Intelligent loading device for drug-loaded microspheres

Abstract

The invention belongs to the technical field of medical instruments, and relates to an intelligent loading device for drug-loaded microspheres. The device comprises a shell, a loading cabin, a real-time monitoring module and a control module. The real-time monitoring module monitors optical or electrical signals reflecting the combination state of drugs and microspheres in a non-contact or contact mode, the control module judges the loading state according to a preset quantitative relation database and real-time signals, and loading is automatically completed when the target loading rate is reached. The standardization, intelligence and traceability of the loading process of the drug-loaded microspheres are realized.

Description

Technical Field

[0001] This invention belongs to the field of medical device technology, specifically relating to a smart loading device for drug-loaded microspheres. Background Technology

[0002] Drug-eluting beads (DEB) interventional embolization is one of the core methods for the local treatment of solid tumors such as intermediate and advanced liver cancer. Its clinical efficacy largely depends on the quality of the key preoperative preparation step of "drug loading"—that is, the process of sufficient and stable binding of chemotherapy drugs with the embolizing microspheres in vitro. However, a fundamental dilemma that has long faced this field is that current clinical practice relies entirely on medical staff manually timing and judging by experience using common timing tools such as stopwatches and alarm clocks.

[0003] This method has inherent systemic flaws. First, the fixed countdown pattern cannot respond to the significant differences in different drug microsphere combinations (e.g., irinotecan typically requires about 30 minutes, while doxorubicin requires about 120 minutes), nor can it adapt to kinetic changes caused by variables such as ambient temperature and oscillation frequency. Second, the entire loading process is a black box, making it impossible to perceive key quality control parameters in real time, resulting in the coexistence of underloading and overloading risks. Research data shows (Guiu B, JVIR 2015) that an irinotecan loading rate below 90% can lead to a 35% decrease in local tumor drug concentration, directly affecting efficacy. Another study (DeBaere T, JVIR 2016) showed that excessive doxorubicin loading (>150 minutes) leads to a 22% increase in microsphere aggregation rate, posing a surgical risk of microcatheter blockage. This indicates that both underloading and overloading can affect drug efficacy.

[0004] Current technologies cannot provide effective real-time monitoring solutions for the physicochemical properties of different drugs. For drugs with inherent fluorescence properties, such as irinotecan, the decrease in free drug concentration during loading leads to a regular decay of the fluorescence signal in the mixture. This provides a physical basis for direct, real-time monitoring of loading rate using fluorescence spectroscopy, but current technologies have failed to utilize this principle for automated judgment. For drugs such as doxorubicin, which are mainly loaded through ion exchange mechanisms and do not possess strong fluorescence properties, the loading process and the risk of microsphere aggregation caused by overloading are more suitable for monitoring using bioimpedance spectroscopy. By measuring changes in the impedance of the mixture, the degree of loading can be indirectly inverted and early microsphere aggregation signals can be sensitively captured, but existing general-purpose timing devices completely lack this function.

[0005] The aforementioned technological gaps directly lead to the standardization dilemma of clinical procedures and the compliance challenges of data traceability. Manual timing is susceptible to fatigue and multitasking, resulting in a high error rate; manually recorded operation logs are prone to tampering and loss, making it difficult to meet the stringent requirements of modern medical quality management systems (such as GMP and FDA 21 CFR Part 11) for data integrity and process traceability. Therefore, the entire DEB technology field urgently needs a comprehensive solution that can overcome the limitations of single drugs, integrate multiple intelligent monitoring methods, and transform the loading process from experience-dependent to data-driven. Summary of the Invention

[0006] The present invention aims to provide an intelligent loading device and method, which adapts drugs with inherent fluorescence characteristics (such as irinotecan) through fluorescence monitoring and drugs that rely on ion exchange mechanisms and require early warning of aggregation risks (such as doxorubicin) through impedance monitoring, so as to dynamically determine and control the loading endpoint and achieve precise, safe and standardized loading.

