Construction method of central venous catheter intelligent decision model based on multi-source data

Abstract

The present application relates to the technical field of central venous catheter, and particularly relates to a method for constructing a central venous catheter intelligent decision model based on multi-source data. The method comprises the following steps: acquiring central venous catheter case data, and processing the data, obtaining patient sign data and catheter structure feature data by extracting the sign data of the case patient and analyzing the structure features of the central venous catheter; collecting a venous ultrasound image according to the patient sign data, and estimating the specification of the central venous catheter through ultrasound image analysis, while calculating the probability of catheter position displacement; collecting a mis-touch area according to the displacement probability data, and analyzing the abnormal state of the catheter in combination with the venous ultrasound image data; performing risk assessment of the catheter according to the abnormal state data, and constructing a central venous catheter intelligent decision model based on the catheter risk data; the present application optimizes the central venous catheter decision, so as to realize more accurate central venous catheter decision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of central venous catheter, in particular to a method for constructing a central venous catheter intelligent decision model based on multi-source data. BACKGROUND

[0002] Central venous vascular access device is a catheter with the tip positioned in the central vein. At present, the commonly used central venous vascular access in clinic includes three types: 1. Central venous catheter (CVC) inserted through the internal jugular vein, subclavian vein and femoral vein. 2. Peripherally central catheter (PlCC) inserted through the peripheral vein. 3. Implantable venous access port (IVAP) inserted through the internal jugular vein or subclavian vein. With the development of medicine, pharmacy, material science and other disciplines, various new intravascular catheters and auxiliary medical consumables have been gradually applied in clinic. Critically ill patients often need to indwell central venous catheter (not including catheters only used as intravenous access or blood purification), which can monitor hemodynamics and serve as an infusion access, while also considering the prevention and treatment of catheter indwelling mechanical complications, central line-associated blood-stream infection (CLABSI) and thrombosis. Clinicians need to choose appropriate catheters according to the patient's condition and adjust the type and position of the catheter at different stages of the disease. In order to standardize the management strategy of central venous catheter for critically ill patients, the traditional method for constructing a central venous catheter intelligent decision model based on multi-source data has the problems of inaccurate estimation of central venous catheter specifications and inaccurate risk analysis of central venous catheter. SUMMARY

[0003] Therefore, it is necessary to provide a method for constructing a central venous catheter intelligent decision model based on multi-source data to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, a method for constructing a central venous catheter intelligent decision model based on multi-source data comprises the following steps:

[0005] Step S1: obtaining central venous catheter case data; extracting case patient signs according to the central venous catheter case data to obtain case patient sign data; analyzing the structure characteristics of the central venous catheter according to the central venous catheter case data to obtain the structure characteristic data of the central venous catheter;

[0006] Step S2: collecting case patient venous ultrasound image data according to case patient sign data; estimating central venous catheter specification data according to case patient venous ultrasound image data; calculating catheter position displacement probability data according to central venous catheter specification data and case patient venous ultrasound image data;

[0007] Step S3: collecting central venous catheter mis-touch area data according to central venous catheter displacement probability data; analyzing central venous catheter abnormal state data according to central venous catheter mis-touch area data and case patient venous ultrasound image data;

[0008] Step S4: performing central venous catheter risk assessment based on central venous catheter abnormal state data to obtain central venous catheter risk data; constructing a central venous catheter intelligent decision-making model based on central venous catheter risk data.

[0009] The present application can comprehensively monitor and evaluate the use state and related risks of central venous catheters through multi-source data fusion and intelligent analysis, thereby improving the safety and accuracy in the clinical treatment process. By obtaining case data of central venous catheters and performing sign extraction and catheter structure feature analysis, not only is the foundation data provided for subsequent risk assessment, but also the physiological characteristics of the patient and the structural characteristics of the catheter are comprehensively mastered, laying a good foundation for accurate analysis. Through the collection and analysis of venous ultrasound image data, the catheter specification can be estimated in combination with the patient's sign data, and the specification and position change of the central venous catheter can be accurately mastered. This process helps to understand the position of the catheter in the blood vessel and its stability, providing reliable data basis for subsequent catheter displacement probability calculation. Based on the catheter displacement probability data, the mis-touch area data is further collected, the area contacted by the catheter is analyzed, and the abnormal state is identified. This analysis can help identify potential risks caused by abnormal catheter position, such as blood vessel damage and blood flow obstruction. Risk assessment based on abnormal state data helps to comprehensively evaluate various problems encountered by the catheter during use, such as thrombosis, infection, blood vessel damage and other potential complications, enhancing the safety during treatment. Therefore, the present application optimizes the construction method of a traditional central venous catheter intelligent decision-making model based on multi-source data, solves the problems of inaccurate central venous catheter specification estimation and inaccurate central venous catheter risk analysis in the traditional construction method of a central venous catheter intelligent decision-making model based on multi-source data, and improves the accuracy of central venous catheter specification estimation and the accuracy of central venous catheter risk analysis.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: Obtain central venous catheter case data;

[0012] Step S12: Perform case desensitization processing according to the central venous catheter case data to obtain central venous catheter case desensitization data;

[0013] Step S13: Perform case patient sign extraction according to the central venous catheter case desensitization data to obtain case patient sign data;

[0014] Step S14: Perform central venous catheter structure feature analysis according to the central venous catheter case desensitization data to obtain central venous catheter structure feature data.

[0015] The present application can ensure patient privacy protection while providing high-quality data basis for subsequent analysis by obtaining central venous catheter case data and performing desensitization processing. The desensitized case data protects the personal information of patients and ensures the security of the data. Based on these desensitized data, the patient's sign information is further extracted to comprehensively understand the patient's health status and provide reliable physiological data support for subsequent analysis. Combined with the central venous catheter case desensitization data, the catheter structure feature analysis helps to accurately grasp the relevant characteristics of the catheter, including the morphology, specification and location characteristics of the catheter, which is of great significance for the use and maintenance of the catheter. Through detailed analysis of the catheter structure features, the relationship between the catheter and the patient's blood vessels is better understood, and potential risk factors or potential problems are found, providing a scientific basis for subsequent treatment and operation. Overall, this series of steps can provide more accurate basic data for the safe use and management of central venous catheters by extracting valuable information from case data, ensuring that potential problems in the use of catheters are identified and avoided in advance.

[0016] Preferably, step S2 comprises the following steps:

[0017] Step S21: Perform case patient vein ultrasound image acquisition according to the case patient sign data to obtain case patient vein ultrasound image data;

[0018] Step S22: Perform vein feature analysis according to the case patient vein ultrasound image data to obtain ultrasound image vein feature data;

[0019] Step S23: Perform central venous catheter specification estimation based on the ultrasound image vein feature data and the central venous catheter structure feature data to obtain central venous catheter specification data;

[0020] Step S24: according to the central venous catheter specification data and the ultrasonic image vein feature data, catheter position displacement probability calculation is performed to obtain central venous catheter displacement probability data.

[0021] The present application collects the ultrasonic image data of the case patient, obtains detailed vein image data, and provides direct image data support for subsequent vein feature analysis. The vein feature analysis helps to deeply understand the morphology, structure and dynamic change of the vein by extracting the key features in the image, thereby providing a reliable basis for accurately evaluating the vein state. The central venous catheter structure features are estimated by using these vein feature data, the adaptability and specification requirements of the catheter are accurately predicted, and reference basis is provided for selecting the appropriate catheter model. Further, the catheter position displacement probability is calculated by combining the catheter specification data and the vein feature data, which helps to predict and evaluate the displacement risk of the catheter. Through these steps, the adaptability and safety of the catheter are comprehensively evaluated, and scientific basis is provided for risk management in the subsequent treatment process. Overall, these steps can improve the accuracy of operation and reduce the potential risk of complications in the process of using the catheter through in-depth analysis of the ultrasonic image data of the vein.

[0022] Preferably, step S22 comprises the following steps:

[0023] Step S221: according to the case patient ultrasonic image data, image texture feature acquisition is performed to obtain ultrasonic image texture feature data;

[0024] Step S222: based on the ultrasonic image texture feature data and the case patient ultrasonic image data, vein vessel edge feature analysis is performed to obtain vein vessel edge feature data;

[0025] Step S223: according to the vein vessel edge feature data and the ultrasonic image texture feature data, contour feature analysis is performed to obtain vein vessel contour data;

[0026] Step S224: based on the vein vessel contour data and the vein vessel edge feature data, vein feature analysis is performed to obtain ultrasonic image vein feature data.

[0027] The present application collects texture features of the case patient's venous ultrasound image, extracts the subtle structural changes in the image, and provides important data support for the morphological analysis of the vein. The texture feature data helps to reveal the organizational characteristics of the venous blood vessels, such as wall smoothness, uniformity of the blood vessel wall, etc., and further reflects the health status of the vein. Using these texture feature data and venous image information for edge feature analysis helps to accurately identify the boundary of the blood vessel, thereby improving the positioning accuracy of the venous blood vessel and reducing errors. Then, combined with the edge feature data and the texture feature data, the contour feature analysis is carried out, and the morphological contour of the blood vessel is further clarified, which provides support for the comprehensive evaluation of the vein morphology. Based on the venous blood vessel contour feature data and the edge feature data, comprehensive venous feature analysis is carried out, which can extract the key characteristics of the vein from all directions and multiple angles, and further provide more comprehensive and accurate image data for clinical application. The implementation of these steps improves the accuracy and depth of the venous image analysis, which helps to effectively identify potential risks.

