38 results about "Postoperative risk" patented technology

The invention relates to the technical field of postoperative risk prediction, in particular to a breast cancer postoperative bleeding risk prediction method and system based on machine learning, and the method comprises the steps: extracting feature variables for breast cancer postoperative bleeding risk prediction from personal data and clinical data through feature importance analysis; utilizing a plurality of machine learning models to construct a plurality of prediction models based on the characteristic variables, and performing performance evaluation to select a prediction model with optimal overall performance; and carrying out local index optimization on the prediction model with the optimal global performance by utilizing reinforcement learning to obtain an optimal prediction model. According to the invention, a plurality of prediction models for breast cancer postoperative bleeding risk prediction are constructed by using a plurality of machine learning models, performance evaluation is carried out on the plurality of prediction models to screen out an optimal prediction model, and then index optimization with insufficient precision is carried out on the optimal prediction model one by one in combination with reinforcement learning. And obtaining a prediction model which is optimized in performance and helps to identify high-risk patients.

The invention relates to the technical field of intelligent medical treatment, in particular to a urinary surgery postoperative risk layering intelligent nursing management system which comprises the steps of preprocessing dynamic sign flow data of a patient to obtain regular sign time sequence data; and separating a core vital sign track, a wound exudation quantitative sequence and an autonomous perception discomfort record through modal analysis. The feature calculation module compares the vital sign trajectory with a standard recovery trajectory library to generate a sign bias spectrum, analyzes the exudation sequence to obtain a wound humidity distribution vector, and calculates a time attenuation factor of the pain event according to the discomfort record. And the space construction and layering module combines the features to construct a multi-dimensional risk base space, and identifies boundary coordinates of high-risk, medium-risk and low-risk risk clusters through clustering. And the strategy generation module divides risk levels and allocates corresponding monitoring and intervention measures according to the risk levels. According to the system, dynamic and multi-dimensional quantitative evaluation of the postoperative recovery state and objective automatic layering of risks are achieved.

The invention discloses a glioma radiotherapy postoperative risk assessment method based on a magnetic resonance image, particularly relates to the field of glioma radiotherapy patient health risk assessment, and is used for solving the problem that an existing assessment mode depends on manual interpretation and is difficult to predict bad clinical outcomes in advance. The method comprises the following steps: performing clinical data gridding reconstruction on a corticoid use cycle of a patient and tumor molecular typing, and combining morphological characteristics of an edema region in a magnetic resonance image to generate time-aligned clinical comprehensive characteristic vectors; mining a frequent association item set between the comprehensive feature vector and the pathological process to construct a mapping relation model; establishing a probability graph reasoning model of the bad outcome based on the pathological process vector sequence and the probability weight; and integrating the two types of models to form a causal reasoning network, and inputting a target patient feature vector to calculate an accumulated risk value of reaching a bad outcome. According to the method, an interpretable individual risk assessment result can be output, and a basis is provided for postoperative follow-up visit and intervention.

The invention relates to the technical field of clinical medicine, and discloses an intertrochanteric fracture postoperative progressive rehabilitation system, which comprises an intertrochanteric fracture comprehensive data acquisition module, a dynamic data monitoring module, a rehabilitation scheme generation module, a rehabilitation execution module, a rehabilitation mode adjustment module, a diet structure adjustment module and a rehabilitation interaction feedback module, the intertrochanteric fracture comprehensive data acquisition module is responsible for acquiring static and semi-dynamic data of a rehabilitation full cycle, the dynamic data monitoring module monitors physiological and motion change data of a patient in daily activities in real time, the rehabilitation scheme generation module generates a personalized stage rehabilitation scheme, and the rehabilitation execution module executes a full-process rehabilitation action. The rehabilitation mode adjusting module is used for adjusting a rehabilitation exercise mode, the diet structure adjusting module is used for optimizing a diet structure, and the rehabilitation interaction feedback module is used for collecting feedback so as to correct the scheme.