[0007] A smart loading device for drug-loaded microspheres, comprising: The hull has hatches on its surface; At least one loading chamber, configured to be inserted into or removed from the housing via an openable door, is used to contain a mixture of drug and microspheres. The loading chamber surface has an injection port, which is an integrally formed elastic seal on the surface of the loading chamber. This elastic seal, in its natural state, closes the injection port to maintain the sealing of the loading chamber. It is configured to be pierced by an injection needle and to return to a sealed state due to its own elasticity after the needle is removed. The operator can inject the mixture through the injection port using a syringe for loading, or aspirate the mixture through the injection port after loading. Throughout the process, the mixture does not come into contact with the outside environment, and the entire injection device is pre-sterilized, thus ensuring the mixture remains in a sealed, sterile environment. A cover is also provided on the top surface of the loading chamber, which can be opened before the loading step to sterilize the interior of the loading chamber.

[0008] It also includes a real-time monitoring module, integrated into the shell or loading chamber, for non-contact or contact monitoring of signals reflecting the binding state of the drug and microspheres during loading. The real-time monitoring module is an optical property monitoring module or an electrical property monitoring module. The control module, electrically connected to the real-time monitoring module, is configured as follows: Access the pre-set quantitative relation database, which stores at least one target loading parameter determined by an adaptive algorithm model based on drug molecular weight and microsphere charge density. The target loading parameter includes theoretical loading time, target loading rate, and environmental control parameters. Based on the signals output by the real-time monitoring module, the real-time loading status of the drug is calculated or determined. The real-time loading status is compared with the target loading rate obtained from the database. When the real-time loading status reaches or exceeds the target loading rate, a loading completion signal is generated.

[0009] This device directly captures optical or electrical signals reflecting the binding state of the drug and microspheres through a real-time monitoring module, and the control module performs quantitative calculations and automatic judgments, thus replacing the traditional, crude operation mode that relies on manual timing and subjective observation. This design fundamentally eliminates inconsistencies in judgment caused by differences in operator experience and fatigue, ensuring high consistency and reliability of results for each loading process. Unlike fixed-duration countdown control methods, this device compares the real-time loading status with a preset target loading rate based on the physicochemical properties of the drug / microspheres (including molecular weight, charge density, etc.), and automatically terminates the loading process when the critical point of the optimal loading rate is reached. Therefore, on the one hand, it avoids the impact on treatment efficacy due to insufficient loading leading to a decrease in local drug concentration in the tumor; on the other hand, it precisely prevents surgical risks such as microsphere aggregation and microcatheter blockage caused by overloading, thus providing assurance for treatment safety and effectiveness at the device level. By introducing a pre-built quantitative relational database and adaptive algorithm model, this device can dynamically generate and execute personalized loading parameters (such as theoretical loading time and target loading rate) for different drug-microsphere combinations (such as irinotecan microspheres with fluorescent properties and doxorubicin microspheres with ion exchange-dependent loading). This function overcomes the one-size-fits-all systematic bias of existing general timing methods, realizing a leap from general-purpose equipment to intelligent special-purpose equipment, and significantly expanding the clinical application scope of this device.

[0010] A smart loading device for drug-loaded microspheres, the device further includes an environmental control module, which includes: The temperature control unit is used to regulate the temperature inside the loading chamber; Oscillating unit, used to drive the container to oscillate; The control module is also configured to dynamically control the operating parameters of the temperature control unit and the oscillation unit based on the real-time loading status and / or preset drug microsphere combination information.

[0011] The loading chamber is machined from a medical-grade aluminum alloy. The chamber body must be made of a high thermal conductivity material to act as a heat exchanger, ensuring efficient and uniform heat transfer between the temperature control unit and the internal drug mixture. The temperature control unit can be a semiconductor Pel patch, which is tightly attached to the outer wall of the loading chamber and secured with thermally conductive silicone grease. A high-precision temperature sensor (such as a PT1000 platinum resistance thermometer) is directly embedded within the loading chamber to monitor the temperature of the mixture. Based on sensor feedback, the control module controls the direction and magnitude of the current in the Pel patch, thereby regulating the temperature of the loading chamber. The oscillation unit is an oscillation platform powered by a linear motor. The motor receives commands from the control module and generates up-and-down oscillations at a specific frequency and amplitude. The binding of the drug to the microspheres is a complex physicochemical process, and its reaction rate and stability are highly dependent on ambient temperature and mixing uniformity. This solution, through the coordinated operation of the temperature control unit and the oscillation unit, creates a highly repeatable standardized environment for each loading, eliminating batch-to-batch variations introduced by environmental fluctuations and ensuring the consistency and reliability of the loading results. The environmental control module does not operate independently but receives instructions from the control module. The control module can directly access the optimal environmental parameters from the database based on pre-set drug microsphere combination information and dynamically adjust these parameters based on real-time feedback to optimize the loading process. For example, if the loading rate is detected to be too slow, the temperature can be appropriately increased to accelerate it; if uneven mixing is detected, the oscillation mode can be adjusted. This significantly improves the intelligence and efficiency of the loading process.