[0028] Preferably, step S222 comprises the following steps:

[0029] According to the case patient's venous ultrasound image data, the ultrasound image gray value is calculated, so as to obtain the venous ultrasound image gray value data;

[0030] According to the venous ultrasound image gray value data, the ultrasound image global gray value standard deviation is calculated, so as to obtain the ultrasound image global gray value standard deviation data;

[0031] According to the ultrasound image global gray value standard deviation data, the ultrasound image gray value contrast is analyzed, so as to obtain the ultrasound image gray value contrast data;

[0032] According to the ultrasound image texture feature data, the texture feature direction gradient change is calculated, so as to obtain the image texture feature direction gradient change data;

[0033] Based on the image texture feature direction gradient change data and the ultrasound image gray value contrast data, the venous blood vessel edge feature analysis is carried out, so as to obtain the venous blood vessel edge feature data.

[0034] The present application provides detailed quantitative data for the intensity distribution of the image by calculating the gray value of the ultrasound image of the vein, which helps to analyze the subtle differences of the vein wall and the dynamic changes of the blood flow. Further, by calculating the global gray value standard deviation of the image, the dispersion degree of the gray value in the image is revealed, so as to evaluate the definition and contrast of the image. This process helps to remove noise and improve image quality. Next, the gray value contrast analysis can quantitatively process the local differences of the image, helping to accurately identify the boundary between the vein structure and the surrounding tissue, and improving the accuracy of blood vessel recognition. The texture feature direction gradient change calculation reveals the change rule of the texture in different directions in the image, further extracts detailed information, and reflects the density change of the tissue around the blood vessel. Combined with these data, the vein edge feature analysis helps to accurately outline the edge contour of the vein, improves the resolution of the image, and provides more accurate vein position and shape data.

[0035] Preferably, step S23 comprises the following steps:

[0036] Step S231: performing image vein fine-grained analysis according to the ultrasound image vein feature data to obtain ultrasound image vein fine-grained data;

[0037] Step S232: performing image vein wall thickness detection based on the ultrasound image vein fine-grained data to obtain ultrasound image vein wall thickness data;

[0038] Step S233: performing image vein position measurement on the ultrasound image vein feature data to obtain ultrasound image vein position data;

[0039] Step S234: performing image vein depth detection according to the ultrasound image vein position data to obtain ultrasound image vein depth data;

[0040] Step S235: performing central vein catheter specification estimation based on the ultrasound image vein depth data to obtain central vein catheter specification data.

[0041] The present application can improve the clarity and detail performance of the image by analyzing the vein image in a fine-grained manner, thereby providing more accurate basic data for subsequent analysis. The vein wall thickness detection helps to determine the health status of the blood vessel and timely identify whether the blood vessel wall has thickening phenomenon, which is very important for evaluating the elasticity and potential lesions of the blood vessel. The vein position measurement can accurately determine the spatial position of the vein in the image, providing key data for positioning and further improving the visibility and positioning accuracy of the blood vessel structure in the image. The depth detection ensures the position of the vein in the three-dimensional space and its relative relationship with the surrounding tissues, which helps to comprehensively understand the spatial distribution and state change of the blood vessel. By combining the central vein catheter specification estimation with the vein depth data, more accurate catheter specification information can be obtained, and the adaptability and safety of the catheter can be further optimized.

[0042] Preferably, step S3 comprises the following steps:

[0043] Step S31: collecting the central vein catheter mis-touch area according to the central vein catheter displacement probability data, to obtain central vein catheter mis-touch area data;

[0044] Step S32: estimating the local thrombosis probability according to the central vein catheter mis-touch area data and the case patient vein ultrasound image data, to obtain ultrasound image local thrombosis probability data;

[0045] Step S33: analyzing the central vein catheter abnormal state according to the ultrasound image local thrombosis probability data and the central vein catheter mis-touch area data, to obtain central vein catheter abnormal state data.

[0046] The present application can accurately identify the potential changes of the catheter position by analyzing the central vein catheter displacement probability data, and further help to locate the mis-touch area to avoid damage to the surrounding tissues. By collecting the mis-touch area, the dangerous area of the blood vessel and the catheter contact can be identified in time, which provides effective basis for the subsequent prevention work. By combining the ultrasound image data and the contact area information, the probability of local thrombosis is evaluated, so as to discover the potential thrombosis risk in advance, which is of great significance for preventing thrombus complications. Further analysis of the local thrombosis probability data helps to evaluate whether the central vein catheter is in an abnormal state, and reveals the problems of blood flow obstruction or catheter blockage. The central vein catheter abnormal state analysis can provide scientific basis for identifying the abnormal function of the catheter.

[0047] Preferably, step S32 comprises the following steps:

[0048] Collecting the vein morphology change to obtain ultrasound image vein morphology change data;

[0049] Step S322: vein vessel puncture detection is performed according to the vein shape change data of the ultrasound image, and vein vessel puncture data of the ultrasound image is obtained;

[0050] Step S323: vein vessel injury detection is performed according to the vein vessel puncture data of the ultrasound image, and vein vessel injury data of the ultrasound image is obtained;

[0051] Step S324: vessel spasm analysis is performed according to the vein vessel injury data of the ultrasound image, and vessel spasm probability data of the ultrasound image is obtained;

[0052] Step S325: vein blood flow obstruction estimation is performed according to the central venous catheter mis-touch region data and the vein vessel injury data of the ultrasound image, and vein blood flow obstruction data is obtained;

[0053] Step S326: local thrombosis probability estimation is performed based on the vessel spasm probability data of the ultrasound image and the vein blood flow obstruction data, and local thrombosis probability data of the ultrasound image is obtained.

[0054] The present application can accurately capture the morphological characteristics of the vein in different states through the collection of vein shape changes, which provides basic data for further analysis of the health status of the vein. Through vein puncture detection of the ultrasound image vein shape change data, it is effective to judge whether the catheter touches the blood vessel, so as to identify the puncture risk and reduce unnecessary damage. On the basis of the blood vessel puncture data, the blood vessel injury detection can early identify the micro-injury or rupture of the blood vessel, prevent further damage to the blood vessel or cause more serious complications. Further analysis of the blood vessel injury, probability analysis of the blood vessel spasm, and prediction of the blood flow problem provide guidance for prevention and intervention. At the same time, by combining the central venous catheter mis-touch region data and the blood vessel injury data, the risk of blood flow obstruction can be predicted in advance, so as to reduce the complications caused by blood flow obstruction. Finally, based on the combination of the blood vessel spasm probability and the blood flow obstruction data, the probability of local thrombosis can be accurately estimated, which provides strong data support for subsequent treatment and intervention measures.

[0055] Preferably, step S33 comprises the following steps:

[0056] Step S331: vein vessel thrombus enlargement analysis is performed according to the local thrombosis probability data of the ultrasound image, and vein vessel thrombus enlargement data is obtained;

[0057] Step S332: vein vessel occlusion detection is performed based on the vein vessel thrombus enlargement data, and vein vessel occlusion data is obtained;

[0058] Step S333: central venous catheter function loss detection is performed according to the vein vessel occlusion data and the vein vessel thrombus enlargement data, and central venous catheter function loss data is obtained;

[0059] Step S334: estimating the probability of venous infection according to the local thrombosis probability data of the ultrasound image and the central venous catheter mis-touch area data, to obtain venous infection probability data;

[0060] Step S335: analyzing the abnormal state of the central venous catheter based on the venous infection probability data and the central venous catheter function loss data, to obtain central venous catheter abnormal state data.

[0061] The present application can monitor the development and change of thrombus in the vein through the analysis of vein thrombus enlargement, thereby warning the growth trend of thrombus and providing early indication for timely treatment. The detection of vein vessel occlusion combined with the data of thrombus enlargement can accurately identify whether thrombus causes blood flow obstruction, thereby further determining whether intervention measures need to be taken to avoid the aggravation of occlusion. Through the combination of thrombus enlargement and vessel occlusion data, central venous catheter function loss detection can be performed to timely discover the decline of catheter function caused by thrombus or occlusion, thereby avoiding the occurrence of serious complications. In addition, by combining the local thrombosis probability data and the central venous catheter mis-touch area data, the risk of venous infection can be evaluated to discover the signs of infection early and avoid the spread of infection and endanger the health of patients. Finally, based on the venous infection probability data and the catheter function loss data, the abnormal state analysis of the central venous catheter can be performed to comprehensively evaluate the state of the catheter and provide a reliable basis for subsequent treatment strategies.

[0062] Preferably, step S4 comprises the following steps:

[0063] Step S41: detecting the venous function impairment based on the central venous catheter abnormal state data, to obtain venous function impairment data;

[0064] Step S42: estimating the venous catheter complication according to the venous function impairment data and the central venous catheter abnormal state data, to obtain central venous catheter complication data;

[0065] Step S43: performing the risk assessment of the central venous catheter based on the venous catheter complication data and the venous function impairment data, to obtain central venous catheter risk data;

[0066] Step S44: constructing the intelligent decision-making model of the central venous catheter based on the central venous catheter risk data, to obtain the intelligent decision-making model of the central venous catheter.

[0067] The present application can early identify the venous dysfunction caused by the abnormal state of the central venous catheter through the detection of venous function impairment, so as to intervene in time and reduce the long-term impact on the health of the patient. This step helps to confirm whether the catheter has caused a change in hemodynamics, so as to determine whether the catheter position needs to be adjusted. In combination with the venous function impairment data and the catheter abnormal state data, the complication estimation is carried out, the type and development trend of the complication are identified, the occurrence of the venous related problem is predicted in advance, the clinical early warning capability is improved, and the potential serious complication is avoided. Further risk assessment is carried out based on the complication data, and personalized risk assessment is provided for different patients.

