Real-time monitoring system and monitoring method for organ activity of kidney transplantation donor

Abstract

The invention discloses a kidney transplantation donor organ activity real-time monitoring system and method in the technical field of isolated organ perfusion, and the system comprises a multi-modal sensing module which is used for sensing multi-modal data of molecular action generation, cell mechanical characteristics and organ functions of an isolated kidney in a cross-scale manner; the perfusion dynamic regulation and control module is used for fusing multi-modal data based on a learning algorithm and optimizing perfusion parameters in real time; and the bionic normal-temperature perfusion module is used for simulating a human physiological environment and maintaining kidney activity by using a micro-fluidic chip technology and a self-adaptive oxygenation technology based on the optimized perfusion parameters. A cross-scale multi-modal monitoring system is constructed, multi-modal data of the in-vitro kidney is sensed, dynamic perfusion regulation and control based on a learning algorithm are developed by fusing the multi-modal data, self-adaptive optimization of perfusion parameters is achieved, a perfusion environment similar to human physiology is created for the in-vitro kidney, and therefore the DGF risk is effectively reduced, and the in-vitro kidney perfusion system is suitable for in-vitro kidney perfusion. The success rate of kidney transplantation and the long-term survival rate of donor organs are improved.

Description

Technical Field

[0001] This invention belongs to the field of ex vivo organ perfusion technology, specifically a real-time monitoring system and method for the viability of kidney transplant donor organs. Background Technology

[0002] In the field of kidney transplantation, delayed graft function recovery (DGF) is a common and challenging early complication after surgery. As a manifestation of early acute kidney injury (AKI) in transplanted kidneys, it can have many adverse effects on patients. For example, DGF can lead to oliguria in transplanted patients, which greatly increases graft immunogenicity and consequently raises the risk of acute rejection. Moreover, DGF is an independent risk factor affecting the long-term survival of transplanted kidneys.

[0003] A deeper investigation into the causes of DGF reveals that the ischemic and hypoxic processes experienced before organ procurement, during preservation, and after transplantation, especially the resulting renal tubular ischemia-reperfusion injury (IRI), are key factors leading to DGF. Currently, real-time monitoring systems and methods for monitoring kidney transplant donor organ viability have a series of shortcomings that urgently need to be addressed.

[0004] On the one hand, traditional hypothermic mechanical perfusion (HMP) technology has significant limitations in the organ preservation stage. It can only assess organ viability through the indirect indicator of perfusion resistance index, failing to provide real-time monitoring of dynamic changes such as renal tubular cell apoptosis and the release of inflammatory factors, which directly reflect kidney damage. Studies have confirmed that the risk of diabetic glomerulonephritis (DGF) increases significantly once cold ischemia time exceeds 24 hours; however, current monitoring systems struggle to provide early warnings of damage during organ preservation, failing to provide timely critical information to healthcare professionals for intervention.

[0005] On the other hand, although organ preservation solutions and hypothermic mechanical perfusion solutions are widely used in clinical practice, the incidence of disease-free blood vessels (DGF) remains high with prolonged cold ischemia time. Furthermore, hypothermic mechanical perfusion has very limited effectiveness in improving the long-term prognosis of transplanted kidneys, as exemplified by the existing LifePort Kidney Transporter device. In particular, for highly immunogenic donor kidneys such as those from DCD donors, current technologies are severely lacking in repair capabilities, making it difficult to effectively reverse complement-mediated microvascular damage, thus failing to reliably guarantee the quality and function of the transplanted kidney, as illustrated in the patent document CN104430302B.

[0006] Furthermore, most existing monitoring methods rely on manual adjustment of perfusion parameters at the operational level. Setting important parameters such as flow rate and oxygen concentration requires highly skilled operators to perform complex and precise operations to ensure the accuracy of the monitoring data. This high dependence on manual operation greatly reduces the reproducibility of monitoring operations, affects the stability and reliability of monitoring results, and is detrimental to the development of precise and efficient monitoring of kidney transplant donor organ viability.

[0007] Therefore, it is necessary to propose a real-time monitoring system and method for kidney transplant donor organ viability that can monitor organ multimodal data in the early stage of ex vivo organ transplantation, assess organ viability, and adaptively adjust perfusion parameters and perfusion components based on multimodal data to improve and repair acute damage to ex vivo organs, reduce the risk of DGF, reduce manual operation, and enhance the real-time monitoring of perfusion. Summary of the Invention

[0008] To address the aforementioned issues, the present invention aims to provide a real-time monitoring system and method for the viability of kidney transplant donor organs. This system constructs a multi-scale, multimodal monitoring system to accurately perceive multimodal data from isolated kidneys, ranging from molecular interactions to cellular mechanical properties and organ function. By integrating this multimodal data, a perfusion dynamic control system based on a multi-agent reinforcement learning algorithm is developed to achieve adaptive optimization of perfusion parameters. Furthermore, microfluidic chip technology and adaptive oxygenation technology are employed to create a perfusion environment for isolated kidneys that closely resembles human physiology, enabling targeted repair of highly immunogenic donor kidneys. This effectively reduces the risk of DGF (digestive globulin-related lesions) and improves the success rate of kidney transplantation and the long-term survival rate of donor organs.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows: a real-time monitoring system for the viability of kidney transplant donor organs, comprising a multimodal sensor module, a perfusion dynamic control module, and a biomimetic room-temperature perfusion module;

[0010] A multimodal sensing module is used to perceive multimodal data on molecular interactions, cellular mechanical properties, and organ function of isolated kidneys across scales.