[0012] A smart loading device for drug-loaded microspheres includes a temperature control unit comprising several semiconductor Pellets and a temperature sensor. The semiconductor Pellets are attached to the outer wall of the loading chamber via thermally conductive silicone grease. The temperature sensor is embedded in the loading chamber and is used to directly monitor the temperature of the drug-microsphere mixture. The temperature sensor is a high-precision platinum resistance sensor. The oscillation unit includes a linear motor and an oscillation platform. An electric locking block is provided on the oscillation platform to fix the loading chamber. The linear motor is controlled by a control module to generate a specific frequency and amplitude to drive the oscillation platform to oscillate in the vertical direction.

[0013] This device employs an adaptive algorithm model based on drug molecular weight and microsphere charge density to pre-determine target loading parameters. This allows for precise and personalized customization of the loading process for different drug-microsphere combinations, fundamentally overcoming the systematic biases caused by existing general timing methods that ignore the specificity of drugs and microspheres. The device's real-time monitoring module acquires and analyzes optical or electrical signals reflecting the drug-microsphere binding state during loading, comparing them in real-time with the target loading rate in the database. This forms a dynamic feedback and closed-loop control intelligent judgment system, accurately capturing the critical point of loading completion. This effectively avoids the risk of drug concentration decrease due to insufficient loading or microsphere aggregation caused by overloading, ensuring the uniformity and reliability of loading quality. Ultimately, this approach elevates the operation process from one relying on manual experience to a data-driven, automated, standardized operating mode. While significantly reducing operational complexity and human error, it provides crucial equipment assurance for the reliability and efficacy consistency of drug-loaded microsphere chemoembolization.

[0014] A smart loading device for drug-loaded microspheres includes a loading chamber comprising at least a first loading chamber. A transparent observation window is located on the top of the first loading chamber and can be directly integrated into the chamber cover. The observation window is made of quartz glass, which has high light transmittance. Quartz glass was chosen because it has extremely high transmittance for both ultraviolet and visible light, and its fluorescence background is very low, thus not interfering with the measurement of drug fluorescence. The inner surface of the observation window is coated with a hydrophobic coating, which effectively prevents drug solution or microspheres from adhering to or remaining on the window surface, avoiding obstruction of the light path or interference with fluorescence signals due to drug drying and crystallization. The optical property monitoring module is a fluorescence spectral sensor. Its optical probe faces the transparent observation window and illuminates the interior of the first loading chamber, exciting light and receiving fluorescence signals from the mixture to analyze changes in drug concentration.

[0015] Before use, drug embolization microspheres require a certain loading time to fully integrate with the drug; the loading time varies depending on the drug and the microsphere. For drugs with inherent fluorescence properties (e.g., irinotecan), a fluorescence spectroscopy sensor emits excitation light of a specific wavelength (e.g., ultraviolet or blue light in the wavelength range of 350-400 nm) into the drug-microsphere mixture within the loading chamber. Once excited, the drug molecules emit a fluorescence signal of a specific wavelength. This fluorescence signal is received by a photodetector in the sensor. The control module reads the signal intensity (or spectral characteristics) acquired by the fluorescence spectroscopy sensor in real time. It has a pre-stored quantitative relationship model of "fluorescence intensity - drug concentration." As the microspheres adsorb the drug, the concentration of free drug in the solution decreases, causing the measured fluorescence signal intensity to weaken. The control module compares the continuously monitored fluorescence signal decay curve with a preset target loading rate threshold (e.g., the fluorescence intensity value corresponding to a 98% loading rate). When the real-time fluorescence signal intensity decays to and stabilizes at this preset threshold, the control module determines that the real-time loading status meets the completion conditions and generates a loading completion signal, triggering an audible and visual alarm to alert the operator.