[0068] The present application is to improve the safety and accuracy in the clinical treatment process by comprehensively monitoring and evaluating the use state of the central venous catheter and its related risks through multi-source data fusion and intelligent analysis. By acquiring the central venous catheter case data and carrying out the sign extraction and catheter structure feature analysis, not only the basic data for subsequent risk assessment is provided, but also the physiological characteristics of the patient and the structural characteristics of the catheter are comprehensively mastered, which lays a good foundation for accurate analysis. Through the collection and analysis of the venous ultrasound image data, the catheter specification can be estimated in combination with the sign data of the patient, and the specification and position change of the central venous catheter can be accurately mastered. This process helps to deeply understand the position of the catheter in the blood vessel and its stability, and provides reliable data basis for subsequent catheter displacement probability calculation. Based on the catheter displacement probability data, the mis-touch area data is further collected, the area contacted by the catheter is analyzed, and the abnormal state is identified. This analysis can help to identify the potential risks caused by the abnormal position of the catheter, such as blood vessel injury and blood flow obstruction. The risk assessment based on the abnormal state data helps to comprehensively evaluate various problems encountered by the catheter in the use process, such as thrombus, infection, blood vessel injury and other potential complications, and enhances the safety in the treatment process. Therefore, the present application optimizes the construction method of the traditional central venous catheter intelligent decision model based on multi-source data, solves the problems of inaccurate central venous catheter specification estimation and inaccurate central venous catheter risk analysis in the traditional construction method of the central venous catheter intelligent decision model based on multi-source data, and improves the accuracy of the central venous catheter specification estimation and the accuracy of the central venous catheter risk analysis. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 The step flowchart of the construction method of the central venous catheter intelligent decision model based on multi-source data is shown in the figure.

[0070] Figure 2 The detailed implementation step flowchart of step S2 is shown in the figure. Figure 1

[0071] ​Figure 3 To Figure 1 Detailed implementation steps of step S4 are shown in a flowchart.

[0072] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0073] The technical method of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0074] In addition, the accompanying drawings are only schematic illustrations of the present application, and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0075] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be called a second element, and similarly a second element can be called a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0076] To achieve the above-mentioned purpose, please refer to Figures 1 to 3 A method for constructing a central venous catheter intelligent decision model based on multi-source data, comprising the following steps:

[0077] Step S1: acquiring central venous catheter case data; extracting case patient signs according to the central venous catheter case data to obtain case patient sign data; analyzing the structure characteristics of the central venous catheter according to the central venous catheter case data to obtain central venous catheter structure characteristic data;

[0078] Step S2: according to the case patient sign data, the case patient vein ultrasound image data is obtained by collecting the case patient vein ultrasound image; according to the case patient vein ultrasound image data, the central venous catheter structure feature data is estimated to obtain the central venous catheter specification data; according to the central venous catheter specification data and the case patient vein ultrasound image data, the catheter position displacement probability is calculated, and the central venous catheter displacement probability data is obtained;

[0079] Step S3: according to the central venous catheter displacement probability data, the central venous catheter mis-touch area data is obtained by collecting the central venous catheter mis-touch area; according to the central venous catheter mis-touch area data and the case patient vein ultrasound image data, the central venous catheter abnormal state analysis is carried out, and the central venous catheter abnormal state data is obtained;

[0080] Step S4: based on the central venous catheter abnormal state data, the central venous catheter risk data is obtained by carrying out the central venous catheter risk assessment; based on the central venous catheter risk data, the central venous catheter intelligent decision model is constructed, and the central venous catheter intelligent decision model is obtained.

[0081] In the embodiment of the application, reference Figure 1 As shown in the figure, it is a step flow schematic diagram of the construction method of the central venous catheter intelligent decision model based on multi-source data, in the example, the construction method of the central venous catheter intelligent decision model based on multi-source data includes the following steps:

[0082] Step S1: obtain the central venous catheter case data; according to the case patient sign data, the case patient sign data is obtained by extracting the case patient sign; according to the central venous catheter case data, the central venous catheter structure feature analysis is carried out, and the central venous catheter structure feature data is obtained;

[0083] In the embodiments of the present application, the central venous catheter case data is obtained by extracting relevant data from real clinical patient cases through the case database of medical institutions or special data collection equipment. The obtained data should include the basic information of the patient (such as age, gender, medical history, etc.), the venous condition in the case and the related medical image data. These data include the placement of the venous catheter, ultrasound image data, etc. Then after data integration, data desensitization processing is performed to remove all personal sensitive information, and through data cleaning and preprocessing, effective and required case data is screened out. The case data is subjected to sign extraction, and medical imaging analysis tools (such as image processing software) are used to extract the sign data of the vein, including the size, position, shape of the vein and the physiological characteristics of the patient (such as blood pressure, heart rate, etc.). At the same time, the structural characteristics of the venous catheter are analyzed in detail by using imaging analysis technology, and the size, shape, direction and structure of the surrounding tissue of the central venous catheter are obtained. These data constitute the structural characteristic data of the central venous catheter.

[0084] Step S2: collecting the case patient venous ultrasound image data according to the case patient sign data; estimating the central venous catheter structural characteristic data according to the case patient venous ultrasound image data; calculating the catheter position shift probability data according to the central venous catheter specification data and the case patient venous ultrasound image data;

[0085] In the embodiments of the present application, after the case patient sign data is extracted, the venous ultrasound image is collected. High-resolution ultrasound equipment is used to scan the patient's vein in real time to obtain high-definition vein image data. During the image collection process, the position, shape, surrounding tissue and relative position of the venous catheter of the vein are focused on. Through image processing and analysis software, combined with image registration technology, the collected vein image is matched with the existing venous catheter structural characteristic data. Then, according to these image data and the existing structural characteristic data, the specification of the central venous catheter is estimated through geometric modeling and machine learning method to obtain accurate catheter size data. This process uses image segmentation technology based on morphological analysis, and through feature extraction and pattern recognition algorithm, the specific specification and shape change of the venous catheter are calculated. At the same time, the shift probability of the catheter position is calculated by using the vein image data, and the probability calculation model and image registration technology are applied to analyze the stability of the catheter in different positions, and the probability data of the central venous catheter shift is obtained.

[0086] Step S3: Collecting central venous catheter mis-touch area data according to the central venous catheter displacement probability data; analyzing the abnormal state of the central venous catheter according to the central venous catheter mis-touch area data and the case patient vein ultrasound image data to obtain central venous catheter abnormal state data;

[0087] In the embodiment of the application, the central venous catheter displacement probability data obtained is used for collecting the catheter mis-touch area, defining the standard of the catheter mis-touch area, and marking the vein catheter mis-touch area through an image processing tool. Then, based on the patient vein ultrasound image data and combined with the accurate calculation result of the catheter position, the area of the catheter mis-touch is analyzed. This process involves image region segmentation technology, and by comparing with the case data, the adjacent tissue contacted by the catheter is found and calibrated. Next, based on the data of the mis-touch area, the abnormal state analysis of the central venous catheter is performed. This analysis process uses a multi-dimensional algorithm to evaluate whether the central venous catheter has an abnormal state, including blood flow obstruction, injury and other problems caused by the catheter, and abnormal state data is obtained.

[0088] Step S4: Performing central venous catheter risk assessment based on the central venous catheter abnormal state data to obtain central venous catheter risk data; and constructing a central venous catheter intelligent decision-making model based on the central venous catheter risk data to obtain the central venous catheter intelligent decision-making model.

[0089] In the embodiment of the application, after the central venous catheter abnormal state data, the risk assessment of the central venous catheter is performed. The data analysis tool and algorithm model are used to comprehensively evaluate the abnormal state of the catheter and analyze the complication risk caused by the abnormal state. This process relies on a machine learning model to evaluate the combination of different risk factors by inputting the catheter abnormal state data, patient sign data and image data. The algorithm quantifies the risks faced by the catheter by comparing the risk situations in different cases and combining standard clinical processes. Finally, according to the risk assessment result, a central venous catheter intelligent decision-making model is constructed. Based on the evaluation data, the model uses an advanced decision support system (such as a rule-based system or a deep learning model) to optimize catheter management and provide personalized risk warning.

[0090] Preferably, step S1 comprises the following steps:

[0091] Step S11: Obtaining central venous catheter case data;

[0092] Step S12: Performing case desensitization processing according to the central venous catheter case data to obtain central venous catheter case desensitization data;

[0093] Step S13: case patient sign data is obtained by performing case patient sign extraction according to the central venous catheter case desensitization data.

[0094] Step S14: central venous catheter structure feature data is obtained by performing central venous catheter structure feature analysis according to the central venous catheter case desensitization data.