[0011] The perfusion dynamic control module is used to optimize perfusion parameters in real time by fusing multimodal data based on learning algorithms;

[0012] The biomimetic ambient temperature perfusion module is used to simulate the human physiological environment and maintain kidney activity based on optimized perfusion parameters, using microfluidic chip technology and adaptive oxygenation technology.

[0013] Furthermore, the multimodal sensing module includes molecular-level monitoring units, cell-level monitoring units, and organ-level monitoring units;

[0014] A molecular-level monitoring unit is used to detect oxidative stress and complement activation levels in real time using ROS-responsive fluorescent probes and complement-specific antibody probes.

[0015] A cell-level monitoring unit is used to dynamically measure the stiffness and viscoelasticity of renal tubular epithelial cells using optical coherence elastography.

[0016] An organ-level monitoring unit is used to calculate corrected creatinine clearance based on a microfluidic biomimetic nephron model.

[0017] Furthermore, the perfusion dynamic regulation module includes a multi-agent reinforcement learning algorithm and a kidney digital twin model.

[0018] A multi-agent reinforcement learning algorithm is used to fuse multimodal data of metabolites, oxygen partial pressure, fluorescence signals and mechanical properties to build a dynamic optimization model of perfusion parameters, train and adjust dynamic perfusion parameters;

[0019] A digital twin model of the kidney is used to construct a three-dimensional hemodynamic simulation based on CT / MRI data of an isolated kidney, and to predict the impact of interventions on local perfusion.

[0020] Furthermore, the objective function of the multi-agent reinforcement learning algorithm is:

[0021] min {P,O,F} (α·ΔROS+β·ΔcGFR+γ·Cell_Stiffness)

[0022] In the formula, the perfusion pressure P, oxygen concentration O, and flow rate F are dynamically adjusted, and the weighting coefficients α, β, and γ are optimized through training with clinical data.

[0023] Furthermore, the biomimetic room-temperature infusion module includes a dual-layer microfluidic chip and an adaptive oxygenation unit.

[0024] A dual-layer microfluidic chip is used to simulate the blood flow shear force in the glomerulus and to control the temperature and oxygen partial pressure in a closed loop. The glomerular filtration pressure and filtration fraction are adjusted by air pressure drive.

[0025] An adaptive oxygenation unit is used to dynamically adjust the oxygen release rate in response to medullary oxygen saturation using hemoglobin oxygen carrier perfusion fluid via an electrochemical oxygen pump.

[0026] Furthermore, the dual-layer microfluidic chip includes a vascular layer and a renal tubule layer, with a deformable PDMS membrane disposed between the vascular layer and the renal tubule layer. The PDMS membrane is used to regulate the glomerular filtration pressure and filtration fraction by air pressure drive.

[0027] Furthermore, the oxygen delivery efficiency of the adaptive oxygenation unit is 3 times or more than that of traditional erythrocyte perfusion fluid, maintaining an isolated kidney ATP level >2.5 μmol / g tissue.

[0028] Furthermore, the multimodal sensing module and the biomimetic room temperature infusion module are connected via a detachable interface, which supports quick replacement of the microfluidic chip, and the interface sealing pressure is ≥1.5atm.

[0029] The basic principle of this invention is as follows: The real-time monitoring system for kidney transplant donor organ activity of this invention achieves comprehensive monitoring and precise control of isolated kidneys through the coordinated operation of a multimodal sensor module, a perfusion dynamic control module, and a biomimetic ambient temperature perfusion module. The multimodal sensor module can sense multimodal data on molecular interactions, cellular mechanical properties, and organ function of isolated kidneys across scales, providing real-time data support for subsequent perfusion parameter optimization. Specifically, the molecular-level monitoring unit utilizes ROS-responsive fluorescent probes and complement-specific antibody probes to accurately detect oxidative stress and complement activation levels, thereby revealing changes at the molecular level of the kidney; the cellular-level monitoring unit uses optical coherence elastography to dynamically measure the stiffness and viscoelasticity of renal tubular epithelial cells, reflecting the physical state of the cells; and the organ-level monitoring unit calculates and corrects creatinine clearance rate using a microfluidic biomimetic nephron model, visually presenting the overall function of the kidney.