[0016] A smart loading device for drug-loaded microspheres includes a control module specifically configured to: calculate the filtered fluorescence intensity signal in real time; compare the fluorescence intensity signal with a target fluorescence intensity threshold calculated based on the initial fluorescence intensity and the target loading rate; and when the fluorescence intensity signal is continuously lower than the fluorescence intensity threshold, and the real-time calculated loading rate is lower than a preset minimum loading rate threshold, the control environment regulation module performs one or more intervention actions, including increasing the frequency of the oscillation unit or adjusting the temperature setpoint of the temperature control unit. The control module does not directly use the raw signal collected by the fluorescence spectroscopy sensor, but first performs digital filtering to suppress instantaneous noise introduced by factors such as liquid surface fluctuations, microbubbles, and temporary obstruction of the light path by particulate matter. The control module calculates the signal level at theoretical completion based on the initial signal and a preset target loading rate (e.g., 98%), as a judgment threshold. When the filtered fluorescence intensity is lower than this target threshold, it indicates that loading should have ended from a concentration perspective; simultaneously, if the real-time loading rate is continuously lower than the set minimum threshold, it indicates that the process is stagnating. When both conditions are met, the system can determine that a "loading stagnation" anomaly has occurred, meaning that the loading process has not met expectations and has lost effective propulsion. Once a "loading stagnation risk" is detected, the system will proactively trigger an intervention mechanism, issuing instructions to the actuators (such as the environmental control module) to dynamically adjust the oscillation frequency and temperature. Increasing the oscillation frequency enhances fluid shear force, disrupts the local concentration boundary layer, and promotes collisions and mass transfer efficiency between drug molecules and microspheres; while appropriately increasing the temperature increases molecular thermal energy, overcomes energy barriers, and accelerates the adsorption kinetics process. Through a coherent technical path of "signal filtering—dual threshold judgment—dynamic parameter adjustment," the system constructs an intelligent closed-loop control strategy. Its core lies in real-time monitoring of the filtered fluorescence signal and comprehensively evaluating two key criteria: "whether the signal is below the target threshold" and "whether the loading rate is too low," thereby accurately diagnosing the loading stagnation state. Once an anomaly is detected, the system immediately and proactively adjusts key parameters such as the oscillation frequency or temperature to break the stagnation state and optimize the reaction process. This closed-loop control mechanism significantly enhances the robustness and success rate of the loading process, ensures the consistency and reliability of results, and achieves a leap from passive monitoring to proactive optimization management, fundamentally eliminating the risk of loading failure caused by environmental fluctuations or batch differences.

[0017] A smart loading device for drug-loaded microspheres includes a second loading chamber and an electrical property monitoring module that is a bioimpedance sensor. The bioimpedance sensor has at least two pairs of measuring electrodes embedded in the sidewall of the second loading chamber, with its measuring surface flush with the inner wall. A control module is configured to drive each electrode pair to perform impedance measurements in a time-division multiplexing manner and calculate an impedance value representing the overall state of the mixture based on the measurement results of each electrode pair. The bioimpedance sensor directly responds to changes in the physical state of the microspheres—particle size, distribution density, and aggregation tendency—by measuring changes in the impedance or dielectric constant of the mixture during loading. This is fundamentally different from optical methods that only monitor drug concentration in solution. This technology allows the system to capture early signals of abnormal microsphere aggregation in real time. According to data from DeBaere T, JVIR 2016, excessive loading of drugs such as doxorubicin can increase the microsphere aggregation rate by 22%, posing a risk of clogging microcatheters. This technical solution can provide timely warnings before this risk occurs (i.e. when the impedance value exceeds the preset threshold), providing an indispensable data foundation for subsequent active intervention, thereby fundamentally avoiding the surgical operation risks caused by microsphere aggregation.

[0018] Most drug-loaded microspheres (such as DC Bead and HepaSphere) adsorb charged drug molecules (such as doxorubicin) through ion exchange. This process alters the charge distribution and ion concentration on the microsphere surface and in the overall solution, causing a regular change in the overall impedance characteristics of the mixture. Bioimpedance sensors can indirectly determine the degree of drug loading with high sensitivity by tracking the dynamic trajectory of this electrical parameter. This allows the device of this invention to be precisely adapted to a large class of drug-microsphere combinations loaded via ion exchange, especially suitable for drugs that do not possess strong fluorescence properties and cannot be analyzed using fluorescence spectroscopy, thus greatly expanding the device's technological inclusivity and clinical application scope.