[0095] In the embodiments of the present application, central venous catheter related case data is extracted from hospital databases, electronic medical record systems (EMR) or medical imaging systems (PACS). These data usually include patient's basic information (such as age, gender, hospitalization time, surgical history), catheter type, usage, catheter placement location, surgical record, treatment plan, follow-up record and ultrasound image data, etc. Using data interface technology (such as HL7 protocol or FHIR standard), these data are extracted from different systems and formatted uniformly. Ensure the integrity and accuracy of the data, remove irrelevant data and mark. The de-identification processing of central venous catheter case data adopts standard de-identification technology. In this step, personal identity information is encrypted or replaced to ensure data privacy. Common de-sensitization methods include masking processing of data fields or replacing sensitive information such as name, ID number, contact information through encryption algorithm. In specific operation, hash algorithm is used to process sensitive fields (such as patient ID, name, hospital number, etc.), so that the real identity cannot be deduced reversely. At the same time, non-sensitive data (such as diagnosis results, treatment plan, etc.) are still retained. This operation complies with data protection regulations (such as GDPR or HIPAA) and ensures the availability of data for subsequent analysis. After de-identification processing, the central venous catheter case de-identification data is obtained. Based on the obtained and de-identified central venous catheter case data, the patient's physical signs are extracted, and the physiological signs of the patient are extracted from the case by using medical image processing tools and algorithms (such as OpenCV or Matlab-based image processing toolkits), including blood pressure, heart rate, blood oxygen saturation and other conventional physical sign information. For image data, the specific position and shape characteristics of the venous catheter in the ultrasound image are extracted by image segmentation technology. Then, the diameter, length, shape, lumen state and structure characteristics of the surrounding tissue of the vein are extracted by pattern recognition algorithm. In the process of extracting the physical signs, automatic algorithm is applied to automatically analyze the image, reduce human intervention, improve efficiency and accuracy, and obtain the physical sign data of the case patient. When analyzing the structural characteristics of the central venous catheter case data, the morphology and structural characteristics of the catheter are analyzed by using imaging technology. This analysis is completed by image enhancement and detail enlargement of the ultrasound image, and the outline of the catheter and the surrounding tissue are clearly displayed by using filtering technology, and the size of the catheter is measured by using geometric analysis method, such as diameter, curvature, length, etc., and the relationship between the catheter and the surrounding tissue is evaluated, such as whether there are compression or damage signs. Through the analysis of multiple image frames, the three-dimensional structure data of the catheter are accurately obtained. In addition, combined with the physical sign data of the patient, statistical analysis method is used to explore the correlation between the structural characteristics of the catheter and the health status of the patient, and the structural characteristic data of the central venous catheter are obtained.

[0096] Preferably, step S2 comprises the following steps:

[0097] Step S21: Acquire venous ultrasound images of the patient based on the patient's vital signs data to obtain venous ultrasound image data of the patient.

[0098] Step S22: Perform vein feature analysis based on the venous ultrasound image data of the case patient to obtain venous feature data from the ultrasound image;

[0099] Step S23: Based on the venous feature data of the ultrasound image, estimate the central venous catheter specifications by analyzing the structural feature data of the central venous catheter to obtain the central venous catheter specification data;

[0100] Step S24: Calculate the probability of catheter displacement based on the central venous catheter specification data and the venous feature data of the ultrasound image to obtain the central venous catheter displacement probability data.

[0101] As an example of the present invention, reference is made to... Figure 2 As shown, in this example, step S2 includes:

[0102] Step S21: Acquire venous ultrasound images of the patient based on the patient's vital signs data to obtain venous ultrasound image data of the patient.

[0103] In this embodiment of the invention, an ultrasound diagnostic device is used to scan the patient. The ultrasound device employs a high-frequency linear array probe, which can provide high-resolution real-time images, and is particularly suitable for the detection of blood vessels and veins. Based on the patient's specific physical characteristics, such as body size, position, and health status, an appropriate probe frequency and imaging mode (e.g., B-mode ultrasound imaging) are selected to ensure accurate display of the veins and surrounding tissues. During ultrasound image acquisition, a professional technician operates the ultrasound device to scan common venous catheter implantation areas such as the neck, chest, or upper limbs, ensuring correct scanning angles and positions to obtain clear venous images. During image acquisition, special attention is paid to venous blood flow, catheter location, and the structure of surrounding tissues.

[0104] Step S22: Perform vein feature analysis based on the venous ultrasound image data of the case patient to obtain venous feature data from the ultrasound image;

[0105] In the embodiment of the present application, an image processing algorithm is used to pre-process the ultrasound image, including denoising, contrast enhancement, and edge sharpening operations, to improve image quality. Using image segmentation techniques, the venous region is extracted, and the boundary of the venous blood vessel is separated by threshold segmentation, edge detection, or region growing algorithm. Morphological analysis methods are used to further describe the venous morphology, and the geometric features of the vein are extracted, including the diameter, curvature, length, and wall thickness of the vein. For blood flow information, the blood flow velocity and direction are measured by Doppler ultrasound, and the hemodynamic characteristics are analyzed by time-velocity spectrum. In addition, the spatial relationship between the vein and the surrounding structure is extracted, such as the smoothness of the vein wall, whether the vessel wall is deformed, etc. This process applies image processing platforms such as OpenCV, Matlab, etc. to extract key feature data of the vein through automated methods, and stores the data as structured data.

[0106] Step S23: Based on the venous feature data of the ultrasound image, the central venous catheter structure feature data is estimated for the central venous catheter specification, and the central venous catheter specification data is obtained.

[0107] In the embodiment of the present application, based on the venous ultrasound image feature data, the specification of the central venous catheter is estimated, and the venous structure identified in the ultrasound image is geometrically modeled. This process measures the diameter, depth, and curvature of the vein, and combines the patient's physical data to estimate the catheter size and specification. For this purpose, a geometric-based algorithm (such as least squares fitting) is used to fit the morphological characteristics of the catheter. In addition, considering the individual differences of different patients, by comparing the known central venous catheter standard size data (such as the inner diameter, outer diameter, and length of the catheter), and combining the venous profile of the ultrasound image, a regression analysis method is used to further estimate the specification of the central venous catheter.

[0108] Step S24: According to the central venous catheter specification data and the ultrasound image venous feature data, the catheter position shift probability is calculated, and the central venous catheter shift probability data is obtained.

[0109] In the embodiments of the present application, on the basis of obtaining the specification data of the central venous catheter and the venous feature data, when the catheter position displacement probability calculation is performed, the relative position change of the catheter is calculated by using the venous position in the venous ultrasound image and the positioning information of the catheter relative to the vein by using a geometric analysis method. Through the measurement of the two-dimensional coordinates of the catheter in the image and the three-dimensional spatial position, combined with the physical sign data of the patient, the deviation between the initial position and the current position of the catheter is determined. In order to estimate the probability of catheter displacement, a statistical model based on the Monte Carlo method is used to simulate the catheter position deviation under different conditions. Through multiple simulations and probability statistics, the probability and degree of catheter displacement are obtained, and the risk level of catheter displacement is further accurately calculated. This process applies programming environments such as Matlab and Python, combines numerical calculation and the Monte Carlo method, and completes the quantitative analysis of the catheter displacement probability.

[0110] Preferably, step S22 comprises the following steps:

[0111] Step S221: Acquire image texture features according to the case patient venous ultrasound image data to obtain ultrasound image texture feature data;

[0112] Step S222: Perform venous blood vessel edge feature analysis based on the ultrasound image texture feature data and the case patient venous ultrasound image data to obtain venous blood vessel edge feature data;

[0113] Step S223: Perform contour feature analysis according to the venous blood vessel edge feature data and the ultrasound image texture feature data to obtain venous blood vessel contour data;

[0114] Step S224: Perform venous feature analysis based on the venous blood vessel contour data and the venous blood vessel edge feature data to obtain ultrasound image venous feature data.

[0115] In the embodiments of the present application, the acquired ultrasound images are preprocessed using image processing tools, including noise removal, contrast enhancement, and edge sharpening operations, to improve image quality. Texture analysis methods, such as gray level co-occurrence matrix (GLCM) or local binary pattern (LBP) techniques, are used to extract texture features from the ultrasound images. Specifically, the gray level co-occurrence matrix analyzes the gray level variation patterns of pixel pairs in the image to obtain feature parameters that describe texture, such as contrast, energy, entropy, and correlation. The local binary pattern describes the local texture structure of the image through the pixel relationships in local regions. In ultrasound images, texture features help identify the uniformity of the blood vessel wall, the pattern of blood flow within the vessel lumen, and their contrast with surrounding tissues. Using programming tools such as Matlab or Python, combined with image processing libraries (such as OpenCV, scikit-image, etc.), texture feature data is automatically extracted. Based on the ultrasound image texture feature data and the case patient's venous ultrasound image data, edge detection of the venous blood vessels is performed. Using the Canny edge detection algorithm or the Sobel operator, the contour information of the venous blood vessels is obtained by identifying the changes in the venous edges in the ultrasound images. At this time, further smoothing of the image is required to reduce noise interference and ensure the accuracy of edge detection. After edge detection, the image contour tracking algorithm (such as the Ramer-Douglas-Peucker algorithm) is used to finely extract the blood vessel edges, obtaining more accurate blood vessel edge data. By analyzing the boundaries of the blood vessels and surrounding tissues in the image, the edge features of the venous blood vessels are obtained, such as the smoothness, sharpness of the edges, and the contrast between the blood vessels and surrounding tissues. Using the image processing libraries in Matlab or Python (such as OpenCV, scikit-image, etc.), edge feature analysis is performed, and the results are converted into numerical data form. Based on the venous blood vessel edge feature data and the ultrasound image texture feature data, contour feature analysis is performed. By combining edge feature data and texture data, the overall contour of the venous blood vessels is further analyzed. Through morphological transformation techniques, such as dilation, erosion, and opening operations, small area noise and defects are eliminated, and a clearer blood vessel contour is extracted. On this basis, polynomial fitting or spline interpolation algorithms (such as B-spline curves or Bezier curves) are used to model the blood vessel contour, obtaining a continuous description of the blood vessel contour. By calculating the smoothness, curvature, and changes in the diameter of the blood vessel lumen, the regularity of the blood vessel shape and its impact on catheter placement are evaluated. The obtained contour feature data includes the precise size, curvature of the blood vessels, and the boundaries of surrounding tissues. Through data analysis software (such as Matlab) combined with texture feature data, the contour data of the venous blood vessels is obtained.When analyzing the venous features based on the venous vessel contour data and the venous vessel edge feature data, the contour and edge feature data are integrated, and the overall structure of the venous vessel is analyzed. By analyzing the shape, diameter, wall thickness, and relationship with the surrounding tissue of the venous vessel, the geometric features of the venous vessel are extracted. This process can use morphological analysis tools (such as Voronoi diagram-based segmentation techniques) to identify the opening and branch positions of the venous vessel, and further analyze the changes of the vessel branches, bends, and surrounding tissues. To accurately extract the features of the venous vessel, curve fitting or morphological contour analysis methods are used to extract the geometric properties of the vessel from the venous edge and contour, such as the size, shape, continuity, etc. of the vessel. Combined with the texture features of the ultrasound image, the uniformity of the vessel wall and the distribution of the blood flow are evaluated. The image processing tools and algorithms in the analysis process, such as Matlab's Image Processing Toolbox and Python's scikit-image library, can realize the extraction and analysis of such venous features.