[0030] The perfusion dynamic regulation module, based on a multi-agent reinforcement learning algorithm, integrates multimodal data, including metabolites, oxygen partial pressure, fluorescence signals, and mechanical properties, to construct a dynamic optimization model for perfusion parameters. This enables real-time optimization and adjustment of key parameters such as perfusion pressure, oxygen concentration, and flow rate. Furthermore, this module combines a digital twin model of the kidney with 3D hemodynamic simulations based on CT / MRI data of the ex vivo kidney to predict the impact of interventions on local perfusion, providing a scientific basis for precise regulation of perfusion parameters.

[0031] The biomimetic ambient temperature perfusion module employs microfluidic chip technology and adaptive oxygenation technology to simulate the human physiological environment and maintain kidney activity. Specifically, the dual-layer microfluidic chip regulates glomerular filtration pressure and filtration fraction via pneumatic drive, simulating blood flow shear force in the glomerulus; the adaptive oxygenation unit dynamically adjusts the oxygen release rate based on medullary oxygen saturation, maintaining ATP levels in the isolated kidney and providing stable energy support.

[0032] The beneficial effects of the basic solution are: 1. The real-time monitoring system for kidney transplant donor organ activity of the present invention can achieve precise monitoring and intelligent regulation of isolated kidneys, effectively solving the problem of inaccurate monitoring in the prior art. Through cross-scale monitoring by multimodal sensor modules, it can comprehensively and in real-time acquire information at the molecular, cellular, and organ levels of isolated kidneys, providing rich data for accurately assessing kidney activity.

[0033] 2. The dynamic control module for irrigation is based on a multi-agent reinforcement learning algorithm, which can automatically optimize irrigation parameters according to real-time monitoring data, improve irrigation efficiency and accuracy, reduce errors caused by manual intervention, and enhance the reproducibility of monitoring irrigation.

[0034] 3. The biomimetic ambient temperature perfusion module simulates the human physiological environment, providing perfusion conditions closer to physiological states for the isolated kidney, which helps maintain kidney activity, reduces the risk of DGF (digestive fibrosis), and improves transplant success rate. The highly efficient oxygen delivery capacity of the adaptive oxygenation unit can better meet the oxygen demand of the isolated kidney, maintain its ATP level, and further protect kidney function.

[0035] 4. The detachable interface design between the multimodal sensing module and the biomimetic room temperature infusion module facilitates quick replacement of the microfluidic chip, improving the system's flexibility and practicality.

[0036] 5. The monitoring system of the present invention has significant advantages in improving the accuracy of monitoring the viability of kidney transplant donor organs, the level of intelligent perfusion regulation, and maintaining the viability of isolated kidneys, and is expected to provide strong support for the success of kidney transplantation surgery and the long-term survival of donor organs.

[0037] Furthermore, it also includes a preprocessing analysis module for reconstructing a three-dimensional vascular tree model based on donor kidney angiography data and pre-calculating the optimal perfusion path.

[0038] The beneficial effects of the basic scheme are as follows: 1. By introducing a preprocessing analysis module, this invention further enhances the assessment capability and perfusion efficiency of the real-time monitoring system for donor organ viability in kidney transplantation. Based on the three-dimensional vascular tree model reconstructed from donor kidney angiography data, the optimal perfusion path can be accurately pre-calculated, optimizing blood flow distribution during perfusion and improving perfusion efficiency and uniformity. This not only helps improve the perfusion quality of the donor organ, ensuring the kidney receives sufficient nutrition and oxygen supply before transplantation, but also reduces local ischemia or damage caused by poor perfusion, further improving the functional recovery and survival rate of the transplanted kidney.

[0039] 2. The addition of the pretreatment analysis module makes the entire monitoring system more functionally complete. By optimizing the perfusion pathway, it provides comprehensive protection for the kidney transplantation process, offering strong technical support for improving transplant success rates and patient prognosis.

[0040] A method for real-time monitoring of kidney transplant donor organ viability, based on the aforementioned real-time monitoring system for kidney transplant donor organ viability, includes the following steps.

[0041] Step 1: Pretreatment of ex vivo organs. The ex vivo kidney is connected to a biomimetic ambient temperature perfusion module. ROS-responsive fluorescent probes are injected into the perfusion fluid, and the optimal perfusion path is calculated.

[0042] Step 2: Dynamic parameter control. The initial perfusion parameters of the dynamic perfusion control module are perfusion pressure of 80 mmHg, flow rate of 150 mL / min, and oxygen partial pressure of 70 mmHg. Multimodal data is collected and parameters are optimized every 5 seconds through a multi-agent reinforcement learning algorithm. When ROS fluorescence intensity > 500 AU or cell stiffness > 3 kPa, the oxygen concentration is automatically increased to 90 mmHg and the flow rate is increased by 20%. When complement signal > 100 pg / mL is detected, the perfusion fluid composition is automatically adjusted and complement inhibitors are added.

