156 results about "Vena porta" patented technology

The invention relates to the technical field of image analysis, in particular to a three-dimensional hepatic duct, hepatic artery, portal vein and hepatic vein image segmentation method and system. The method comprises the following steps: acquiring multi-phase medical image data of a liver region; performing region identification on the image data, and extracting hepatic duct, hepatic artery, portal vein and hepatic vein regions; executing structure tracking on the hepatic duct area to obtain a hepatic duct positioning result; carrying out improved contrast extraction on the hepatic artery region to obtain hepatic artery positioning data; performing directional segmentation analysis on the portal vein region to obtain a portal vein positioning result; performing branch angle extraction on the hepatic vein region in combination with the portal vein direction to obtain hepatic vein positioning data; taking the hepatic artery as a reference structure, performing elastic registration on each blood vessel structure, and generating registration data under a unified coordinate; and completing three-dimensional visualization based on a registration result. According to the method, multi-stage image joint segmentation and registration are realized, and the method has the technical advantage of high structural recognition integrity.

The invention discloses a 4D-Flow-CFD simulation model construction method and system for quantitative evaluation of cirrhosis portal vein hydrodynamics, and belongs to the technical field of medical image processing, the construction method comprises the following steps: step 1, placing a patient on a scanner, and sequentially performing structural image and 4D-Flow MRI (four-dimensional blood flow magnetic resonance) scanning; 2, integrating the 4D-Flow MRI and CFD (computational fluid dynamics) technology, performing post-processing on a 4D-Flow MRI image by adopting Circle software, and screening out hemodynamic characteristic parameters capable of representing the liver cirrhosis process; and 3, constructing a liver portal vein system finite element model by using the screened haemodynamic characteristic parameters so as to establish a quantitative evaluation model of portal vein haemodynamics of the patients with cirrhosis, so that a new haemodynamic detection method is explored for clinical systematic evaluation of cirrhosis, and a new thought is provided for accurate evaluation of cirrhosis.

The invention relates to the technical field of image feature recognition, in particular to an auxiliary method for sketching adjacent regions before hepatic segment resection. According to the method, a CT image under each layer of the liver is analyzed, and the respiration influence degree is quantified by using the shape change and area change of a blood vessel region. According to the method, matching analysis is carried out on target blood vessel areas between adjacent layers, then change trend characteristics of respiration influence degrees are determined, all target blood vessel areas are traversed in a mode of matching and then combining, and an optimal combination mode for representing portal veins can be determined based on the change trend characteristics; therefore, the blood vessel region represented by the portal vein is screened out, and the effective assistance of the sketching of the adjacent region can be realized through three-dimensional reconstruction. According to the method, the accurate portal vein region is screened out and three-dimensional reconstruction is carried out through correlation analysis on the blood vessel regions in the CT images of different layers, and sketching of adjacent regions can be effectively assisted.

The invention provides a fracture repair and topology analysis method for a blood vessel segmentation mask in a liver CT image, and the method specifically comprises the steps: 1) carrying out the skeletonization of the liver blood vessel segmentation mask, and extracting a blood vessel skeleton, a skeleton end point, and a blood vessel radius rough value at a skeleton point; 2) solving a connected blood vessel center line (1) for the purpose of fracture repair without considering removal of wrong adhesion of hepatic veins and portal veins; 3) considering possible wrong adhesion for removing the hepatic vein and the portal vein, setting a stricter connection condition to obtain a broken blood vessel center line, and solving a broken blood vessel center line (2) in which the hepatic vein and the portal vein are obviously distinguished by taking classification as a purpose; and 4) combining the center lines (1) and (2), reconstructing missing blood vessel segments in the original blood vessel segmentation mask, and combining the reconstructed blood vessel segments into the original segmentation mask to obtain a final blood vessel mask and center line topological structure after fracture repair.

The invention discloses a postoperative portal vein pressure prediction system, which relates to the technical field of portal vein pressure prediction, and is characterized in that three-dimensional structure regions of interest of the liver and the spleen are segmented based on preoperative and postoperative abdomen enhanced CT vein phase image data of a patient; image omics features including first-order statistical features, texture features and shape features are extracted from the segmented regions of interest of the three-dimensional structures of the liver and the spleen, and core feature screening is carried out on the extracted features; integrating the screened core features with clinical hemodynamic parameters and surgical parameters to construct a multi-modal prediction model of the portal vein pressure gradient; and predicting and outputting a portal vein pressure gradient predicted value by using the multi-modal prediction model of the portal vein pressure gradient, and determining a postoperative portal vein pressure gradient risk grade of the patient. The portal vein pressure gradient prediction method solves the problems of insufficient dynamic evaluation capability, multi-modal information integration and risk layering application in the prior art, and realizes non-invasive and accurate prediction of the portal vein pressure gradient.