[0116] Preferably, step S222 comprises the following steps:

[0117] According to the case patient venous ultrasound image data, the ultrasound image gray value data is calculated, and the venous ultrasound image gray value data is obtained;

[0118] According to the venous ultrasound image gray value data, the ultrasound image global gray value standard deviation data is calculated, and the ultrasound image global gray value standard deviation data is obtained;

[0119] According to the ultrasound image global gray value standard deviation data, the ultrasound image gray value contrast data is analyzed, and the ultrasound image gray value contrast data is obtained;

[0120] According to the ultrasound image texture feature data, the texture feature direction gradient change data is calculated, and the image texture feature direction gradient change data is obtained;

[0121] Based on the image texture feature direction gradient change data and the ultrasound image gray value contrast data, the venous vessel edge feature data is analyzed, and the venous vessel edge feature data is obtained.

[0122] In the embodiment of the present application, after obtaining the ultrasound image data, the image preprocessing step is used to remove noise and clutter to improve the image quality. The gray scale conversion method is used to process the color ultrasound image and convert it into a single channel gray scale image. In the conversion process, the weighted average method (such as the commonly used weighted coefficients 0.299, 0.587, 0.114) is used to gray the RGB image to obtain the gray value of each pixel point. By calculating the gray value of each pixel point of the image, the gray distribution data of the image is obtained. Using image processing tools (such as the rgb2gray function of Matlab or the cv2.cvtColor method in the OpenCV library of Python), the ultrasound image is efficiently converted into a gray scale image, and the gray value information of each pixel is exported. After obtaining the gray value data of the venous ultrasound image, the gray value of the image is statistically analyzed. The standard deviation of the gray value is calculated using the gray histogram of the image, and the standard deviation is used as an indicator to measure the degree of change of the image gray scale distribution. The calculation process includes: counting the frequency of each gray value in the image, then calculating the mean value of the image gray value according to the frequency of each gray value; then, the difference square of each gray value and the mean value is calculated, and the weighted average is obtained to get the global gray value standard deviation. The standard deviation reflects the gray fluctuation in the image, and a higher standard deviation indicates a larger brightness change in the image, and a lower standard deviation indicates a more uniform image. This operation is calculated by the std() function in Matlab or the numpy.std() method in Python, and the result is a single value representing the global gray scale standard deviation of the image. Using the global gray value standard deviation data of the ultrasound image, the gray scale contrast of the image is further analyzed. The gray scale contrast is a measure of the brightness difference in the image, and is usually defined as the difference between the maximum and minimum gray values in the image. When calculating, the maximum and minimum gray values are found through the gray value matrix of the image. The formula for calculating the contrast is: contrast = maximum gray value - minimum gray value, or some weighted contrast indicators such as root mean square deviation (RMS) are used to measure the contrast change in the image. Based on the gray scale standard deviation of the image, a higher standard deviation value will produce a larger gray scale contrast. This operation uses the range() function of Matlab or the numpy.ptp() method of Python to calculate the gray scale contrast data of the venous ultrasound image. When analyzing the gradient change of the texture feature direction of the ultrasound image, the texture feature is extracted from the image, and the commonly used methods are gray level co-occurrence matrix (GLCM) or local binary pattern (LBP). The gradient change of the texture feature is calculated, and the texture change in different directions of the image is particularly concerned. The gradient change is calculated by applying the Sobel operator or gradient convolution to the texture feature image to obtain the gradient change of the image in the horizontal, vertical and diagonal directions. For the gradient of each direction, the change amplitude of the pixel gray value is calculated to analyze the directional variation characteristics of the image texture.The direction gradient change data of the texture features is calculated to further reveal the roughness of the blood vessel wall, the direction of blood flow, and the boundary changes with the surrounding tissues. This process is implemented through the gradient() function in Matlab or numpy.gradient() in Python to obtain the data of the direction gradient change of the texture features. By combining the direction gradient change data of the texture features with the grayscale value contrast data of the ultrasound image, the edge features of the venous blood vessels are further analyzed, and the edge region of the blood vessels is extracted based on the direction gradient change data of the texture features. For the blood vessel part in the ultrasound image, the texture features usually show obvious directional differences, while the edge part has a relatively sharp grayscale change. By fusing these two groups of data, an edge detection algorithm (such as Canny or Sobel algorithm) is used to accurately locate the edge position of the blood vessels. Subsequently, the sharpness, curve change, and smoothness of the blood vessel edge are analyzed by combining the edge position with the grayscale contrast. Morphological operations (such as dilation and erosion) are used to optimize the edge and remove small-scale noise while retaining the true blood vessel edge. This operation is performed through the edge() function in Matlab or cv2.Canny() in Python to obtain the edge feature data of the venous blood vessels.

[0123] Preferably, step S23 comprises the following steps:

[0124] Step S231 : performing image venous fine-grained analysis according to the ultrasound image venous feature data to obtain ultrasound image venous fine-grained data;

[0125] Step S232: performing image venous wall thickness detection based on the ultrasound image venous fine-grained data to obtain ultrasound image venous wall thickness data;

[0126] Step S233: performing image venous position measurement on the ultrasound image venous feature data to obtain ultrasound image venous position data;

[0127] Step S234: performing image venous depth detection according to the ultrasound image venous position data to obtain ultrasound image venous depth data;

[0128] Step S235: performing central venous catheter specification estimation based on the ultrasound image venous depth data to obtain central venous catheter specification data.

[0129] In the embodiments of the present application, when performing fine-grained analysis on the vein features in the ultrasound image, the vein region in the image is extracted, and the information of each pixel point of the image is further refined. High-precision segmentation algorithms are used to identify the edges and internal structure of the vein region, such as a convolutional neural network (CNN) model based on deep learning or a classic edge detection method (such as Canny edge detection). By performing pixel-by-pixel processing on the image at a small scale, the micro changes of the vein wall can be analyzed, such as the morphological fluctuations or micro lesions of the vein lumen. In image processing, a local feature extraction method (such as local binary pattern LBP) is used to perform texture analysis on the neighborhood around each pixel, and then fine-grained vein structure data is obtained. These fine-grained features will provide accurate regional information for subsequent steps such as vein wall thickness measurement and depth detection. This process can be implemented through functions such as regionprops() in Matlab or skimage.measure.label() in Python to perform fine segmentation and analysis and obtain fine-grained data of the vein. When using the obtained fine-grained vein data to detect the thickness of the vein wall, the measurement region of the vein wall needs to be defined. By performing high-precision segmentation on the image, the boundary line between the vein lumen and the surrounding tissue is identified, and a gradient algorithm based on gray value change (such as Sobel operator) is applied in this region to detect the thickness of the vein wall. The vein wall in the image usually shows obvious gray value change, and the boundary position of the vein wall is accurately determined by calculating the change rate of the gray value. The calculation method of the vein wall thickness is usually to convert the pixel unit to the actual length unit such as millimeter or centimeter according to the physical scale of the image. By measuring the wall thickness at different positions of the vein multiple times, the vein wall thickness data of the ultrasound image is obtained. This step uses the imgradient() function in Matlab or the cv2.Sobel() function in Python to complete the image gradient calculation, and combines the distance transformation method to obtain the accurate thickness value of the vein wall. According to the vein feature data in the ultrasound image, the position of the vein is measured from the vein region segmentation result. Through edge detection of the vein, the starting point and the ending point of the vein are accurately obtained, and the center axis position of the vein is calculated. Common algorithms include Canny operator based on edge detection or Snakes model based on active contour model to further refine the position of the vein. The position measurement of the vein needs to combine the resolution of the image to convert the image coordinate system to the actual physical coordinates. By measuring the longitudinal position of the vein, the position distribution of the vein in the image is determined, and this step combines morphological operations (such as erosion and dilation) to optimize the detection results and remove misidentified regions.The boundary data of the vein is extracted using the bwboundaries() function in Matlab or skimage.measure.regionprops() in Python to obtain the specific position coordinates of the vein in the image. When detecting the depth of the vein based on the position data of the vein in the ultrasound image, the position data of the vein in the image is used to determine the depth of the vein relative to the probe. Depth detection usually relies on the two-dimensional gray scale distribution of the ultrasound image, and the brightness value in the image is inversely proportional to the depth of the object. In the depth measurement process, the actual depth of the vein needs to be obtained by calibrating the depth scale of the image based on the propagation speed of the ultrasound wave and the reflection principle. In specific operation, the depth of the vein in the image is obtained by calculating the vertical distance from the position of the vein to the ultrasound probe. This step uses image calibration technology to convert the pixel value in the image to the actual depth value, and the commonly used method includes calibration by a reference object with known depth. Through the depth information of the image, the insertion depth and position of the central vein catheter are further inferred. In Matlab, the imdistline() tool is used to assist in measuring the depth, and in Python, scipy.ndimage.measurements.center_of_mass() can be used to calculate the depth. After obtaining the depth data of the vein, the specification of the central vein catheter is estimated. According to the depth information of the vein, combined with the diameter, shape and position of the vein, the specification of the central vein catheter is inferred. In this process, the size data of the vein, such as the inner diameter and outer diameter of the blood vessel, need to be obtained. Then, according to the depth information of the blood vessel, the suitable catheter specification at different depths is inferred. The estimation of the catheter specification is usually calculated based on empirical formulas or standard values in medical databases, combined with the depth and morphological information of the vein, to give the matching catheter specification. This process involves complex geometric models and biomechanical principles, and needs to consider factors such as the diameter, curvature and material properties of the catheter. By combining image depth and size information, interpolation or regression analysis methods are used to obtain the specification of the central vein catheter.