[0043] Step 3: Activity assessment. Corrected cGFR is measured using a biomimetic nephron model, with a threshold > 40 mL / min / 1.73 mcg. 2 The density of capillaries and the continuity of blood flow were verified by OCT-Angio imaging.

[0044] The beneficial effects of the basic scheme are: 1. Based on an advanced monitoring system, this invention achieves accurate assessment, intelligent regulation, and effective protection of isolated kidneys. In the pretreatment stage, by connecting the isolated kidney to a biomimetic ambient temperature perfusion module and injecting ROS-responsive fluorescent probes into the perfusion fluid, the level of oxidative stress can be monitored in real time.

[0045] 2. In the dynamic regulation phase, the perfusion dynamic regulation module starts with initializing perfusion parameters and uses a multi-agent reinforcement learning algorithm to collect multimodal data every 5 seconds and optimize perfusion parameters, achieving automation and intelligence in the perfusion process. This ability to adjust perfusion parameters in real time ensures that the kidney receives optimal perfusion conditions in an ex vivo state, reducing damage caused by insufficient or excessive perfusion. When ROS fluorescence intensity or cell stiffness exceeds a set threshold, the system automatically increases oxygen concentration and flow rate to meet the immediate needs of the kidney; and when complement signals exceed a threshold, complement inhibitors are automatically added to further protect the kidney from immune damage.

[0046] 3. During the evaluation phase, cGFR was measured and corrected based on a biomimetic nephron model, and capillary density and blood flow continuity were verified using OCT-Angio imaging, providing comprehensive data support for assessing kidney function and vascular status. This integrated evaluation method can more accurately determine kidney viability and transplant success rate, providing a scientific basis for clinical decision-making. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of a real-time monitoring system for the viability of kidney transplant donor organs in an embodiment of the present invention.

[0048] Figure 2 This is a schematic diagram of the multimodal sensing module in an embodiment of the present invention.

[0049] Figure 3This is a schematic diagram of the infusion dynamic control module in an embodiment of the present invention.

[0050] Figure 4 This is a schematic diagram of the biomimetic room temperature infusion module in an embodiment of the present invention.

[0051] Figure 5 This is a flowchart illustrating the method for real-time monitoring of kidney transplant donor organ viability in an embodiment of the present invention. Detailed Implementation

[0052] The following detailed description illustrates the specific implementation method:

[0053] Example 1

[0054] The basics are as follows: Figure 1 , Figure 2 , Figure 3 and Figure 4 The system described is a real-time monitoring system for the viability of kidney transplant donor organs. It is characterized by comprising a multimodal sensing module for sensing multimodal data on molecular interactions, cellular mechanical properties, and organ function of isolated kidneys across scales. The multimodal sensing module includes a molecular-level monitoring unit for real-time detection of oxidative stress and complement activation levels using ROS-responsive fluorescent probes and complement-specific antibody probes; a cellular-level monitoring unit for dynamically measuring the stiffness and viscoelasticity of renal tubular epithelial cells using optical coherence elastography; and an organ-level monitoring unit for calculating corrected creatinine clearance based on a microfluidic biomimetic nephron model.

[0055] The perfusion dynamic regulation module is used to optimize perfusion parameters in real time by fusing multimodal data based on a learning algorithm. This module includes a multi-agent reinforcement learning algorithm, which fuses multimodal data on metabolites, oxygen partial pressure, fluorescence signals, and mechanical properties to construct a dynamic optimization model for perfusion parameters. The module trains and adjusts the dynamic perfusion parameters. The objective function of the multi-agent reinforcement learning algorithm is:

[0056] min {P,O,F} (α·ΔROS+β·ΔcGFR+γ·Cell_Stiffness)

[0057] In the formula, the perfusion pressure P, oxygen concentration O, and flow rate F are dynamically adjusted, and the weighting coefficients α, β, and γ are optimized through training with clinical data; the kidney digital twin model is used to construct a three-dimensional hemodynamic simulation based on CT / MRI data of the ex vivo kidney to predict the impact of intervention measures on local perfusion.

[0058] The biomimetic ambient temperature perfusion module, based on optimized perfusion parameters, utilizes microfluidic chip technology and adaptive oxygenation technology to simulate the human physiological environment and maintain kidney activity. The module includes a dual-layer microfluidic chip to simulate blood flow shear force in the glomerulus and control temperature and oxygen partial pressure in a closed loop. It regulates glomerular filtration pressure and fraction of filtration via pneumatic pressure. The dual-layer microfluidic chip comprises a vascular layer and a renal tubular layer, with a deformable PDMS membrane installed between them. The PDMS membrane is used to regulate glomerular filtration pressure and fraction of filtration via pneumatic pressure. The adaptive oxygenation unit uses hemoglobin-based oxygen carrier perfusion fluid and dynamically adjusts the oxygen release rate in response to medullary oxygen saturation via an electrochemical oxygen pump. The oxygen delivery efficiency of the adaptive oxygenation unit is three times or more than that of traditional erythrocyte perfusion fluid, maintaining an ATP level >2.5 μmol / g tissue in isolated kidney tissue.