The invention provides an intelligent decision-making method and system for free flap riser anastomosis based on hemodynamics. According to the method, firstly, the blood flow direction, the flow speed and the blood vessel wall tension parameter of a target blood vessel are collected, and a dynamic correlation model of intravascular fluid mechanics and external tissue deformation is established; a three-dimensional blood flow pressure gradient map is generated based on the model, then, a system automatically screens an anastomosis path meeting a blood flow resistance threshold value and the endothelial cell survival rate requirement, and the optimal anastomosis pose is determined by analyzing the blood vessel wall elasticity matching degree and the turbulence generation probability. Then, guiding the accurate adjustment of the clamping direction of the blood vessel and the suture needle pitch in the operation, ensuring that the pressure of the lumen of the anastomosed blood vessel is kept balanced when the anastomosed blood vessel is axially stretched, meanwhile, effectively inhibiting local vortex, and remarkably improving the long-term smoothness of the anastomosis of the blood vessel. According to the method, mechanical precise matching and blood flow stable control of free flap vascular anastomosis are achieved, and the postoperative risk is remarkably reduced.

The invention discloses an intelligent evaluation method for screening multiple diseases of lumbar vertebra based on deep learning, and relates to the technical field of intelligent screening of lumbar vertebra diseases, and the method comprises the steps: collecting multi-source data, and obtaining lumbar vertebra CT, MRI, bone mineral density images, clinical diagnosis and treatment data and operation related parameters; preprocessing data, denoising images, segmenting key structures, and standardizing non-image data; a deep learning model is constructed, image features are extracted through CNN branches, non-image features are processed through Transform branches, and disease classification, severity assessment and risk prediction are achieved in combination with multi-task learning; integrating the multi-modal features through a feature fusion optimization algorithm, and outputting an evaluation result; doctor interaction correction and model iteration are supported, and the generalization ability of the system is improved. According to the method, multi-source data are integrated, the accuracy and efficiency of lumbar vertebra multi-disease assessment are improved, postoperative risk prediction is optimized, clinical diversified requirements are met, and powerful support is provided for precise diagnosis and treatment and rehabilitation.

The invention belongs to the technical field of medical information processing, and discloses a wound repair analysis method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining three-dimensional point cloud data, spectral data and patient physiological signals obtained through multiple times of scanning of a to-be-repaired wound; performing dynamic correction on the three-dimensional point cloud data based on the physiological signal of the patient and the spectral data to generate a standardized wound three-dimensional model; performing space-time dimension wound assessment based on the model and the dynamic data of the wound to obtain a quantitative assessment report; and finally, generating an individualized repair scheme through double-path driving of historical data and operator experience. According to the method, the scanning precision is improved through a dynamic correction algorithm fused with physiological signals, advanced prediction of postoperative risks is achieved through a space-time dimension evaluation model, the individual suitability of the scheme is improved through a double-path scheme generation mechanism, and the precision and success rate of plastic wound repair are remarkably improved.

The invention relates to the technical field of postoperative risk prediction, in particular to a cardiovascular intervention postoperative nursing risk prediction system based on physiological feedback. The method comprises the following steps: firstly, acquiring physiological data of historical patients and risk values of postoperative care, and grouping; the aggregation of the risk values of the patients is further analyzed, and the initial typical degree is obtained in combination with the risk values; further analyzing the reference weight of other groups for each physiological data of the target group, analyzing the correlation between the physiological data of other groups and the maximum initial typical degree in other groups, and obtaining the prediction importance of each physiological data of the target group; the final typical degree is further obtained; and finally, based on the final canonical degree fusion risk value of the patients in each group, obtaining a risk optimization value and constructing a prediction decision tree, thereby improving the representativeness and individualized prediction capability of the decision tree.

The invention belongs to the technical field of biological medicine, and discloses a portal cholangiocarcinoma postoperative risk prediction model based on a gamma-glutamyltransferase (GGT) dynamic change track and a construction method thereof. The method comprises the following steps: (1) collecting clinical characteristic data and survival characteristic data (including survival time and survival state) of a patient suffering from porta hepatis cholangiocarcinoma subjected to radical resection; 2) on the basis of GGT detection results at different time points in a perioperative period, dividing patients into different GGT dynamic change track types by using a latent category hybrid model; and 3) taking the clinical characteristic data and the GGT dynamic change track type as independent variables, taking survival data as response variables, screening prediction factors through LASSO regression, and constructing a multi-factor Cox regression model, namely a postoperative risk prediction model. According to the method, the dynamic evolution rule of the GGT is introduced, so that the limitation of single-time static detection is made up, the quantitative evaluation of the survival risk is realized, and a scientific basis is provided for individualized treatment and follow-up visit management.