[0130] Preferably, step S3 comprises the following steps:

[0131] Step S31: collecting the central vein catheter mis-touch area data according to the central vein catheter displacement probability data;

[0132] Step S32: estimating the local thrombosis probability data of the ultrasound image according to the central vein catheter mis-touch area data and the case patient vein ultrasound image data;

[0133] Step S33: according to the local thrombosis probability data of the ultrasound image and the central venous catheter mis-touch region data, central venous catheter abnormal state analysis is performed to obtain central venous catheter abnormal state data.

[0134] In the embodiments of the present application, after obtaining the central venous catheter displacement probability data, the area of the mis-touch needs to be further analyzed for the displaced catheter. The mis-touch area refers to the position of the catheter after displacement, which is in contact with the surrounding blood vessel wall, heart wall or anatomical structure, thereby causing tissue damage. Based on the catheter displacement probability data, the image registration technology is used to accurately align the central venous catheter position with the venous anatomical structure. The image segmentation and registration algorithm is used to process the venous ultrasound image of the patient, and the actual position of the catheter is overlaid with the image data of the blood vessel wall. By calculating the offset and relative position of the catheter relative to the blood vessel wall, the potential mis-touch area is identified. This process can use the threshold-based image processing method combined with morphological operations to highlight the mis-touch area. In Matlab, the imoverlay() function superimposes the mis-touch area on the image, and skimage.measure.label() in Python can also be used to mark the position of the mis-touch area to obtain the mis-touch area data. After obtaining the contact area data of the central venous catheter, it is necessary to further evaluate whether these areas have the risk of local thrombosis. Thrombosis usually occurs at the site where the catheter contacts the vein wall or where the blood flow is slow. Therefore, by combining the position information of the catheter contact area and the hemodynamic characteristics in the ultrasound image, the risk of thrombosis can be estimated. Using blood flow velocity analysis algorithm, the blood flow pattern in the venous ultrasound image is evaluated, combined with the diameter of the blood vessel lumen and the pressure point of the catheter contact, to speculate the site of thrombosis. This analysis can use the method based on fluid mechanics simulation to calculate the area where the catheter contacts the blood vessel wall, combined with the local blood flow deceleration area, to evaluate the probability of thrombosis. Common image processing methods include gray level co-occurrence matrix (GLCM) analysis and texture analysis to extract local features of the blood vessel wall. In addition, physiological parameters (such as blood viscosity, blood flow rate, etc.) are combined with the blood flow pattern in the ultrasound image to quantify the risk of thrombosis. The regionprops() function of Matlab or the scipy.ndimage library of Python is used for local thrombus evaluation to obtain the local thrombosis probability data. Based on the obtained local thrombosis probability data and the central venous catheter mis-touch area data obtained in step S31, the analysis of the abnormal state of the central venous catheter is carried out, the two data are fused, and the abnormal risk of the catheter is comprehensively evaluated by weighted average method or logistic regression analysis combined with the mis-touch area and the thrombosis probability. Specifically, if the catheter is in the mis-touch area and the probability of local thrombosis is high, the catheter will malfunction or cause complications. This analysis can use statistical methods such as Bayesian network or multiple regression analysis to quantify the probability of abnormal state. Image data and hemodynamic data can be fused and analyzed by specific algorithms to obtain the abnormal state of the catheter under different conditions.Flow velocity analysis, vein wall morphological changes, and the relative position of the catheter and the vessel wall in the ultrasound image are used to further strengthen the evaluation. The fitlm() function in Matlab or the sklearn.linear_model.LogisticRegression model in Python is used to implement regression analysis and obtain catheter abnormal state data.

[0135] Preferably, step S32 comprises the following steps:

[0136] Step S321: Collect vein morphological changes in ultrasound images according to central venous catheter mis-touch area data and case patient vein ultrasound image data to obtain vein morphological changes in ultrasound images data;

[0137] Step S322: Detect vein blood vessel puncture according to vein morphological changes in ultrasound images data to obtain vein blood vessel puncture in ultrasound images data;

[0138] Step S323: Detect vein blood vessel injury according to vein blood vessel puncture in ultrasound images data to obtain vein blood vessel injury in ultrasound images data;

[0139] Step S324: Analyze blood vessel spasm according to vein blood vessel injury in ultrasound images data to obtain blood vessel spasm probability in ultrasound images data;

[0140] Step S325: Estimate vein blood flow obstruction according to central venous catheter mis-touch area data and vein blood vessel injury in ultrasound images data to obtain vein blood flow obstruction data;

[0141] Step S326: Estimate local thrombosis probability based on blood vessel spasm probability in ultrasound images data and vein blood flow obstruction data to obtain local thrombosis probability in ultrasound images data.

[0142] In the embodiments of the present application, after obtaining the central venous catheter mis-touch area data, it is necessary to further evaluate the influence of catheter displacement on the shape of the blood vessel. The change of the vein shape is judged by the change of the blood vessel contour, the shape of the inner cavity and the blood vessel wall in the ultrasound image. In this step, the ultrasound image of the patient's vein is processed, the boundary of the blood vessel is identified, and compared with the normal vein shape. Image processing includes edge detection, morphological operation (such as dilation, erosion, etc.), so as to highlight the shape characteristics of the blood vessel. In order to obtain the change of the vein shape, by comparing the vein images before and after the catheter placement, the image difference analysis method is used to determine the change of the blood vessel after the catheter mis-touch. The specific operation uses the edge detection algorithm (such as Canny()) and findContours() in the OpenCV library of Python to extract the vein shape data, and at the same time, morphological transformation is carried out to obtain the data set of the change of the blood vessel shape. Based on the ultrasound image vein shape change data, further detection is carried out on whether the vein blood vessel is punctured. In this step, the puncture detection is mainly carried out by judging whether there are signs of damage or rupture of the vein wall, using the blood vessel edge information extracted from the vein shape change data, combining the intensity change characteristics in the ultrasound image, detecting whether there are abnormal blood vessel wall penetration or puncture marks. Blood vessel puncture is usually accompanied by local blood vessel wall rupture or penetration, and is shown in the image as a region with obvious contrast difference with the surrounding blood vessel tissue. This detection uses mutation analysis of image gray value or change of image texture features (such as gray level co-occurrence matrix, local binary pattern) to carry out. Through the analysis of these features, combined with the time series data of the ultrasound image, it is judged whether the blood vessel is punctured. After detecting the blood vessel puncture, further analysis is carried out on whether there is blood vessel damage. Blood vessel damage is usually accompanied by blood vessel wall rupture, bleeding or blood flow abnormalities. In this process, the blood vessel region in the ultrasound image needs to be further analyzed in detail, and the specific position of the vein injury is located through the blood vessel puncture region data. Then, using the high frequency detail information (such as tissue echo, gray value change, etc.) of the ultrasound image, the tissue characteristics of the damage site are analyzed, the abnormal shape of the blood vessel wall, the bleeding condition and the signs of blood flow interruption are identified. Through the analysis of local gray value difference, echo intensity change or high frequency component of the image, the specific shape of the damage region is detected. Common processing methods include using region growing algorithm, image segmentation technology, etc. in image processing. The regionprops() function in Matlab can be used for region recognition, and the region data of the damage is extracted. Blood vessel injury is usually accompanied by the occurrence of blood vessel spasm, especially during the operation of the central venous catheter, the external pressure and the change of blood flow cause the contraction reaction of the blood vessel. By analyzing the shape change of the blood vessel in the ultrasound image, it can be detected whether the blood vessel appears spasm.Specifically, vasospasm can cause the lumen of the blood vessel to narrow, which is manifested as the inward contraction of the vessel wall in the image, causing poor blood flow. Based on the venous vessel injury data, further analysis is performed using the morphological features of the veins in the image (such as vessel diameter, wall thickness, etc.) to identify whether the contraction phenomenon occurs. Image contrast analysis or changes in arterial and venous resolution are used to identify vasospasm. In Matlab or Python, the regionprops() function is used to quantitatively analyze the changes in vessel diameter, and combined with blood flow velocity data (such as Doppler signals) to further confirm whether vasospasm occurs. By quantifying these changes, the probability data of vasospasm is obtained. Venous blood flow obstruction is usually caused by catheter mis-touch, blood vessel injury or vasospasm. In this step, the central venous catheter mis-touch area data is combined with the venous vessel injury data to evaluate the obstruction of venous blood flow. By analyzing the area where the catheter contacts the vessel wall and the location of the vessel injury, it is inferred whether the blood flow is affected. The obstruction of venous blood flow is usually accompanied by a decrease in local blood flow velocity or a reverse flow phenomenon. In order to estimate the severity of blood flow obstruction, a blood flow simulation method based on image processing is adopted, combined with the geometric structure of the blood vessel, the hemodynamic characteristics and the location of the vessel wall injury, and the fluid mechanics model is used to calculate the resistance and flow changes of the blood flow. Through the flow velocity information in the ultrasound image (such as Doppler image) and the change of blood vessel diameter, the degree of blood flow obstruction is quantified. Numerical solution methods such as scipy.integrate.solve_ivp() in Python are used to simulate hemodynamics to obtain quantitative data of blood flow obstruction. After obtaining the probability data of vasospasm and the data of venous blood flow obstruction, the probability of thrombus formation is further predicted. The formation of thrombus is usually closely related to the slowing down of blood flow and the injury of blood vessel lumen. Combined with the area of vasospasm and blood flow obstruction, a physiological model is used to predict the risk of local thrombus formation. Based on image texture features, hemodynamic analysis and local pressure changes, combined with the morphological data of venous vessels, the probability of thrombus formation is obtained. In this process, regression analysis or Bayesian network and other statistical methods are used to integrate different data sources to further improve the prediction accuracy of thrombus formation risk. Comprehensive analysis is conducted by combining the blood flow velocity changes in the ultrasound image, the changes in the shape of the blood vessel, and the model of blood rheology.