[0019] A smart loading device for drug-loaded microspheres includes a control module specifically configured to: calculate in real time the impedance value representing the overall state of the mixture and its impedance ratio relative to a reference impedance; compare the impedance ratio with a preset aggregation risk threshold; and when the impedance ratio reaches or exceeds the aggregation risk threshold, control the oscillation unit to execute one or more strong oscillation pulses to disperse the microspheres. By controlling the entire oscillation platform to execute short-duration, high-intensity strong oscillation pulses, powerful shear forces can be instantly and uniformly transmitted to every part of the mixture within the loading chamber, achieving efficient physical dispersal of aggregated microsphere clumps. Impedance monitoring and platform-level oscillation are seamlessly linked through the control module. Platform oscillation is the most direct and effective mechanical method for achieving global mixing. Once the system determines a risk (impedance exceeding the threshold), the most powerful intervention measure (platform strong oscillation) can be initiated within seconds, forming a rapid closed loop from risk perception to thorough handling. This automated and powerful intervention efficiency and speed far surpass the lagging and inefficient manual handling methods such as human observation and manual shaking, significantly improving the system's intelligence level and processing reliability.

[0020] A smart loading device for drug-loaded microspheres, the device further includes a human-machine interaction module, the human-machine interaction module includes: a touch screen, used to display a drug microsphere combination selection menu, a real-time loading rate curve and remaining estimated time; An audible and visual alarm is used to provide an audible and visual alert when a loading completion signal is generated.

[0021] The touchscreen provides a drug-microsphere combination selection menu, eliminating the need for operators to memorize complex loading parameters for different drugs; all initialization settings can be completed simply by selecting options. Essentially, this presents the vast quantitative database to the user in the most intuitive way. The touchscreen displays the loading rate curve and remaining estimated time in real time, transforming the microscopic drug combination process into clearly visible graphics and data. This allows medical staff to monitor the loading progress and status in real time without any guesswork. This transparent process monitoring not only enhances operator confidence but also facilitates operating room scheduling with its clearly defined "remaining estimated time," improving the planning and efficiency of the entire treatment process. When the control module generates a loading completion signal, an audible and visual alarm provides proactive notification. This multimodal (visual and auditory) alarm ensures timely and accurate reception of notifications even in noisy environments or when the operator's gaze is not focused on the equipment, effectively avoiding the risk of overloading due to failure to detect completion and ensuring that the loaded product is used clinically at the optimal time.

[0022] A smart drug-loaded microsphere loading device includes a control module configured to: before activating the real-time monitoring module, determine whether a pre-rinsing step completion confirmation signal based on a touchscreen has been received; if not received, the loading start function is locked. The control module adds a judgment node before initiating the core real-time monitoring and loading process, requiring the receipt of a pre-rinsing step completion confirmation signal based on the touchscreen. If not received, the loading start function is locked. The pre-rinsing step (typically rinsing the microspheres with sterile water or a specific solvent) is a crucial preparatory operation to remove the preservation solution inside the microspheres, activate the microspheres, and ensure successful subsequent drug loading. Skipping this step will directly lead to loading failure or a sharp drop in efficiency. This solution forces the operator to execute and confirm this step through software logic, eliminating the possibility of entire batches of microspheres being scrapped due to negligence or omission, ensuring that each loading begins under correct initial conditions, and guaranteeing the basic reliability and consistency of the loading results.

[0023] A smart drug-loaded microsphere loading device includes a data communication interface. A control module is configured to generate a structured operation log and upload it to an external system via the data communication interface. The operation log includes at least: drug microsphere combination, theoretical loading time, actual loading time, final loading status determination result, and environmental parameter curves during the loading process. The structured operation log generated by the control module comprehensively records the entire data chain from input (drug microsphere combination), process (theoretical / actual loading time, environmental parameter curves) to result (final loading status determination). This overcomes the drawbacks of traditional manual records being easily lost and altered, creating a detailed "electronic birth certificate" for each drug-loaded microsphere product used by patients. This log fully complies with regulatory requirements for the authenticity, completeness, and traceability of electronic records, providing indisputable objective evidence for hospitals to respond to Good Manufacturing Practice audits, drug regulatory inspections, and handle potential medical disputes, greatly improving the compliance and safety of medical practices.