The invention provides a portal vein pressure adjusting support and a production method thereof, and belongs to the technical field of medical instruments.The adjusting support comprises a cylindrical support body, the support body is provided with a first section and a second section which are integrally arranged, and the side wall of the first section is provided with a film-shaped component used for sealing and isolating the outer side and an inner cavity of the support body; a pressure adjusting valve is arranged in an inner cavity of the first section, is opened for blood flow to pass through when the pressure in the portal vein is larger than a set threshold value, and is closed to cut off the blood flow when the pressure in the portal vein is smaller than the set threshold value; the stent can automatically adjust the opening and closing state of the shunt channel according to the portal vein real-time pressure, and ineffective shunt is reduced to the maximum extent while it is ensured that the portal vein pressure is reduced to a safe range, so that the occurrence risk of hepatic encephalopathy is reduced.

The invention discloses a clinical prediction model for the probability of occurrence of spleen hyperfunction of a hepatolenticular degeneration patient and a design method, and belongs to the technical field of medical information. The clinical prediction model comprises an input module for receiving patient parameter indexes consisting of an aspartate aminotransferase and platelet ratio index (APRI), a hepatic fibrosis 4 index (FIB-4), a liver hardness value (LSM) and a portal vein diameter (PVD); the processing module is connected with the input module and calculates the prediction probability of the patient according to the model; and the output module is connected with the processing module and outputs the risk probability of occurrence of the hypersplenic function of the hepatolenticular degeneration patient obtained by the clinical prediction model. The clinical prediction model established by the invention has good distinction degree, calibration degree and clinical practicability, has relatively high accuracy, and can be conveniently used for early recognition and risk prediction of hypersplenic function of hepatolenticular degeneration patients.

The invention discloses a monitoring method and system based on portal vein blood pressure data, and relates to the technical field of data monitoring. A lumped parameter model is constructed according to postoperative clinical data, and parameters related to the lumped parameter model are optimized and calibrated; performing steady-state simulation by using the optimized and calibrated lumped parameter model, and calculating portal vein blood pressure and other related hemodynamic parameters; the portal vein blood pressure is monitored in real time on the basis of the lumped parameter model application condition determined through steady state simulation, and the disease change and the treatment effect of the patient are evaluated; whether the monitoring result is abnormal or not is judged according to the monitoring result and patient personalized parameters, and then clinical feedback and suggestions are provided; a patient personalized updating and adjusting period is set based on the patient personalized parameters, and the lumped parameter model is updated and optimized based on the patient personalized updating and adjusting period. Continuous monitoring and personalized treatment of the condition of the patient are achieved, and the diagnosis accuracy and the treatment effect are remarkably improved.

A perfusion loop assembly for ex vivo liver perfusion includes a pump providing perfusion fluid through a line branching at a branching point into a first branch line and a second branch line. The first branch line provides a first portion of the perfusion fluid to the hepatic artery of the liver, the first branch line coupled with a gas exchanger, where the first branch line includes a flow rate sensor and / or a pressure sensor. The second branch line provides a second portion of the perfusion fluid to the portal vein of the liver; the second branch line includes a valve for controlling flow of perfusion fluid into the portal vein. The second branch line includes a flow rate sensor and / or a pressure sensor. A liver chamber assembly holds the liver ex vivo, and an outlet line for the perfusion fluid connects the liver chamber assembly and the pump.

The present invention relates to an ex vivo liver perfusion system, a joint pipe assembly and a connection method thereof, wherein the joint pipe assembly comprises: an intubation seat and a docking seat. The intubation seat is provided with a first compression end face, and a through hole is located in the middle of the first compression end face. The docking seat is provided with a docking hole, and the docking hole and the through hole are arranged correspondingly. The docking seat is also provided with a second compression end face. The docking hole is located in the middle of the second compression end face, and the second compression end face is compressed and matched with the first compression end face. The docking seat is also provided with a connecting pipe connected to the docking hole, and the connecting pipe is used to be connected to a consumable pipeline. In this way, the inner wall of the portal vein or the proper hepatic artery is not contacted during the perfusion process, thus avoiding damage to the endothelial tissue of the portal vein or the proper hepatic artery, and no suture operation is required, which greatly improves the work efficiency of assembly. In addition, the joint pipe assembly is firmly combined with the artery or portal vein of the ex vivo liver, and there is no risk of falling off during the perfusion process.

