonic breast tumor image after the single-scale superpixel segmentation with an ultrasonic breast tumor strong-corrosion interest region, and co-screening out a second suspected calcification point closer to the target based on the gray scale comparison characteristic and the gray scale distance characteristic; and obtaining the micro calcification point of the ultrasonic breast tumor with the accurate target and the compact edge. The method ensures that the detected tiny calcification point is accurate and reliable, is ingenious and novel to realize, and has a good application prospect.

The invention relates to a disaster relief triage processing device which is composed of an input device, a display device, an alarm device, a signal colleting device, a communication port and a signal processing device. The signal processing device comprises a storage device, a data collecting device, a first data processing device and a data forwarding device, wherein the data collecting device is used for converting wounded injury data collected by the signal colleting device and various instruments in a disaster relief site to be in a standard data format and transmitting the wounded injury data to the storage device for storing and is also used for receiving setting data; the data colleting device comprises an objective data colleting device and a basic data colleting device; the first data processing device is used for processing associated data collected in the disaster site to form suggestion data; the first data processing device comprises a first data type identifying device, a first data weighting device and a first data synthesizing device; and the data forwarding device is used for responding data requests in the outside of the disaster relief triage processing device and sending signals to the display device and the alarm device. The invention also provides a method for carrying out data processing by utilizing the disaster relief triage processing device.

The invention discloses a disposable anoscope which is used for various disease checks of the lower sections of the anal canal and the rectum and provides a space for operation treatment. The disposable anoscope comprises an inner cylinder and an outer cylinder, wherein the inner cylinder is a cylinder of which the side wall is provided with a longitudinal strip opening, and the cross section of the inner cylinder is in a C shape; the outer cylinder is also a cylinder of which the side wall is provided with a longitudinal strip opening, and the cross section of the outer cylinder is also in aC shape; and the inner cylinder is mutually sleeved with the outer cylinder to form a structure that the inner cylinder can rotate along the inner wall of the outer wall. When the inner cylinder rotates to lead the openings of the inner cylinder and the outer cylinder to superpose, a check window through which direct observe can be performed is formed, and the size of the check window can be adjusted and controlled according to the requirement, thereby the observation visual field range is wide and controllable. When the openings of the inner cylinder and the outer cylinder are completely closed, the anoscope can be conveniently inserted into the anus. The disposable anoscope solves the problems of inconvenient clinical check and difficult treatment operation of the currently used anoscope.

The invention provides an antibody composition and application thereof to detection of acute B lymphocytic leukemia. The antibody composition comprises four groups of antibodies; the first group of antibodies includes a CD38 antibody, an anti-CD10 antibody, an anti-CD34 antibody, an anti-CD19 antibody, an anti-CD20 antibody and an anti-CD45 antibody; the second group of antibodies includes an anti-cytoplasm CD79a antibody; the third group of antibodies includes an anti-CD81 antibody, an anti-CD34 antibody, an anti-CD10 antibody and an anti-CD45 antibody; and the fourth group of antibodies includes an anti-TdT antibody and an anti-cytoplasm CD79a antibody. The antibody composition disclosed by the invention can be applied to detecting acute B lymphocytic leukemia and / or minimal residual disease by flow cytometry, and can be used for effectively selecting B cells in a gating manner, including B cells of CD19- and B cells of CD19+, and the B cells cover the most complete immature B cell population and even B cell population, so that the missed diagnosis probability is reduced.