[0059] It also includes a preprocessing analysis module for reconstructing a three-dimensional vascular tree model based on donor kidney angiography data and pre-calculating the optimal perfusion path.

[0060] The multimodal sensing module and the biomimetic room temperature infusion module are connected via a detachable interface, which supports quick replacement of the microfluidic chip, and the interface sealing pressure is ≥1.5atm.

[0061] The specific implementation process is as follows: This system achieves real-time monitoring and optimized maintenance of the active state of isolated kidneys through the deep integration of multimodal sensing, dynamic control algorithms and biomimetic perfusion technology.

[0062] The ROS-responsive fluorescent probe in the molecular-level monitoring unit of the multimodal sensing module has an excitation wavelength of 980 nm and an emission band of 1000-1700 nm (NIR-II window). It achieves a sensitivity of up to 1 nM ROS concentration and can penetrate kidney tissue to a depth of 5-8 mm, enabling real-time monitoring of renal tubular oxidative stress levels. The complement C5a-specific antibody probe uses quantum dot (CdSe / ZnS core-shell structure) labeled anti-human C5a monoclonal antibody, with a detection limit as low as 5 pg / mL, dynamically tracking complement activation signals. The probe targets and binds to the target molecule through perfusion fluid circulation, and the fluorescence signal is acquired by a fiber optic spectrometer, with data transmitted to the control module in real time.

[0063] The optical coherence elastography (OCE) technology of the cell-level monitoring unit has a sampling frequency ≥10Hz, a spatial resolution of 10μm, and a detection range of 0.1-10kPa. Kidney tissue is irradiated with low-coherence light, and the reflected light interference signal is analyzed to generate a thermal map of stiffness and viscoelasticity distribution. The Young's modulus threshold is set at 3kPa (exceeding this threshold indicates early ischemic injury), and the viscoelastic loss modulus threshold is set at 50Pa (exceeding this threshold indicates cellular structural damage). Real-time monitoring of changes in the mechanical properties of renal tubular epithelial cells provides cellular mechanical data support for dynamic regulation.

[0064] The microfluidic biomimetic nephron model chip of the organ-level monitoring unit measures 10cm × 5cm × 1cm, simulates a glomerular filtration pressure of 10-20 mmHg, and has a filtration membrane pore size of 5-10 nm (biomimetic podocyte slit membrane) with an accuracy error ≤5%. The corrected creatinine clearance rate (cGFR) is calculated in real time using a pressure sensor (range 0-50 mmHg) and a flow meter (accuracy ±0.1 mL / min). The calculation formula is as follows:

[0065]

[0066] It dynamically simulates glomerular filtration function and calculates organ-level functional indicators by combining metabolite concentrations, replacing traditional lagging biochemical tests.

[0067] The input data for the multi-agent reinforcement learning algorithm in the perfusion dynamic regulation module includes ROS concentration, cGFR, cell stiffness, oxygen partial pressure (40-100 mmHg), and flow rate (50-200 mL / min). The output variables include perfusion pressure (P), oxygen concentration (O), and flow rate (F), with adjustment steps of ±5 mmHg, ±10 mmHg, and ±20 mL / min, respectively.

[0068] The digital twin model of the kidney reconstructs a three-dimensional vascular network based on CT / MRI data of an ex vivo kidney (0.5 mm slice thickness), with a mesh generation accuracy of 0.1 mm. 3 Import hemodynamic parameters (viscosity 3.5 cP, density 1.06 g / cm³). 3 This simulation model simulates local blood flow distribution and predicts the impact of interventions (such as prostaglandin E1 injection at 0.1 μg / kg) on ​​blood flow, for example, a 15%-20% increase in medullary blood flow after vasodilation. It also integrates with a multi-agent reinforcement learning algorithm to provide simulation validation for parameter optimization.

[0069] The vascular layer of the bilayer microfluidic chip in the ambient temperature perfusion module is a fractal vascular network 3D printed from hydrogel (PEGDA-based), with 6 levels of branching, a minimum channel diameter of 50 μm, and simulated blood flow shear force of 0.1-10 dyn / cm. 2 It integrates a PID temperature control system (accuracy ±0.5℃) and an oxygen partial pressure sensor (accuracy ±1mmHg) to maintain 37℃ and oxygen partial pressure 70-80mmHg.