The present invention provides a liver hypothermic machine perfusion preservation device and a method. The device includes a perfusion loop and a controller, wherein the perfusion loop includes a main pump, a main reservoir, an auxiliary pump and an auxiliary reservoir. The main pump is connected to an inlet of the main reservoir and configured to pump a perfusate from an organ cassette into the main reservoir; an outlet of the main reservoir is connected to a portal vein and a hepatic artery of a liver, and the main reservoir is configured to supply the perfusate to the portal vein and the hepatic artery; and the auxiliary reservoir and the auxiliary pump provide an additional hepatic artery perfusion flow path. According to the present invention, for a poor internal environment of the hepatic artery, the main reservoir and the auxiliary reservoir cooperate with each other to carry out perfusion, so that the perfusion efficiency and effect of the hepatic artery can be improved. The device according to the present invention also has a molecular-sieve oxygen generation function, and a free adjustment range of oxygen concentration from 21% to 90% can be achieved by means of combined control of a molecular-sieve oxygen generation channel and an air channel, thus providing a more flexible solution for the perfusion.

The invention belongs to the technical field of medical diagnosis, and discloses a marker for evaluating the curative effect of treating hepatocellular carcinoma by combining TACE with Czodili monoclonal antibody and bevacizumab, and a construction method and application of an FAAP prognostic scoring model. The invention develops and verifies an FAAP prognostic scoring system, and a novel composite model integrates four indexes of fibrin degradation product / cholinesterase ratio * 1000 (FCR), aspartate aminotransferase (AST), alpha fetoprotein (AFP) and portal vein tumor thrombus (PVTT), and is used for predicting the survival condition of uHCC patients subjected to TACE-Czodimab-bevacizumab triple therapy. The tool has an instant clinical value, can optimize patient selection, guides adaptive treatment upgrading, and provides a standardized curative effect evaluation basis for a clinical test for exploring a TACE-immunotherapy-anti-angiogenesis combined scheme.

The invention provides a portal vein right atrium circulating tube under an endoscope, and belongs to the field of medical instruments. Comprising a tube body and two insertion tubes, the two insertion tubes are fixedly connected to the two ends of the tube body respectively, the tube body is straight and made of hard materials, and the two insertion tubes can be bent and made of elastic materials; when the device needs to be used, due to the fact that the tube body and the two insertion tubes are straight when not used, the two insertion tubes are inserted into the portal vein or the right atrium respectively, after insertion is completed, the annular plate and the connecting plate are driven by the driving assembly to slide on the tube body, and the connecting plate drives the guide wire to slide in the guide groove; the guide wire firstly slides into the guide groove in the tube body from the guide groove in the insertion tube and then penetrates out of the tube body from the guide groove in the tube body, when the guide wire completely leaves the guide groove in the insertion tube, the insertion tube rebounds and bends, the tube body is limited, the insertion tube is prevented from being inserted into the portal vein or the right atrium to fall off, and therefore the use effect is improved.

The present invention relates to a dual-input four-chamber model for the evaluation of dynamic PET / CT imaging of hepatocellular carcinoma. The present invention collects HCC dynamic PET / CT data for pathological diagnosis, including 5 minutes of dynamic data supplemented by 1 minute of static data 60 minutes after injection and a single 5 minutes of dynamic data; secondly, the tracer time-radioactivity curve data of normal liver tissue, hepatic artery, portal vein, and hepatocellular carcinoma are extracted from the PET / CT images; then the gravitational search algorithm is used to solve the kinetic parameters of the liver 4-chamber model; finally, the correlation between the kinetic parameters K1~k3 of the two time groups and the difference in the tracer kinetic parameters of the hepatocellular carcinoma tissue image and the surrounding healthy liver tissue image are analyzed and compared, and compared with the commonly used liver 3-chamber model kinetic parameters, the present invention can better characterize the metabolism of HCC and normal liver tissue, and has certain value for clinical auxiliary diagnosis of HCC.