[0143] Preferably, step S33 comprises the following steps:

[0144] Step S331: Perform venous vessel thrombus enlargement analysis according to the local thrombus formation probability data of the ultrasound image to obtain venous vessel thrombus enlargement data;

[0145] Step S332: Perform venous vessel occlusion detection based on the venous vessel thrombus enlargement data to obtain venous vessel occlusion data;

[0146] Step S333: detecting the central venous catheter function loss according to the venous vessel occlusion data and the venous vessel thrombus enlargement data, to obtain central venous catheter function loss data;

[0147] Step S334: estimating the venous infection probability according to the ultrasonic image local thrombosis probability data and the central venous catheter mis-touch region data, to obtain venous infection probability data;

[0148] Step S335: analyzing the central venous catheter abnormal state based on the venous infection probability data and the central venous catheter function loss data, to obtain central venous catheter abnormal state data.

[0149] In the embodiments of the present application, the thrombus is analyzed for enlargement by the local thrombus formation probability data in the ultrasound image. Thrombus enlargement is usually accompanied by compression of the local blood vessel or poor blood flow, leading to expansion of the thrombus volume. This process analyzes the local thrombus formation probability data, uses the thrombus shape change and local blood flow velocity change information in the image to infer whether the thrombus is expanding. The basis for judging thrombus enlargement includes changes in the blood vessel lumen, the fuzzy or irregular shape of the thrombus edge, and the gray value changes in the thrombus area in the ultrasound image. Accurate thrombus enlargement data is obtained, the thrombus area is extracted using image segmentation technology, and the degree of thrombus enlargement is quantified by comparing image data at different times. The consequence of thrombus enlargement leads to complete occlusion of the blood vessel. Based on the thrombus enlargement data of the venous vessel, the blood vessel occlusion is detected. Occlusion usually manifests as complete occlusion of the blood vessel lumen, leading to a complete stop or severe slowing of blood flow. The diameter information of the blood vessel lumen in the image and the hemodynamic data (such as Doppler ultrasound measurement data) are used to identify whether there is complete occlusion. The detection method includes calculating the stenosis of the blood vessel and the flow velocity change, and if the blood vessel lumen is completely closed, it indicates that the blood vessel is occluded. In order to accurately extract the occluded area, the image segmentation technology is further used to analyze the morphological changes of the blood vessel lumen area and compare it with the normal blood vessel morphology to determine whether there is a blood flow interruption. The cv2.findContours() function in Matlab or Python can be used to extract the blood vessel lumen area, and further combined with the blood flow velocity information to analyze whether there is an occlusion. Through the blood flow velocity, blood vessel lumen morphology, etc. data, the severity of the blood vessel occlusion is quantified. Central venous catheter dysfunction is usually caused by blood flow obstruction due to blood vessel occlusion or thrombus enlargement. In this step, the central venous catheter dysfunction degree is evaluated by the venous vessel occlusion data and the thrombus enlargement data, the blood vessel occlusion degree is analyzed, and the catheter position and hemodynamic model are combined to determine whether the catheter function is limited. Thrombus enlargement and occlusion cause blockage of the fluid in the catheter, thereby affecting its normal function. By comparing the blood flow data (such as blood flow velocity, pressure distribution, etc.) in the blood vessel with the expected function of the catheter, it is evaluated whether the catheter has a dysfunction. This analysis combines the changes in blood vessel resistance, blood flow velocity, and the geometry of the catheter, and thrombus formation and central venous catheter miscontact area both cause venous infection. In this step, the local thrombus formation probability data and the catheter miscontact area data in the ultrasound image are combined with the bacterial infection risk of the blood vessel lumen to estimate the probability of venous infection. Venous infection usually occurs at the contact between the catheter and the blood vessel wall, especially in the thrombus formation site, and the catheter miscontact area is analyzed to evaluate the local infection risk. Secondly, based on the vein wall morphology, thrombus formation, and local flow data in the image, the bacterial proliferation and infection probability in these areas are inferred. Quantitative analysis is performed by image feature extraction combined with a physiological model of microbial infection.The scipy library of Python can be used for data analysis, combined with Bayesian network or regression model to calculate the probability of venous infection, and the estimated value of infection is obtained according to the combined data of catheter position, thrombus and mis-touch area. The abnormal state of central venous catheter is usually manifested as infection, blood flow obstruction, function loss and other problems. In this step, the comprehensive analysis of catheter abnormal state is carried out by combining the venous infection probability data and the central venous catheter function loss data. Venous infection and function loss are usually manifestations of catheter abnormalities, which affect the overall function of the catheter. By integrating the data of venous infection and catheter function loss, it is analyzed whether the catheter is in an abnormal state, and the severity of the abnormality is evaluated. By constructing a comprehensive risk assessment model, considering multiple factors such as infection, thrombus, catheter displacement, etc., the comprehensive abnormal state of the catheter is obtained.

[0150] Preferably, step S4 comprises the following steps:

[0151] Step S41: detecting venous function impairment based on central venous catheter abnormal state data to obtain venous function impairment data;

[0152] Step S42: estimating venous catheter complications according to the venous function impairment data and the central venous catheter abnormal state data to obtain central venous catheter complication data;

[0153] Step S43: performing central venous catheter risk assessment based on the venous catheter complication data and the venous function impairment data to obtain central venous catheter risk data;

[0154] Step S44: constructing a central venous catheter intelligent decision-making model based on the central venous catheter risk data to obtain a central venous catheter intelligent decision-making model.

[0155] As an example of the present application, reference is made to Fig. 1, which shows the steps S4 in this example comprising: Figure 3

[0156] Step S41: detecting venous function impairment based on central venous catheter abnormal state data to obtain venous function impairment data;

[0157] ​In the embodiment of the present application, the detection of venous function impairment is based on the analysis of known central venous catheter abnormal state data, mainly through the physiological model and hemodynamic model of venous function. In this step, according to the abnormal state data of the central venous catheter, such as catheter position deviation, thrombus enlargement, local blood flow obstruction, etc., the functional changes of the venous system are analyzed. By comparing the normal venous function data and the abnormal state data, the changes of the elasticity of the blood vessel, the blood flow velocity, the blood vessel pressure, etc. are identified, and then it is judged whether the venous function is impaired. Venous function impairment is usually accompanied by changes in hemodynamics, including phenomena such as blood flow velocity decrease, abnormal pressure increase or decrease, etc. By using ultrasonic Doppler blood flow velocity measurement, pressure sensor data, combined with image feature analysis technology (such as blood vessel lumen analysis based on threshold segmentation), it is determined whether the vein has functional impairment.

[0158] Step S42: estimating the central venous catheter complication according to the venous function impairment data and the central venous catheter abnormal state data, to obtain central venous catheter complication data;

[0159] In the embodiment of the present application, the estimation of the venous catheter complication is based on the combined analysis of the venous function impairment data and the catheter abnormal state data. In this step, by analyzing the type and degree of venous function impairment, combined with the state of the catheter itself, such as position shift, infection, thrombus, etc., the complications caused are identified. For example, thrombus formation or catheter mis-touch leads to venipuncture, thrombus migration or local infection, and then causes blood flow obstruction or loss of catheter function, by analyzing the dynamic changes of venous blood flow, combined with the information of the shape, elasticity, thickness, etc. of the blood vessel wall, it is judged whether there is a potential complication of the catheter. For the venous function impairment data, it is usually necessary to track the use state and complications of the catheter in real time through a dynamic monitoring system (such as hemodynamic simulation, blood viscosity model, etc.). In addition, machine learning algorithms (such as support vector machine, decision tree) are used to quantitatively estimate the complications, and the model input data includes blood flow velocity, catheter position, thrombus size, infection risk, etc.

[0160] Step S43: based on the venous catheter complication data and the venous function impairment data, the central venous catheter risk assessment is performed, to obtain central venous catheter risk data;

[0161] In the embodiments of the present application, the risk assessment of central venous catheters combines the analysis results of venous function impairment and catheter complications, mainly relies on a risk assessment model, calculates the comprehensive risk of catheter abnormalities by evaluating the weights of different factors. In this step, the venous function impairment data and complication data are scored by weighting. According to the severity of each type of impairment (such as blood flow abnormalities, catheter blockage, infection, etc.), a weight is assigned to each indicator. Generally, these weights are derived from clinical research and empirical data. Then, the overall risk of central venous catheters is calculated by weighted scoring. The evaluation model is constructed by regression analysis, decision tree model or neural network. Common regression analysis methods include Logistic regression or Cox regression model, which models the influencing factors of venous function impairment and complications to obtain catheter risk data. In addition, based on historical case data and medical records, a deep learning model such as LSTM (Long Short-Term Memory Network) is trained to model time series data and predict the risk faced by the catheter in the future.