[0070] A deformable PDMS membrane (200 μm thick) in the nephron layer regulates the glomerular filtration pressure via a pneumatically driven system (pressure range 0-30 kPa), with fluctuations ranging from 10-20 mmHg ± 0.5 mmHg. The dynamic adjustment formula for the filtration fraction (FF) is as follows:

[0071]

[0072] The target FF range is 15%-25%, simulating a physiological state.

[0073] The adaptive oxygenation unit uses hemoglobin oxygen carrier (HBOC) perfusion solution with an oxygen-carrying capacity ≥60 mL O2 / dL, free hemoglobin <10 mg / dL, and a viscosity of 2.8 cP (close to plasma). The electrochemical oxygen pump responds to the medullary oxygen saturation (SmO2) detected by a ROS-responsive fluorescent probe, dynamically adjusting the oxygen release rate (0.1-5 mL / min) to maintain ATP >2.5 μmol / g tissue. When SmO2 <60%, the oxygen pump power increases to 120%, restoring oxygenation levels within 30 seconds.

[0074] CTA / MRA data reconstruction uses segmentation software (such as Mimics) to extract the renal artery to interlobular artery branches, generating the optimal perfusion path (prioritizing perfusion of ischemic areas). The path planning algorithm (Dijkstra's algorithm) calculates the path of least resistance to guide the initial flow rate setting (e.g., 150 mL / min for the main renal artery).

[0075] This invention provides end-to-end monitoring from molecular events to organ function, surpassing traditional technologies in sensitivity and timeliness. The multi-agent reinforcement learning algorithm has a response time of less than 1 second, improving efficiency by 10 times compared to manual control. A dual-layer microfluidic chip and HBOC oxygenation technology maintain the viability of isolated kidneys for over 72 hours. The modular design adapts to different donor types, supporting rapid transfer to transplant centers.

[0076] Example 2

[0077] The difference from the above embodiments is that, as shown in the appendix Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5 The following is illustrated: A method for real-time monitoring of kidney transplant donor organ viability, based on the aforementioned real-time monitoring system for kidney transplant donor organ viability, includes the following steps:

[0078] Step 1: Pretreatment of isolated organs. The isolated kidney was connected to a biomimetic ambient temperature perfusion module. ROS-responsive fluorescent probes were injected into the perfusion fluid, and the optimal perfusion path was calculated.

[0079] Step 2: Dynamic parameter control. The initial perfusion parameters of the dynamic perfusion control module are perfusion pressure of 80 mmHg, flow rate of 150 mL / min, and oxygen partial pressure of 70 mmHg. Multimodal data is collected and parameters are optimized every 5 seconds through a multi-agent reinforcement learning algorithm. When ROS fluorescence intensity > 500 AU or cell stiffness > 3 kPa, the oxygen concentration is automatically increased to 90 mmHg and the flow rate is increased by 20%. When complement signal > 100 pg / mL is detected, the perfusion fluid composition is automatically adjusted and complement inhibitors are added.

[0080] Step 3: Activity assessment. Corrected cGFR is measured using a biomimetic nephron model, with a threshold > 40 mL / min / 1.73 mcg. 2 OCT-Angio imaging was used to verify capillary density and blood flow continuity.

[0081] The specific experimental procedure is as follows: I. Experimental Preparation

[0082] 1. Harvesting and processing of isolated kidneys: After harvesting, the kidneys were stored and transported in UW solution at 4°C. The weight, volume, and cold ischemia time were recorded. The vascular bed was flushed with physiological saline containing heparin (10 U / mL) to remove thrombi.

[0083] 2. Equipment Preparation: Connect and debug all modules of the real-time monitoring system for kidney transplant donor organ viability, including the multimodal sensor module, perfusion dynamic control module, biomimetic ambient temperature perfusion module, and preprocessing analysis module. Prepare the necessary perfusion fluid, fluorescent probes, and other reagents. Prepare the OCT-Angio imaging equipment for subsequent evaluation.

[0084] II. Experimental Procedure

[0085] 1. Connect the isolated kidney to the biomimetic ambient temperature perfusion module, ensuring a tight and leak-free connection.

[0086] 2. Inject ROS-responsive fluorescent probes into the perfusion solution to ensure uniform distribution.

[0087] 3. Start the dynamic control module for perfusion and set the initial perfusion parameters: perfusion pressure 80 mmHg, flow rate 150 mL / min and oxygen partial pressure 70 mmHg.

[0088] 4. Multimodal data (including metabolites, oxygen partial pressure, fluorescence signal, and mechanical properties) are collected every 5 seconds using a multi-agent reinforcement learning algorithm, and perfusion parameters are optimized according to preset conditions. When ROS fluorescence intensity > 500 AU or cell stiffness > 3 kPa, the oxygen concentration is automatically increased to 90 mmHg and the flow rate is increased by 20%; when complement signal > 100 pg / mL is detected, the perfusion fluid composition is automatically adjusted and complement inhibitors are added.