[0024] A smart loading control method for drug-loaded microspheres, comprising the following steps: Receive instructions for selecting drug microsphere combinations; Based on a pre-set quantitative relational database, the target loading parameters corresponding to the combination are determined. The target loading parameters include theoretical loading time, target loading rate, and environmental control parameters. The environmental control module ensures that the environmental parameters of the loading chamber containing the drug and microsphere mixture meet the environmental control parameters. The real-time monitoring module is activated to acquire optical or electrical signals reflecting the binding state of the drug and microspheres in a non-contact or contact manner. Calculate the real-time loading status of the drug based on optical or electrical signals; The real-time loading status is compared with the target loading rate. When the real-time loading status reaches or exceeds the target loading rate, a loading completion signal is generated.

[0025] This method begins by receiving a selection instruction for the drug-microsphere combination, and then determines the corresponding target loading parameters based on a pre-defined quantitative relationship database. Therefore, the control logic of the entire loading process is dynamically generated according to the characteristics of the selected drug-microsphere combination, rather than using a fixed preset scheme. This method fundamentally overcomes the inherent limitations of existing technologies that ignore the specificity of drugs and microspheres and uniformly adopt fixed-duration schemes. Regardless of whether the drug to be loaded is irinotecan (requiring a shorter loading time) or doxorubicin (requiring a longer loading time), the system can automatically call upon the optimal process parameters that are suitable for it, thereby providing a personalized and precise control strategy for each loading, ensuring the accuracy of loading quality from the source.

[0026] The system not only sets ideal loading conditions in the initial stage, but also dynamically tracks and adjusts based on real-time feedback throughout the loading process. With the help of dynamic feedback and closed-loop decision-making mechanisms, it can promptly correct deviations caused by environmental fluctuations or batch differences in materials, ensuring that the loading process always proceeds along the optimal path. This achieves a qualitative leap from "ensuring the process occurs" to "guaranteeing the process is optimal," effectively improving the loading success rate and overall efficiency.

[0027] This method transforms the most critical step in the loading process—judging the timing of loading completion—which traditionally relies on human experience, into a standardized algorithm that calculates the loading status in real time based on optical or electrical signals and compares it with the target loading rate. This effectively eliminates inconsistencies in judgment caused by differences in operator experience and fatigue levels. Regardless of the operator, as long as this method is followed, highly consistent and stable loading results can be obtained. This ability to transform tacit knowledge into a standardized process is a key foundation for achieving standardized treatment across institutions and regions, and also provides the technical prerequisite for the automated execution and remote monitoring of this method.

[0028] The core advantage of this invention lies in its integration of two specific technical pathways—fluorescence monitoring and impedance monitoring—to construct a dynamic feedback intelligent loading system. Fluorescence is used to directly and accurately track the concentration decay of fluorescent drugs such as irinotecan, ensuring the loading rate meets standards. Simultaneously, impedance monitoring is used to indirectly monitor the loading process of non-fluorescent drugs such as doxorubicin and provide sensitive early warnings of microsphere aggregation risks. This addresses two major clinical challenges at the source: decreased efficacy due to insufficient loading and catheter blockage caused by overloading. It comprehensively upgrades operations that rely on manual experience into a data-driven, process-controllable, and traceable operational mode. Attached Figure Description

[0029] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0030] Figure 1 This is an internal perspective view of the overall device of the present invention; Figure 2 This is a schematic diagram of the housing of the present invention; Figure 3 This is a schematic diagram of the first loading chamber of the present invention; Figure 4 This is a schematic diagram of the loading chamber and oscillation unit of the present invention; Figure 5 This is a front view schematic diagram of the oscillation unit and loading chamber of the present invention; Figure 6 This is a schematic diagram of the fluorescence spectroscopy sensor of the present invention irradiating the first loading chamber; Figure 7 This is a schematic diagram of the electrode arrangement inside the second loading chamber of the present invention.