[0162] Step S44: Constructing a central venous catheter intelligent decision-making model based on the central venous catheter risk data to obtain a central venous catheter intelligent decision-making model.

[0163] In the embodiments of the present application, this step involves inputting the central venous catheter risk data obtained in the previous step into the intelligent decision-making model for comprehensive judgment and formulation. The central venous catheter intelligent decision-making model is based on the input risk data, combines multi-source data (such as patient signs, venous ultrasound images, hemodynamic data, etc.), and generates catheter use and intervention schemes through multi-dimensional evaluation and decision-making algorithms. The construction of the intelligent decision-making model usually uses machine learning or optimization algorithms, such as decision trees, support vector machines, random forests, genetic algorithms, reinforcement learning, etc., through data preprocessing (such as standardization, missing value filling, etc.) to ensure the quality of the input data. Then, a decision tree model or a random forest model is constructed to predict catheter risks by training historical data and give intervention suggestions according to different risk levels. At the same time, the central venous catheter intelligent decision-making model is trained according to the training historical data and the catheter risk, so that the central venous catheter intelligent decision-making model gives suggestions on catheter specifications, puncture site selection according to the patient's treatment plan.

[0164] The above is only a specific embodiment of the present application, which enables those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a central venous catheter intelligent decision model based on multi-source data, characterized in that, The method comprises the following steps: Step S1: acquiring central venous catheter case data; extracting case patient signs according to the central venous catheter case data to obtain case patient sign data; and analyzing central venous catheter structure characteristics according to the central venous catheter case data to obtain central venous catheter structure characteristic data; Step S2: collecting case patient venous ultrasound images according to the case patient sign data to obtain case patient venous ultrasound image data; estimating the specifications of the central venous catheter according to the case patient venous ultrasound image data and the central venous catheter structure characteristic data to obtain central venous catheter specification data; and calculating the catheter position displacement probability according to the central venous catheter specification data and the case patient venous ultrasound image data to obtain central venous catheter displacement probability data; Step S3: collecting central venous catheter mis-touch regions according to the central venous catheter displacement probability data to obtain central venous catheter mis-touch region data; and analyzing the abnormal state of the central venous catheter according to the central venous catheter mis-touch region data and the case patient venous ultrasound image data to obtain central venous catheter abnormal state data; Step S4: performing central venous catheter risk assessment based on the central venous catheter abnormal state data to obtain central venous catheter risk data; and constructing a central venous catheter intelligent decision-making model based on the central venous catheter risk data to obtain the central venous catheter intelligent decision-making model.

2. The method of claim 1, wherein the method is used for constructing a central venous catheter intelligent decision model based on multi-source data. Step S1 comprises the following steps: Step S11: acquiring central venous catheter case data; Step S12: performing case desensitization processing according to the central venous catheter case data to obtain central venous catheter case desensitization data; Step S13: extracting case patient signs according to the central venous catheter case desensitization data to obtain case patient sign data; Step S14: analyzing central venous catheter structure characteristics according to the central venous catheter case desensitization data to obtain central venous catheter structure characteristic data.

3. The method of claim 1, wherein the method further comprises: determining a plurality of features of the central venous catheter based on the multi-source data; and determining a plurality of weights of the features based on the multi-source data. Step S2 comprises the following steps: Step S21: collecting case patient venous ultrasound images according to the case patient sign data to obtain case patient venous ultrasound image data; Step S22: analyzing venous characteristics according to the case patient venous ultrasound image data to obtain ultrasound image venous characteristic data; Step S23: estimating the specifications of the central venous catheter based on the ultrasound image venous characteristic data and the central venous catheter structure characteristic data to obtain central venous catheter specification data; Step S24: calculating the catheter position displacement probability according to the central venous catheter specification data and the ultrasound image venous characteristic data to obtain central venous catheter displacement probability data.

4. The method of claim 3, wherein the method further comprises: Step S22 comprises the following steps: Step S221: collecting image texture feature data according to the case patient venous ultrasound image data; Step S222: analyzing venous blood vessel edge features based on the ultrasound image texture feature data and the case patient venous ultrasound image data to obtain venous blood vessel edge feature data; Step S223: analyzing contour features according to the venous blood vessel edge feature data and the ultrasound image texture feature data to obtain venous blood vessel contour data; Step S224: based on the venous vessel contour data and the venous vessel edge feature data, venous feature analysis is performed to obtain the ultrasound image venous feature data.

5. The method of claim 4, wherein the method further comprises: Step S222 includes the following steps: According to the case patient venous ultrasound image data, the ultrasound image gray value is calculated, and the venous ultrasound image gray value data is obtained; According to the venous ultrasound image gray value data, the ultrasound image global gray value standard deviation is calculated, and the ultrasound image global gray value standard deviation data is obtained; According to the ultrasound image global gray value standard deviation data, the ultrasound image gray value contrast analysis is performed, and the ultrasound image gray value contrast data is obtained; According to the ultrasound image texture feature data, the texture feature direction gradient change calculation is performed, and the image texture feature direction gradient change data is obtained; Based on the image texture feature direction gradient change data and the ultrasound image gray value contrast data, the venous vessel edge feature analysis is performed, and the venous vessel edge feature data is obtained.

6. The method of claim 3, wherein the method further comprises: Step S23 includes the following steps: Step S231: according to the ultrasound image venous feature data, the image venous fine-grained analysis is performed, and the ultrasound image venous fine-grained data is obtained; Step S232: based on the ultrasound image venous fine-grained data, the image venous wall thickness detection is performed, and the ultrasound image venous wall thickness data is obtained; Step S233: the ultrasound image venous feature data is measured, and the ultrasound image venous position data is obtained; Step S234: according to the ultrasound image venous position data, the image venous depth detection is performed, and the ultrasound image venous depth data is obtained; Step S235: based on the ultrasound image venous depth data, the central venous catheter structure feature data is estimated, and the central venous catheter specification data is obtained.

7. The method of claim 1, wherein the method further comprises: determining a plurality of features of the central venous catheter based on the multi-source data; and determining a plurality of weights of the features based on the multi-source data. Step S3 includes the following steps: Step S31: according to the central venous catheter displacement probability data, the central venous catheter mis-touch area collection is performed, and the central venous catheter mis-touch area data is obtained; Step S32: according to the central venous catheter mis-touch area data and the case patient venous ultrasound image data, the local thrombosis probability estimation is performed, and the ultrasound image local thrombosis probability data is obtained; Step S33: according to the ultrasound image local thrombosis probability data and the central venous catheter mis-touch area data, the central venous catheter abnormal state analysis is performed, and the central venous catheter abnormal state data is obtained.

8. The method of claim 7, wherein the method is characterized by, Step S32 includes the following steps: Step S321: according to the central venous catheter mis-touch area data and the case patient venous ultrasound image data, the ultrasound image venous morphological change collection is performed, and the ultrasound image venous morphological change data is obtained; Step S322: according to the ultrasound image venous morphological change data, the venous vessel puncture detection is performed, and the ultrasound image venous vessel puncture data is obtained; Step S323: according to the ultrasound image venous vessel puncture data, the venous vessel injury detection is performed, and the ultrasound image venous vessel injury data is obtained; Step S324: according to the ultrasound image venous vessel injury data, the vascular spasm analysis is performed, and the ultrasound image vascular spasm probability data is obtained; Step S325: according to the central venous catheter error touch area data and the ultrasonic image vein vessel injury data, vein blood flow obstruction estimation is performed, and vein blood flow obstruction data is obtained; Step S326: based on the ultrasonic image blood vessel spasm probability data and the vein blood flow obstruction data, local thrombosis probability estimation is performed, and ultrasonic image local thrombosis probability data is obtained.

9. The method of claim 7, wherein the method further comprises: Step S33 includes the following steps: Step S331: according to the ultrasonic image local thrombosis probability data, vein vessel thrombus enlargement analysis is performed, and vein vessel thrombus enlargement data is obtained; Step S332: based on the vein vessel thrombus enlargement data, vein vessel occlusion detection is performed, and vein vessel occlusion data is obtained; Step S333: according to the vein vessel occlusion data and the vein vessel thrombus enlargement data, central venous catheter function loss detection is performed, and central venous catheter function loss data is obtained; Step S334: according to the ultrasonic image local thrombosis probability data and the central venous catheter error touch area data, vein infection probability estimation is performed, and vein infection probability data is obtained; Step S335: based on the vein infection probability data and the central venous catheter function loss data, central venous catheter abnormal state analysis is performed, and central venous catheter abnormal state data is obtained.

10. The method of claim 1, wherein the method further comprises: determining a plurality of features of the patient based on the plurality of data sources; and determining a plurality of weights of the plurality of features based on the plurality of data sources. Step S4 includes the following steps: Step S41: based on the central venous catheter abnormal state data, vein function damage detection is performed, and vein function damage data is obtained; Step S42: according to the vein function damage data and the central venous catheter abnormal state data, central venous catheter complication estimation is performed, and central venous catheter complication data is obtained; Step S43: based on the central venous catheter complication data and the vein function damage data, central venous catheter risk assessment is performed, and central venous catheter risk data is obtained; Step S44: based on the central venous catheter risk data, a central venous catheter intelligent decision-making model is constructed, and a central venous catheter intelligent decision-making model is obtained.

Images