[0089] 5. Monitor various indicators of the isolated kidney in real time and perfuse continuously for 30 minutes.

[0090] 6. Based on a biomimetic nephron model, the corrected cGFR was determined, with a threshold >40 mL / min / 1.73 m 2 .

[0091] 7. Use OCT-Angio imaging to verify capillary density and blood flow continuity.

[0092] 8. Record all data during the experiment, including changes in perfusion parameters, fluorescence intensity, cell stiffness, complement signal, corrected cGFR, capillary density, etc.

[0093] 9. In addition, pathological analysis of renal tissue puncture biopsy was performed on the isolated kidney room temperature perfusion experimental group to establish a quantitative correspondence between non-invasive perfusion parameters and invasive pathological damage, proving that real-time perfusion monitoring can replace puncture biopsy to assess organ viability.

[0094] A 16G needle was used to puncture and sample the lower pole of the kidney at the cortico-medullary junction (avoiding large blood vessels); symmetrical areas of the same donor kidney were sampled before perfusion and 30 minutes after perfusion; the tissue was immediately fixed in 4% neutral formaldehyde, embedded in paraffin, and sectioned.

[0095] 10. Sampling and staining: HE staining: assess renal tubular necrosis and cast formation; PAS staining: observe brush border integrity; Masson trichrome staining: quantify interstitial fibrosis.

[0096] 11. Scoring criteria: Renal tubular necrosis, 0 = none; 1 ≤ 5%; 2 = 25-50%; 3 ≥ 50%

[0097] Brush-like margin detachment: 0 = intact; 1 = focal; 2 = patchy; 3 = completely absent

[0098] Microthrombus formation: 0 = none; 1 = 1-2 per field of view; 2 = 3-5 per field of view; 3 ≥ 5

[0099] Mitochondrial vacuolation: 0 = none; 1 = mild; 2 = moderate; 3 = severe.

[0100] III. Experimental Results

[0101] The results are shown in the table below:

[0102] Table 1. Infusion Parameter Data

[0103]

[0104] At 10 minutes, ROS reached 600 AU, triggering an increase in oxygen concentration (70→90 mmHg) and flow rate (150→180 mL / min); C5a signal was 110 pg / mL, and eculizumab was injected. At 15 minutes, cell stiffness was 3.2 kPa, and the MARL algorithm was used to increase perfusion pressure to 85 mmHg to improve blood perfusion. At 25 minutes, SmO2 recovered to 75%, and oxygen partial pressure returned to 85 mmHg.

[0105] Table 2. Changes in renal function indicators

[0106]

[0107] Table 3. Evaluation Results

[0108]

[0109] * indicates a significant difference between the experimental group and the control group (t-test, α = 0.05).

[0110] Table 4. Dynamic correlation between perfusion parameters and pathological scores

[0111]

[0112]

[0113] The ATP level in this system group (2.8±0.4μmol / g) was significantly higher than that in the control group (1.1±0.3μmol / g).

[0114] During the experiment, the peak ROS (620 AU) decreased to 200 AU 30 minutes after regulation, a reduction of 67.7%; cGFR increased from 28 to 44 mL / min / 1.73 m 2 The transplantation threshold (>40) has been reached.

[0115] The evaluation results showed that capillary density was increased by 43.3% and cortical stiffness was reduced by 44.7% compared to traditional HMP, validating the microcirculation protective effect of biomimetic perfusion. The system achieved functional recovery within 30 minutes and is suitable for rapid assessment and repair of marginal donor kidneys (such as DCD grade III).

[0116] The results of the puncture experiment demonstrate the feasibility of non-invasive alternatives to invasive procedures:

[0117] ROS concentration was strongly correlated with renal tubular necrosis score (r = 0.91):

[0118] Necrosis score = 0.0025 × ROS - 0.18 (P < 0.001)

[0119] Cell stiffness was strongly correlated with brush border shedding score (r = 0.89):

[0120] Detachment score = 1.2 × hardness - 0.75 (P < 0.001)

[0121] This means that by establishing a quantitative mathematical model (R...) between perfusion parameters (ROS / stiffness / cGFR) and Banff pathology scores... 2 >0.85), which can realize non-invasive monitoring to completely replace puncture biopsy and solve the problem of secondary damage caused by repeated punctures to marginal donor kidneys.

[0122] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0123] The above descriptions are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A real-time monitoring system for the viability of kidney transplant donor organs, characterized in that: It includes a multimodal sensor module, a perfusion dynamic control module, and a biomimetic room-temperature perfusion module; A multimodal sensing module is used to perceive multimodal data on molecular interactions, cellular mechanical properties, and organ function of isolated kidneys across scales. The perfusion dynamic control module is used to optimize perfusion parameters in real time by fusing multimodal data based on learning algorithms; The biomimetic ambient temperature perfusion module is used to simulate the human physiological environment and maintain kidney activity based on optimized perfusion parameters, using microfluidic chip technology and adaptive oxygenation technology.