The application belongs to the technical field of molecular biology, and particularly relates to metabolic markers for evaluating the prognosis of patients with hypertrophic cardiomyopathy and application thereof. The application determines that one or more of 3-hydroxyrilotene, asparagine-lysine, cyclohexylamine, arachidonic acid, aconitic acid, propionyl carnitine, ceramide, tryptophan, 3-hydroxysebacic acid, traumatic acid, phosphatidylcholine, alanine-proline, lysophosphatidylcholine, taurine and S-garlicin cysteine are related to the prognosis of hypertrophic cardiomyopathy, can evaluate the prognosis of patients with hypertrophic cardiomyopathy after undergoing muscle resection of interventricular septum, and shows accurate judgment on all-cause death or rehospitalization, cardiovascular events and postoperative new-onset arrhythmia. Finally, all the prognostic biomarkers are reproduced in an independent cohort, which is helpful for postoperative risk evaluation of patients with hypertrophic cardiomyopathy, has high specificity and high sensitivity.

The application relates to the technical field of medical information processing, and discloses a method and system for dynamically evaluating post-renal biopsy occult bleeding risk based on multi-source data fusion, wherein the method comprises the following steps: constructing a standardized uniform observation vector sequence; constructing a short-time instant risk model; constructing a mechanism to establish a long-time lag risk transfer model; generating a hierarchical treatment decision and a review scheduling plan; gain correcting the risk probability; and outputting a final occult bleeding risk grade. Compared with the scheme for postoperative bleeding risk in the prior art, which only performs static risk evaluation based on a single time section index, especially in the case that postoperative vital signs are temporarily stable but there is delayed hemoglobin decline, the technical problem that the continuous prediction of the risk transfer process of'short-time stability-long-time lag deterioration' cannot be realized is solved. Since the time-varying state transfer probability matrix is introduced in the application, and the risk transfer acceleration is determined, the sensitivity of postoperative risk identification is improved.

The invention discloses a heart valve postoperative death risk prediction method based on physiological index interaction, and relates to the technical field of medical engineering. The method specifically comprises the steps that S1, variable importance analysis is carried out to reduce interference of unimportant feature physiological indexes; s2, analyzing a survival curve for determining the influence of the key feature physiological indexes on the postoperative survival state of the patient; s3, constructing an intelligent discrimination model, and constructing the model by adopting a classification machine learning algorithm; and S4, intelligent discrimination and prediction: screening the postoperative death risk prediction model of the heart valve patient for predicting the postoperative death risk of the heart valve patient, and reducing the discrimination difficulty. According to the method, by analyzing the synergistic effect and antagonism among different characteristic physiological indexes, the heart valve postoperative death risk prediction model with high accuracy and good robustness is screened, and physiological data is provided for postoperative risk analysis and evaluation of heart valve patients.

The invention relates to the technical field of post-anesthesia complication monitoring, in particular to an intelligent pre-warning system for post-anesthesia complication of a surgical patient. The system obtains an intraoperative risk degree according to the size and change of intraoperative body index data of a patient in each first preset time period and past medical history scores and allergy history scores of the patient, and further judges an intraoperative anesthesia complication patient and an intraoperative normal patient in real time; according to the size of the post-operation body index data, the consciousness state score, the nervous system score, the wound recovery score and the waking state of the normal patient in each second preset time period and the post-operation risk degree of the normal patient in the operation, the post-operation risk degree is obtained, and then the post-operation anesthesia complication patient and the normal patient in the operation are judged in real time. According to the method, the intraoperative risk degree and the postoperative risk degree are accurately obtained in real time, so that the post-anesthesia complications of the patient are accurately detected in time, measures are taken in time, and the harm of the post-anesthesia complications is effectively reduced.