The present application relates to a kind of stent delivery device and stent delivery system.The stent delivery device includes inner tube, outer tube, limiting sleeve and release sleeve;Limiting sleeve, release sleeve are sleeved on the outside of inner tube, outer tube is sleeved on the outside of limiting sleeve, release sleeve, limiting sleeve is located at the distal end side of release sleeve;The distal end of release sleeve is provided with restraint wire, and the bare section of stent is restrained outside limiting sleeve by restraint wire, release sleeve can move towards the proximal end of outer tube under the action of external force until the bare section is released by restraint wire.The stent delivery device, when releasing the film section of stent in advance, the bare section is restrained outside limiting sleeve by restraint wire, so that itself cannot be self-inflating, and the stent cannot "front jump", so as to reduce the risk of blocking portal vein branch.

The utility model relates to an automatic adjustment hepatic duodenal ligament portal vein and hepatic arterial blood flow on-off device which comprises a soft rubber sleeve, a handle, a connecting pipe, a first clamping strip, a second clamping strip, a rotating shaft, a flexible wire and a flexible wire control mechanism. The first clamping strip and the second clamping strip are sleeved with the soft rubber sleeve, one end of the connecting pipe is connected to the soft rubber sleeve, the other end of the connecting pipe is connected to the handle, the flexible wire control mechanism is arranged on the handle, the flexible wire is inserted into the connecting pipe in a penetrating mode, one end of the flexible wire is connected to the first clamping strip, and the other end of the flexible wire is connected to the flexible wire control mechanism. According to the utility model, the defects of the existing porta hepatis occlusion technology can be overcome, more accurate, safer and more convenient blood flow occlusion and conduction control can be realized, and the surgical risk and liver injury can be reduced.

The invention discloses CCL2, a receptor specific binding polypeptide thereof and application of the receptor specific binding polypeptide in prevention and treatment of cholestatic liver diseases, and relates to the technical field of biological medicines. The CCL2 and the receptor specific binding polypeptide thereof, which are screened by a molecular simulation technology, can be specifically bound with a key molecule CCL2 for inducing diseases and a plurality of receptors thereof in a cholestatic liver disease occurrence process, so that liver function indexes are recovered; the traditional Chinese medicine composition can be used for relieving liver pathological injury, including inhibiting bile duct hyperplasia, relieving portal vein inflammatory response and relieving portal vein fibrosis, so that the pathological development process of fibrosis and inflammation caused by the liver injury is blocked.

Disclosed is a device for the perfusion of an organ, including: a container of fluid, containing an organ bathed in the perfusion fluid; a first path including an inlet, an outlet and a pump; and a second path including an inlet, an outlet and a pump. The “arterial” outlet of the first path has a diameter smaller than a diameter of the “portal” outlet of the second path. The device additionally includes, between the pump and the outlet of the first path and / or between the pump and the outlet of the second path, an oxygenation unit arranged to oxygenate the fluid emerging from the “arterial” outlet of the first path more than the fluid emerging from the “portal” outlet of the second path. The device can include a communication path between the first path and the second path in order to oxygenate the second path. Use in liver transplantation.

A TIPS (Transient Invasive Stent) for flow restriction includes a bare stent with a fixed stent connected to one side. A flow-restricting stent is disposed inside the fixed stent. The flow-restricting stent has a conical structure, with the larger end of the conical stent connected to the inner side of the fixed stent, and the smaller end of the conical stent facing the other end of the fixed stent. A drug-eluting layer is disposed inside the fixed stent. In this application, the flow-restricting stent is a streamlined cone shape with a diameter that gradually decreases from one end to the other. This conical design reduces blood flow turbulence and recirculation, lowers the risk of thrombosis, maintains high stent patency, and more rationally balances the relationship between portal hypertension and hepatic encephalopathy, thus improving patients' quality of life. In some patients, it can more stably maintain the postoperative portal vein pressure gradient within the safe range of 10-12 mmHg. The conical structure achieves a smooth pressure gradient change through a gradually decreasing diameter, which is more in line with the principles of fluid dynamics and has high clinical value in preventing hepatic encephalopathy.