The invention relates to a micro-robot mechanism, concretely relating to a variable-diameter crawler-type robot mechanism used for colon diagnoses and examinations. Therefore, problems of a conventional micro-robot mechanism used for colon diagnoses and examinations are resolved. Such problems include low execution efficiency and the failure of fulfilling functions of varying diameters and lingering. The variable-diameter crawler-type robot mechanism used for colon diagnoses and examinations comprises a variable-diameter sub-mechanism and three walking sub-mechanisms. The variable-diameter sub-mechanism comprises a front baffle board, a back baffle board, a first direct current motor, a driving gear, a lead screw, a driven gear, a front nut, a back nut, three guide rods and three four-barlinkage mechanisms. Each walking sub-mechanism comprises a cabinet, a second direct current motor, a driving bevel wheel, a left bearing, a right bearing, a rotating shaft, a left synchronous pulley,a right synchronous pulley, a driven bevel wheel, a left resistance-decreasing roller, a right resistance-decreasing roller, a left crawler belt, a right crawler belt and an ear board. The variable-diameter crawler-type robot mechanism used for colon diagnoses and examinations is suitable for noninvasive diagnoses and examinations of colon diseases.

The invention provides an ultrasonic imaging method. The method comprises the steps that three-dimensional ultrasound image data of the face of a fetus are obtained; three-dimensional images, displayed on the same picture, of the left eye and ear and right ear and eye of the left side and the right side of the face are obtained respectively; the position information of the left and right eyes and the position information of the upper edges of the left and right auricles are obtained respectively; the angle between the connecting line of the left eye and the upper edge of the left auricle and the reference face and the angle between the connecting line of the right eye and the upper edge of the right auricle and the reference face are calculated respectively. The invention further provides an ultrasonic imaging device and ultrasonic equipment comprising the same. The low-set ear condition of the fetus can be rapidly diagnosed, and missed diagnosis is avoided.

The invention discloses an intracranial cerebral blood flow three-dimensional image acquisition device and method. The method includes the steps that a transcranial doppler probe is utilized to conduct single-beam or multi-beam ultrasound scanning on intracranial vessels of a detected object, ultrasound echo signals are received, and calculation is conducted according to the received ultrasound echo signals, so that blood flow information is obtained; meanwhile, a space tracking locator is utilized to obtain the relative position points and angle parameters of the transcranial doppler probe, and gathering space points in the detection process are calculated in combination with the relative distance between the space tracking locator and a probe wafer and depth adjustment parameters of the transcranial doppler probe; the gathering space points with blood flow information are integrated into a curve or a curved surface, and three-dimensional image data with the blood flow information are formed. By means of the intracranial cerebral blood flow three-dimensional image acquisition device and method, three-dimensional images of all acoustic window positions can be obtained, under the situation that workload of doctors is not added, comprehensive and visualized information of intracranial cerebral blood flow is obtained automatically, it is convenient to conduct diagnosis, diagnosis leakage and error diagnosis are prevented, information is more visualized and complete, and clinical big data can be accumulated and obtained.

The invention discloses a construction method and a construction system for a Kawasaki disease risk assessment model based on a neural network algorithm. The construction method comprises the steps: extracting valid samples which can be used for modeling and assessment from a sample data set; selecting ten features meeting field medical auxiliary diagnostic application from the feature set of thevalid samples; randomly dividing the incomplete data set of the valid samples into a training set and a validation set; fitting the training set by using the neural network method to construct the model and recording the optimal model parameters by using the ten-fold cross-validation method; and calculating the model classification threshold t by using the validation set according to the ROC curveso as to construct the Kawasaki disease risk assessment model. The corresponding Kawasaki disease risk assessment system is also constructed to be applied to assessment of the data to be assessed soas to obtain the KDx score. The misdiagnosis rate and the missed diagnosis rate of the Kawasaki disease can be reduced so that the patients can obtain effective prevention, intervention and treatmentin the early stage of the disease.

The invention discloses a method and a system for a B-mode ultrasound probe scanning without blind areas. The method comprises the following steps that video frame images of a part to be inspected ofa human body are acquired; a range to be checked of the part to be checked is determined and drawn on the video frame images; in response to the scanning of the B-mode ultrasound probe on the part ofthe human body to be examined, the coverage scanned by the B-mode ultrasound probe is displayed on the video frame images; whether the coverage completely covers the range to be inspected or not is judged; and if the coverage completely covers the range to be inspected, the completion of the inspection is prompted. According to the method and the system, a method for assisting an operator to carryout B-mode ultrasound scanning is provided, by tracking and recording a scanning track of the B-mode ultrasound probe in real time, the operator can see the scanned coverage more intuitively, and thescanning omission is effectively prevented.