The utility model discloses an anastomat assembly with a blood coagulation function. The anastomat assembly comprises a cutting knife assembly, a nail box inner core, a nail bin assembly and a nail abutting seat assembly. Wherein the cutting knife assembly comprises a cutting knife head, a cutting knife middle piece and a cutting knife side piece, the cutting knife head is located at the far end of the cutting knife middle piece, and the cutting knife head, the cutting knife middle piece and the cutting knife side piece are fixed together; the anastomat assembly further comprises a thermocouple assembly and an electric connection assembly, and the thermocouple assembly is connected with the electric connection assembly. The thermocouple assembly is arranged on the cutting tool bit and can heat the cutting tool bit; when the cutting knife assembly moves, the electric connection assembly and the thermocouple assembly keep a circuit closed. The endoscopic anastomat assembly is matched with an electric endoscopic anastomat to achieve a blood coagulation function on cut tissue capillaries, the amount of bleeding in an operation is reduced, postoperative recovery is faster, and the operation risk and the postoperative risk are reduced; the number of times for medical staff to clean errhysis at an operation site during an operation can be reduced, the labor of the medical staff is reduced, the operation duration can be shortened, and the working intensity of the medical staff is reduced.

The application discloses a myopia surgery scheme generation method, a generation system and a storage medium, and comprises the following steps: collecting historical surgery data, wherein the historical surgery data comprises preoperative data, intraoperative data and postoperative data of a plurality of patients, the preoperative data comprises eye parameters, living habit data, previous medical history data and patient basic information; preprocessing the historical surgery data to obtain a training set; constructing a dynamic Bayesian network model based on the training set; collecting preoperative data of a patient to be diagnosed, inputting the dynamic Bayesian network model, and generating a myopia surgery scheme of the patient to be diagnosed based on a surgery scheme library. Through multi-modal data fusion to construct a dynamic Bayesian network, preoperative individualized scheme recommendation, intraoperative real-time parameter optimization and postoperative risk tracking are realized, and the surgery success rate and patient satisfaction are improved.

The application relates to the technical field of postoperative early warning, and particularly discloses a postoperative risk early warning method and system based on machine learning, which belongs to the technical field of postoperative early warning, and the method comprises the following steps: acquiring historical patient data of a plurality of patients who have undergone a to-be-early-warned operation, determining an adverse prognosis set of the to-be-early-warned operation and adverse data of each adverse prognosis in the adverse prognosis set; determining an influence factor vector of each adverse prognosis in the adverse prognosis set and an independent prognosis factor set; determining a plurality of preset non-independent factor sets of each adverse prognosis in the adverse prognosis set; determining a non-independent prognosis factor set of each adverse prognosis in the adverse prognosis set; and realizing postoperative risk early warning. The postoperative risk early warning method can comprehensively cover the postoperative risk influence dimension, accurately capture data correlation, realize intelligent transformation from data to risk prediction, provide more accurate and comprehensive postoperative risk early warning for the clinic, and improve the scientificity, accuracy and reliability of postoperative risk early warning.

The invention relates to the technical field of risk assessment, and discloses a lung transplantation patient postoperative risk auxiliary judgment method based on data analysis, and the method comprises the steps: S1, carrying out the model training according to collected data, obtaining a risk calibration model, and obtaining a real-time risk value according to the risk calibration model and current data; s2, according to a critical threshold value of the real-time risk value and the change characteristics, performing preliminary judgment on the postoperative risk, and when a preliminary judgment result is high risk, sending out first early warning information; and S3, comparing the graph change track of the numerical value of the real-time risk value with a preset risk change track, performing depth judgment according to a comparison result, and sending second early warning information when a depth judgment result is high risk. The postoperative risk is preliminarily judged according to the critical threshold value and the change characteristics of the real-time risk value, depth judgment is performed according to the graphic change track of the numerical value of the real-time risk value and the preset risk change track, and the risk of the patient can be continuously, timely and accurately judged.

The application discloses a kind of nerve electrophysiological monitoring equipment, the nerve electrophysiological monitoring equipment includes: urethra detection device, urethra detection device is provided with bladder pressure detection piece and first electromyography detection piece;Rectum detection device, rectum detection device is provided with rectal pressure detection piece and second electromyography detection piece;Monitoring module, monitoring module is electrically connected with bladder pressure detection piece, first electromyography detection piece, rectal pressure detection piece and second electromyography detection piece respectively.According to the nerve electrophysiological monitoring equipment of the application, the nerve electrophysiological monitoring equipment can simultaneously detect the intravesical pressure and intrarectal pressure, the electromyographic activity of urethral external sphincter and the electromyographic activity of anal external sphincter, without multiple monitoring instruments for detection, thereby improving the integration of nerve electrophysiological monitoring equipment and the safety of operation, reduce postoperative risk.