[0031] Figure descriptions: 1-Housing, 2-Loading chamber, 2a-Injection port, 3-Real-time monitoring module, 4-Touch screen, 5-Oscillation unit, 6-Audio-visual alarm, 7-Semiconductor Pellet patch, 11-Door, 21-First loading chamber, 21a-Transparent observation window, 22-Second loading chamber, 31-Fluorescence spectroscopy sensor, 31a-Optical probe, 32-Bioimpedance sensor, 52-Oscillation platform, 52b-Electrically operated block, 51-Linear motor, E1-First electrode, E2-Second electrode, E3-Third electrode, E4-Fourth electrode. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] The concepts involved in this application will first be described with reference to the accompanying drawings. It should be noted that the following descriptions of various concepts are only for the purpose of making the content of this application easier to understand and do not constitute a limitation on the scope of protection of this application; furthermore, the embodiments and features in the embodiments of this application can be combined with each other unless otherwise specified. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0034] Example 1: The hardware structure of the intelligent drug-loaded microsphere loading device in this embodiment includes: See attached document Figure 1 , 2 As shown in Figures 1 and 3, the housing 1 and the first loading chamber 21 are connected. The first loading chamber 21 is made of medical-grade aluminum alloy by CNC machining to ensure uniform wall thickness and good thermal conductivity. A door 11 is provided on the surface of the housing 1. Opening the door 11 allows the first loading chamber 21 to be placed inside the housing 1 and fixed in place. The first loading chamber 21 provided in this embodiment is approximately cylindrical and is used to contain a mixture of drug and microspheres. An injection port 2a is provided on the side wall of the first loading chamber 21. The injection port 2a is composed of a rubber sealing plug. This sealing plug has a cylindrical structure similar to a medical infusion bottle stopper, with its outer diameter slightly larger than the inner diameter of the injection port 2a. It is tightly embedded in the injection port 2a by pressing. In the non-injection state, the rubber sealing plug fills the entire injection port 2a to achieve a seal. During injection, the syringe needle punctures through the solid part of the rubber sealing plug. After the needle is withdrawn, the puncture channel of the rubber sealing plug will quickly contract and close under its own elasticity, restoring the sealing state.

[0035] This embodiment uses a fluorescence spectroscopy sensor 31 as the optical property monitoring module. (Refer to the appendix.) Figure 6 As shown, the fluorescence spectral sensor 31 is integrated non-contactly above the transparent observation window of the first loading chamber 21. Its optical probe 31a can emit LED light with a wavelength center of 370nm and can also serve as a photodetector with a detection range of 450-550nm. (See attached diagram.) Figure 4 As shown, the temperature control unit includes multiple semiconductor Peltier patches 7 tightly attached to the outer wall of the first loading chamber 21, and a high-precision platinum resistance temperature sensor directly immersed in the inner cavity of the first loading chamber 21. The Peltier patches 7 achieve thermal coupling with the outer wall of the first loading chamber 21 through thermally conductive silicone grease. (Refer to the attached diagram.) Figure 4 , 6 As shown, the oscillation unit 5 includes an oscillation platform 52 driven by a linear motor 51. The entire first loading chamber 21 is fixed on this platform, and the linear motor 51 can drive the platform to perform vertical reciprocating oscillation with an amplitude of 2 cm. The control module adopts an embedded microprocessor system, and the human-machine interaction module includes a touch screen 4 and an audible and visual alarm 6. The data communication interface integrates an Ethernet interface.

[0036] The control module's built-in quantitative relationship database establishes an adaptive algorithm model based on drug molecular weight and microsphere charge density. The core parameter relationships are as follows: The optimal loading time T op The calculation formula is as follows: ; Temperature control parameter T co The calculation formula is as follows: ; Oscillation frequency F o The calculation formula is as follows: ; Reference molecular weight (MW) ref =600Da, reference charge density CD ref = 3.0 μeq / g.

[0037] Among them, MW ir The molecular weight of irinotecan, CD ta T represents the charge density of the Tandem microspheres. b1 The base loading time, ΔT is the empirical temperature constant, T b2 Based on the base temperature, F b Based on the oscillation frequency, K t K is the temperature sensitivity coefficient. o These are the oscillation optimization coefficients.

[0038] Step 1: Parameter Initialization and Dynamic Calculation When the operator selects the "irinotecan-Tandem 100μm" combination via touchscreen 4, the control module retrieves the basic parameters from the database: Irinotecan molecular weight (MW) ir The charge density CD of the Tandem microspheres is 586.68 Da. ta It is 2.8 μeq / g, with a base loading time T. b1 The loading time is 30 minutes (consistent with the typical loading time for irinotecan), with a base temperature T. b2 The temperature is 25°C (which is consistent with the typical operating temperature of irinotecan), and the fundamental oscillation frequency F is... b 100 rpm, temperature sensitivity coefficient K t The oscillation optimization coefficient K is 0.85. o The value is 1.2. The temperature sensitivity coefficient K in the quantization relational database is... t and oscillation optimization coefficient K oThe parameters were determined through multivariate nonlinear regression analysis of a large amount of historical experimental data. This historical experimental data systematically examined the stability and loading kinetics of different drugs (with molecular weight as the key parameter) at different temperatures, and the loading efficiency and integrity of different microspheres (with charge density and physical properties as key parameters) under different oscillation intensities. Specifically, the temperature sensitivity coefficient was determined primarily based on the chemical stability data of the drugs and the temperature dependence of the loading reaction; the oscillation optimization coefficient was determined primarily based on the shear strength, aggregation tendency, and mass transfer efficiency of the microspheres under different oscillation conditions. These coefficients enable the adaptive algorithm model of this invention to go beyond simple linear relationships, more accurately capturing the complex characteristics of drug-microsphere combinations, thereby achieving truly personalized parameter optimization.