2. The real-time monitoring system for kidney transplant donor organ viability according to claim 1, characterized in that: The multimodal sensing module includes molecular-level monitoring units, cell-level monitoring units, and organ-level monitoring units; A molecular-level monitoring unit is used to detect oxidative stress and complement activation levels in real time using ROS-responsive fluorescent probes and complement-specific antibody probes. A cell-level monitoring unit is used to dynamically measure the stiffness and viscoelasticity of renal tubular epithelial cells using optical coherence elastography. An organ-level monitoring unit is used to calculate corrected creatinine clearance based on a microfluidic biomimetic nephron model.

3. The real-time monitoring system for kidney transplant donor organ viability according to claim 1, characterized in that: The perfusion dynamic regulation module includes a multi-agent reinforcement learning algorithm and a kidney digital twin model. A multi-agent reinforcement learning algorithm is used to fuse multimodal data of metabolites, oxygen partial pressure, fluorescence signals and mechanical properties to build a dynamic optimization model of perfusion parameters, train and adjust dynamic perfusion parameters; A digital twin model of the kidney is used to construct a three-dimensional hemodynamic simulation based on CT / MRI data of an isolated kidney, and to predict the impact of interventions on local perfusion.

4. The real-time monitoring system for kidney transplant donor organ viability according to claim 3, characterized in that: The objective function of the multi-agent reinforcement learning algorithm is: minutes {P,O,F} (α·ΔROS+β·ΔcGFR+γ·Cell_Stiffness) In the formula, the perfusion pressure P, oxygen concentration O, and flow rate F are dynamically adjusted, and the weighting coefficients α, β, and γ are optimized through training with clinical data.

5. The real-time monitoring system for kidney transplant donor organ viability according to claim 1, characterized in that: The biomimetic room-temperature infusion module includes a dual-layer microfluidic chip and an adaptive oxygenation unit. A dual-layer microfluidic chip is used to simulate the blood flow shear force in the glomerulus and to control the temperature and oxygen partial pressure in a closed loop. The glomerular filtration pressure and filtration fraction are adjusted by air pressure drive. An adaptive oxygenation unit is used to dynamically adjust the oxygen release rate in response to medullary oxygen saturation using hemoglobin oxygen carrier perfusion fluid via an electrochemical oxygen pump.

6. The real-time monitoring system for kidney transplant donor organ viability according to claim 5, characterized in that: The dual-layer microfluidic chip includes a vascular layer and a renal tubule layer, with a deformable PDMS membrane between the vascular layer and the renal tubule layer. The PDMS membrane is used to regulate the glomerular filtration pressure and filtration fraction by air pressure.

7. The real-time monitoring system for kidney transplant donor organ viability according to claim 5, characterized in that: The oxygen delivery efficiency of the adaptive oxygenation unit is 3 times or more than that of traditional erythrocyte perfusion fluid, maintaining an isolated kidney ATP level >2.5 μmol / g tissue.

8. The real-time monitoring system for kidney transplant donor organ viability according to claim 1, characterized in that: It also includes a preprocessing analysis module for reconstructing a three-dimensional vascular tree model based on donor kidney angiography data and pre-calculating the optimal perfusion path.

9. The real-time monitoring system for kidney transplant donor organ viability according to claim 1, characterized in that: The multimodal sensing module and the biomimetic room temperature infusion module are connected via a detachable interface, which supports quick replacement of the microfluidic chip, and the interface sealing pressure is ≥1.5atm.

10. A method for real-time monitoring of kidney transplant donor organ viability, based on the real-time monitoring system for kidney transplant donor organ viability as described in any one of claims 1-9, characterized in that: Includes the following steps, Step 1: Pretreatment of isolated organs: The isolated kidney was connected to a biomimetic ambient temperature perfusion module, and a ROS-responsive fluorescent probe was injected into the perfusion fluid to calculate the optimal perfusion path; Step 2, Dynamic Parameter Control: The initial perfusion parameters of the dynamic perfusion control module are perfusion pressure of 80 mmHg, flow rate of 150 mL / min, and oxygen partial pressure of 70 mmHg. Multimodal data is collected and parameters are optimized every 5 seconds using a multi-agent reinforcement learning algorithm. When ROS fluorescence intensity > 500 AU or cell stiffness > 3 kPa, the oxygen concentration is automatically increased to 90 mmHg and the flow rate is increased by 20%. When complement signal > 100 pg / mL is detected, the perfusion fluid composition is automatically adjusted and complement inhibitors are added. Step 3: Activity Assessment: Corrected cGFR was measured based on a biomimetic nephron model, with a threshold > 40 mL / min / 1.73 m 2 OCT-Angio imaging was used to verify capillary density and blood flow continuity.

Images