The utility model provides a hip and leg braking device which can be worn on the hip and the unilateral lower limb of a patient and comprises a hip fixator, a thigh fixator, a shank fixator and a supporting plate. The hip fixator is used for wrapping the hip of a patient, the thigh fixator is used for wrapping the thigh on one side of the patient, and the shank fixator is used for wrapping the shank on the side of the patient; the supporting plate penetrates through the hip fixator, the thigh fixator and the shank fixator, is located on the inner sides of the hip fixator, the thigh fixator and the shank fixator and is used for preventing the lower limbs of the patient from bending. According to the hip and leg braking device, the hip, the thighs and the shanks of a patient can be braked at the same time, so that the lower limbs of the patient are effectively prevented from bending, and the postoperative risk is reduced.

The application relates to the technical field of anesthesia post-complication monitoring, in particular to an intelligent early warning system for post-anesthesia complications of surgical patients. The system obtains the intraoperative risk degree according to the size and change of the intraoperative body index data of the patient in each first preset time period, the past medical history score and the allergy history score of the patient, and then judges the intraoperative anesthesia complication patients and the intraoperative normal patients in real time; the postoperative risk degree is obtained according to the size of the postoperative body index data of the intraoperative normal patient in each second preset time period, the consciousness state score, the nervous system score, the wound recovery score and the wakefulness state of the intraoperative normal patient, and the intraoperative risk degree of the intraoperative normal patient, and then the postoperative anesthesia complication patients and the postoperative normal patients are judged in real time. The intraoperative risk degree and the postoperative risk degree are accurately obtained in real time, so that the post-anesthesia complications of the patient can be accurately detected in time, and measures can be taken in time, and the harm of the post-anesthesia complications is effectively reduced.

The present application relates to the field of intelligent medical treatment and artificial intelligence, and discloses a preoperative precise evaluation system for gastric antrum cancer with pyloric obstruction based on AI image recognition, which comprises: a data acquisition and preprocessing module, which acquires structured clinical data and multi-modal image data of a patient and performs preprocessing; a lesion identification and segmentation module, which performs lesion identification and segmentation processing on the preprocessed image data by using an improved nnU-Net framework; a feature extraction and screening module, which extracts image features from the segmentation result and screens out features related to clinical indicators; a multi-modal feature fusion module, which performs feature fusion on the screened image features and the preprocessed structured clinical data; a multi-task risk evaluation module, which establishes a multi-parameter risk prediction model by using a semi-structured distribution regression method and outputs multi-task prediction results; and an evaluation report generation module, which generates a structured evaluation report. The present application realizes intelligent and precise preoperative evaluation and postoperative risk prediction for gastric antrum cancer patients.

The invention provides a postoperative risk prediction method and system based on multi-modal data fusion. The method comprises the following steps: acquiring image data and a clinical index data set; dividing the image data into a plurality of image blocks, and determining a plurality of tumor image blocks from the plurality of image blocks based on an image recognition technology; inputting the plurality of tumor image blocks into a preset first risk prediction model, so that the first risk prediction model performs feature extraction and linear mapping on each tumor image block based on a multi-instance learning framework to generate a corresponding first risk score; inputting the clinical index data set into a preset second risk prediction model, so that the second risk prediction model generates a corresponding second risk score according to a preset prediction factor and the clinical index data set; and performing weighted fusion on the first risk score and the second risk score through a preset statistical model to generate a postoperative risk prediction result, thereby improving the accuracy and reliability of postoperative risk prediction.

The invention discloses a colorectal cancer postoperative risk assessment method, device and equipment and a storage medium, and relates to the technical field of colorectal cancer postoperative assessment, and the colorectal cancer postoperative risk assessment method comprises the following steps: based on a multi-head attention mechanism, segmenting a centrum horizontal CT image of a preoperative and postoperative third lumbar vertebra of a target object, and obtaining a centrum horizontal CT image of a second lumbar vertebra of the target object; obtaining a skeletal muscle area SMA, a visceral fat area VFA and a subcutaneous fat area SFA corresponding to preoperative and postoperative operations; respectively calculating a skeletal muscle index SMI and a visceral fat to subcutaneous fat ratio VSR according to SMA, VFA and SFA before and after the operation; according to the gender of the target object and the preoperative and postoperative SMI and VSR, judging whether the target object is sarcopenia and / or visceral fat obesity; and dividing risk levels according to preoperative and postoperative judgment results. Based on a multi-head attention mechanism, automatic accurate segmentation and risk assessment of the colorectal cancer postoperative body composition are realized, and the analysis efficiency and accuracy are remarkably improved.