The invention provides an all-functional respiratory disease intelligent consulting room system and a working method thereof. The system is characterized by comprising an intelligent consulting room subsystem, an intelligent consulting room management subsystem and a medical institution subsystem. The problems that the treatment process of the existing intelligent consulting room technology is limited and the equipment reusability is poor can be effectively solved. The diagnosis process is simplified through adopting the automatic technology, and medical resources are reasonably allocated. Anauxiliary diagnosis means is provided, so that the formulation efficiency of diagnosis and treatment schemes for chronic respiratory diseases is improved. The effect of qualified doctor resources richin experience and medical resources including large hospitals during the diagnosis and treatment of chronic respiratory diseases is fully utilized. As a result, patients can enjoy the diagnosis and treatment service of large hospitals in communities or at home. Users can select the most appropriate medical service in a plurality of large hospitals conveniently.

The invention discloses a joint detection kit for novel influenza virus and influenza virus A / B as well as preparation method and application of the joint detection kit. The joint detection kit comprises a detection test strip, the detection test strip comprises a sample pad, a combination pad, a detection pad and an absorption pad which are sequentially connected in an overlapped mode, and the detection pad is provided with a detection line and a quality control line which are parallel to each other; the detection line comprises a first detection line, a second detection line and a third detection line which are separated from one another, the first detection line is coated with a new coronavirus antibody, the second detection line is coated with an influenza A virus antibody, and the third detection line is coated with an influenza B virus antibody. The joint detection kit provided by the invention can simultaneously detect three items of the new coronavirus, the influenza A and the influenza B after one-time sampling treatment, is short in detection time and simple to operate, and avoids missed diagnosis or misdiagnosis in single item detection; besides, the repeated contact between medical staff and patients in multiple sample collection processes is reduced, and the risk of epidemic situation spreading is reduced.

The invention provides a reagent composition for screening and / or diagnosing clonal diseases by a one-step method and application of the reagent composition. The reagent composition comprises eight groups of antibodies and is a flow cytometry detection composition capable of being used for one-step screening and / or diagnosis of clonal diseases, a five-tube parallel scheme is used during application, the first group of antibodies and the second group of antibodies are respectively used for different tube samples, the third group of antibodies and the sixth group of antibodies are used for the same tube sample, the fourth group of antibodies and the seventh group of antibodies are used for the same tube sample, and the fifth group of antibodies and the eighth group of antibodies are used for the same tube sample. The reagent composition disclosed by the invention can be applied to flow cytometry one-step screening / diagnosis of various clonal diseases and other abnormalities, comprising acute leukemia, mature lymphocyte tumor, chronic myeloid tumor, night paroxysmal sleep hemoglobinuria, metastatic cancer, precancerous lesions, rare normal human polymorphisms, aplastic anemia, other immune cell subset abnormalities and the like.

The invention relates to a nuclear power device fault diagnosis method and system. The method comprises the steps of obtaining historical operation data and actual operation data of the nuclear powerdevice, and determining the state category of the actual operation data; performing phase space reconstruction on the actual operation data, and determining the reconstructed actual operation data; establishing a time convolution network model formed by stacking the small convolution kernels; constructing a time convolution network base classifier by using the reconstructed actual operation data and the state category; determining a training set and a test set of a secondary classifier according to the five time convolution network base classifiers; training a secondary classifier by using thesecondary classifier training set, testing the secondary classifier by using the secondary classifier test set, and determining a stack generalization ensemble learning model; and performing fault diagnosis on the nuclear power device according to the stack generalization ensemble learning model, and outputting a fault category. According to the invention, the accuracy of fault diagnosis is improved, and misdiagnosis and missed diagnosis are avoided.