The invention provides an image-clinical characterization-combined multi-endpoint prognosis evaluation method for transjugular vein intrahepatic portal vein shunt, which comprises the following steps of: constructing a few-label portal vein segmentation module to obtain a preoperative CT portal vein label of a full-dose patient, and extracting deep learning features and radiomics features of the region; establishing a multi-modal interactive representation learning module for implementation, and performing cross-modal fusion with clinical features to form unified representation; and designing a multi-endpoint prognosis prediction module, inputting the data to a plurality of prognosis task decoders for postoperative survival, portal vein pressure gradient, hepatic encephalopathy prediction and the like, and adopting a multi-task learning optimization model to obtain a postoperative multi-endpoint prognosis evaluation result. According to the method, efficient fusion and multi-endpoint prognosis prediction of images and clinical information can be realized under limited labeling, clinical doctors can be assisted in preoperative patient screening and treatment scheme making, and the method has good clinical application value.

Disclosed is a composition comprising a therapeutically effective amount of zipotentan and dapagliflozin for use in a method of treating liver cirrhosis in a subject in need thereof.

The invention relates to the technical field of data processing, in particular to a postoperation delayed portal vein blood pressure monitoring method and device, equipment and a medium. A lumped parameter model is constructed based on a vein phase image of a patient before a jugular vein intrahepatic portal shunt operation; acquiring actually measured clinical data of the patient after jugular vein intrahepatic portal shunt operation; the constructed lumped parameter model is calibrated on the basis of the actually measured clinical data, and the delayed portal vein blood pressure of the patient after the jugular vein intrahepatic portal shunt operation is solved on the basis of the calibrated lumped parameter model. Therefore, numerical solution is achieved on the constructed lumped parameter model based on postoperative actual measurement clinical data calibration parameters, postoperative delayed portal vein blood pressure can be monitored in a non-invasive mode, three-dimensional simulation is not needed, and consumed time is saved.

The present invention relates to the technical field of pharmaceuticals and specifically describes an application of a 5-hydroxytryptamine receptor 1A in preparing a drug for portal hypertension. The present invention achieved a significant decrease in portal venous pressure in model rats with cirrhotic portal hypertension by means of an HTR1A inhibitor WAY100635 administered at 1 mg / kg / day by intraperitoneal injection or alverine administered at 15 mg / kg / day by oral gavage. A 5-hydroxytryptamine receptor 1A antagonist can be used to prepare an experimental drug for the treatment of cirrhotic portal hypertension. Furthermore, the 5-hydroxytryptamine receptor 1A antagonist can also be used to prepare drugs for symptoms of portal hypertension syndrome, such as esophageal and fundal varices, rupture and bleeding, ascites, or hepatic encephalopathy. The present invention provides a novel treatment means to treat the manifestations of portal hypertension syndrome, such as esophageal and fundal varices, rupture and bleeding, ascites, or hepatic encephalopathy.

The invention discloses a portal vein stent shunt intraoperative blood flow proportion calculation method based on hemodynamics, and the method comprises the steps: obtaining target data which comprises target clinical measurement data and target image data after a target patient TIPS operation; calculating a target boundary condition based on the target data; constructing a target model, wherein the target model is a calculation model for calculating the blood mass flow rate based on geometric discretization and CFD simulation; and based on the target boundary condition and the target model, obtaining the blood proportion from the superior mesenteric vein and the spleen vein in the stent after the TIPS operation of the target patient. The blood proportion of the superior mesenteric vein and the splenic vein in the stent after the TIPS operation of the patient is calculated, and the difference of puncture from different branches of the portal vein in the TIPS operation is effectively judged, so that on the basis of effectively reducing the portal vein pressure, the occurrence of hepatic encephalopathy is reduced by optimizing the TIPS operation scheme.

The invention discloses a method for blood vessel segmentation, feature value extraction and data integration based on V-net, and relates to the technical field of medical image analysis. Comprising the steps that S1, a CT image of a patient is acquired and preprocessed; s2, constructing and training a V-net neural network, inputting the preprocessed CT image into the trained V-net neural network, and reconstructing a blood vessel model; s3, feature value extraction is carried out on the blood vessel model. By constructing and training the V-net neural network, the blood vessels in the patient's abdomen CT medical image are segmented and reconstructed to obtain portal vein, hepatic artery, aorta and other blood vessel models with good segmentation effects, and feature values of the blood vessel models are extracted and data are standardized through an algorithm, so that the accuracy of the blood vessel models is improved. Therefore, the CT image information can show the blood vessel form and the characteristic value in a model and datamation form, and the concerned blood vessel characteristic information can be obtained from the complex CT image.