The invention provides a left auricle plugging device and a left auricle plugging device. The left auricle plugging device comprises a fixing disc and a plugging disc, the fixing disc is connected with the plugging disc, the fixing disc comprises a first support, the plugging disc comprises a second support, the first support and the second support are both of a net-shaped structure, and the radial size of the second support is gradually increased in the direction close to the first support. The left auricle plugging device solves the problem that in the prior art, after a left auricle plugging device is implanted, adverse postoperative risks are likely to be generated.

The invention discloses an intelligent pre-warning method and system for interventional postoperative complications based on multi-source data fusion, and relates to the technical field of risk assessment, and the method comprises the steps: carrying out the model training based on a historical data set, and obtaining an interventional postoperative complication risk assessment model; the characteristic value data set comprises vital sign data, laboratory examination data, intraoperative parameter data and instrument parameter data; performing standardization processing on the characteristic value data set to obtain a characteristic value matrix; and inputting the eigenvalue matrix into a risk assessment model for risk prediction to obtain a complication risk prediction result. The problems that intervention postoperative complication assessment depends on artificial experience, efficiency is low and subjectivity is high are solved, the technical problems that multi-source heterogeneous data are difficult to process and real-time risk prediction is insufficient are avoided, postoperative risk intelligent prediction, quantitative assessment and interpretable report generation based on multi-source data fusion are achieved, clinical decision is assisted, and the risk assessment efficiency is improved. And the prediction accuracy and the management efficiency are improved.

The invention provides a postoperative risk intelligent monitoring system and method based on multi-parameter fusion. The method comprises the following steps: acquiring ultrasonic echo data of a postoperative area; determining a plurality of apodization functions, and performing multi-parameter beam forming on the ultrasonic echo data according to all apodization functions; enhancing the ultrasonic echo data through the optimal apodization function to obtain an optimized ultrasonic signal; determining a total-graph coherence coefficient according to the ultrasonic echo data, determining conditional coherence factors for optimizing the ultrasonic signals, and performing multi-parameter fusion on all the conditional coherence factors based on the total-graph coherence coefficient to obtain a plurality of composite weighting coefficients; generating an optimized ultrasonic image according to the composite weighting coefficient and the optimized ultrasonic signal; and inputting the optimized ultrasonic image into a risk identification model to identify a risk target of the postoperative area. By adopting the scheme of the invention, multi-parameter optimization can be carried out on the ultrasonic image by considering the contrast ratio and noise suppression in a complex postoperative environment, so that accurate postoperative risk monitoring is realized.

The present application relates to the technical field of postoperative risk grading identification, in particular to a system for grading the adverse risk of distal remodeling after aortic dissection surgery. The system first determines a thrombus permeation resistance coefficient according to the gray level difference of CT images of the patient at different times, and extracts the thrombus isolation thickness; based on the thrombus permeation resistance coefficient, solves the anisotropic potential function equation to determine the cumulative leakage cost; performs structure tensor analysis on the gradient in the cumulative loss field formed by the cumulative leakage cost to determine the jet channel consistency factor; combines the cumulative leakage cost, the jet channel consistency factor and the thrombus isolation thickness to calculate the fluid impact load borne by the outer wall of the blood vessel; and based on the fluid impact load and the jet channel consistency factor, determines the risk level of different connected domains in the CT image. The present application improves the accuracy of jet impact risk identification.

The invention provides a postoperative concurrent risk prediction method, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: constructing an initial model based on a Transform architecture, and training the initial model based on a preset training data set to obtain a multi-modal deep learning radiology model; acquiring preoperative medical image data and clinical data of the to-be-detected object; based on the preoperative medical image data, radiomics features are extracted; carrying out feature fusion on the radiomics features and clinical data to obtain multi-modal input features; and inputting the multi-modal input features into a multi-modal deep learning radiology model, and sequentially processing the multi-modal input features by a multi-head attention module, a feedforward neural network module and a classification head module in the multi-modal deep learning radiology model to obtain the risk probability of the target postoperative complications of the to-be-detected object. The automatic prediction from the original image to the postoperative risk probability is realized, the human intervention is reduced, and the clinical availability is improved.