The invention aims to provide a reconstruction algorithm based on deep learning, which takes an enhanced pre-liver MRI image as input, does not need a contrast agent, and efficiently and quickly outputs multi-phase (arterial phase, portal vein phase, delay phase and liver and gall specific phase) enhanced images of the liver for clinical diagnosis and evaluation. In order to achieve the purpose, the technical scheme of the invention provides a virtual multi-phase contrast agent enhanced magnetic resonance imaging reconstruction method without a contrast agent. According to the method, a deep learning model based on a Generative Advanced Network is established, an MRI image before enhancement is taken as input, real enhanced images, acquired through MRI scanning, of all phases are taken as a gold standard training model, and after model training is completed, any image before enhancement can be input, so that corresponding four-phase liver enhanced images are output by the model.

The invention discloses an SMV blood distribution calculation method and a postoperative hepatic encephalopathy risk assessment method, and relates to the technical field of medicines.The method comprises the steps that S1, CTA image data and ultrasonic data of a patient are obtained; s2, performing three-dimensional reconstruction on portal vein system dissection and a TIPS stent; s3, numerical simulation is completed; s4, quantitatively analyzing SMV blood distribution characteristics; s5, SMV blood distribution characteristics are obtained through SMV blood distribution calculation before the TIPS operation, the proportion of SMV blood flowing to the left portal vein branch and the right portal vein branch is analyzed, the position where the stent is implanted is selected, the proportion of SMV blood entering the TIPS stent after the operation is reduced, and the risk of hepatic encephalopathy of the patient after the TIPS operation is evaluated. According to the invention, the flow rule of the SMV blood in the portal vein system can be analyzed, and the volume ratio (percentage) of the SMV blood at any position can be obtained; the'blood ammonia hypothesis' is verified, and the estimation accuracy of the hepatic encephalopathy is improved.

The invention relates to the technical field of microfluidics, and discloses a biliary duct-containing bionic liver chip and application thereof. A central vein pool, a hepatic portal vein channel and a hepatic inherent artery channel are arranged on the surface of a substrate of the bionic liver chip; a bile duct / hepatocyte U-shaped channel; and a stellate cell channel A and a stellate cell channel B, according to the liver chip, bile duct cells and a biliary tract simulation structure used for collecting bile are added while the liver chip simulates the physiological structure of the liver, interaction between the liver cells and the gall cells can be achieved, an in-vitro liver system is more bionic, and therefore the liver chip can better serve related research of biomedicine; comprising liver disease modeling, drug toxicity detection, drug effect evaluation and the like.

The present invention discloses a particle stent for portal vein, a particle stent for cavity and a kit thereof. The particle stent includes a fixing portion; a first parallel portion, including a plurality of first edges parallel to the length direction of the particle stent; a second parallel portion, including a plurality of second edges parallel to the length direction of the particle stent; a supporting portion, extending along the circumferential direction of the particle stent to provide radial supporting force for the first edge and the second edge; a particle filling portion, wherein each group of nickel-titanium alloy wires are mutually wound from the first end to the support starting point to form a first edge; and the first edge is dispersed into two nickel-titanium alloy wires from the support starting point, each nickel-titanium alloy wire is bent at the support end point, and is mutually wound with the adjacent group of nickel-titanium alloy wires from the support end point to the second end to form a second edge. The particle stent can not only realize the precise arrangement of radiation particles, but also avoid the displacement of cancer thrombi.

The invention discloses a multi-modal feature fused portal vein pressure gradient acquisition method and system, and relates to the technical field of portal vein pressure acquisition, and the method comprises the steps: S1, obtaining a CT image of a patient, and carrying out the preprocessing of the CT image, and obtaining the related feature data of the CT image; s2, constructing a multi-modal feature model based on liver blood vessels in the CT image; inputting the preprocessed CT image data, the related feature data and the corresponding to-be-predicted portal vein pressure gradient label value into a model for training; and S3, inputting the preprocessed CT image data and the related feature data into the trained model, and performing prediction to obtain a portal vein pressure gradient value corresponding to the CT image. Through an innovative multi-modal fusion architecture and a two-stage training strategy, pathological characteristics of geometric structure information and physical parameter information of a three-dimensional model are fully utilized, and high-precision and high-robustness non-contact prediction of PPG values under the condition that only CT image blood vessel model input exists is realized.