[0039] Once all values ​​are determined, the control module calculates personalized parameters in real time.

[0040] ; ; The ΔT in the formula is set to 10, which is an empirical constant determined based on engineering practice experience and the clinical operating temperature range.

[0041] ; Step 2: Setting up environmental parameters and starting the process After receiving the pre-rinse completion confirmation signal, the control module simultaneously activates the environmental control module. The control module sends a command to the temperature control unit, setting the target temperature to 24.8℃ (optimized based on irinotecan characteristics), and sends a command to the oscillation unit 5, setting the oscillation frequency to 120 rpm and the amplitude to 2 cm. Once the environmental parameters stabilize, the real-time monitoring module is activated.

[0042] Step 3: Signal Acquisition and Real-time Load Rate Calculation The control module performs the following calculations once per second: Read the raw fluorescence intensity signal I r (t), using moving average filtering. ; I(t) is the filtered fluorescence intensity value at time point t, and Ir(ti) is the original fluorescence intensity signal at time point ti.

[0043] Record the average fluorescence intensity I0 for the first 3 seconds at the start of loading. ; The theoretical fluorescence intensity I corresponding to a target loading rate of 98% t , ; Calculation of real-time loading rate L(t) ; Real-time calculation of loading rate V l , ; If V l When (t) is lower than 90% of the expected value, the oscillation frequency F is automatically adjusted. n , ; Where V e V represents the desired loading rate. a This represents the actual monitored loading rate. V e Based on the theoretical loading rate derived from a preset model or historical data, and considering the characteristics of irinotecan, a loading rate of 2% / minute is expected to be achieved under specific conditions.

[0044] Step 4: Double verification completes the judgment. The system employs a dual verification approach, combining dynamic model prediction with real-time monitoring. Predicted completion time T p , ; T c This is the current time.

[0045] The completion condition requires that L(t) > 97% and |T| be satisfied simultaneously. p -T o |≤6 minutes, loading rate V1(t)≤0.1% / minute.

[0046] Step 5: Process Termination and Data Recording After generating a loading completion signal, the oscillation unit 5 and the temperature control unit gradually stop operating. Then, the audible and visual alarm 6 is triggered, a structured operation log is generated and uploaded, and the log includes the drug microsphere combination, the calculated theoretical loading time (25.2 minutes), the actual loading time, the final loading status determination result, and the environmental parameter curves during the loading process.

[0047] The following calculation uses substitution, assuming the original fluorescence signal collected continuously for 5 seconds is I. r (15:00)=325、I r (15:01)=327、I r (15:02)=323、I r (15:03)=326、I r (15:04) = 324, The filtered signal, .

[0048] Given that the initial fluorescence intensity I0 = 1000, and the target intensity I... t =1000 × 0.08 = 80, , Assuming the expected loading rate V e =2.0% / minute, actual loading rate V a =1.5% / minute (calculated based on recent load rate changes), then The system increases the oscillation frequency from 120 rpm to 139 rpm.

[0049] 28 minutes after loading began, the current real-time load rate L(28:00) = 97.8%, and the current load rate V a =0.3% / minute, current time 28 minutes.

[0050] Predicted completion time .

[0051] Where L(t) = 97.8% ≥ 98%, |T p -T o If |=5.03 minutes≤6 minutes, dL(t) / dt=0.3% / minute≤0.1% / minute, all system judgment conditions are met, loading is complete, and a loading signal is generated.

[0052] The generated structured operation log includes: Drug microsphere combination: Irinotecan-Tandem 100μm; Theoretical loading time: 30.3 minutes (based on dynamic model calculation); Actual loading time: 28.0 minutes; Final loading status: Successful, loading rate ≥ 97.8%; Environmental parameter curves: Recorded the complete process of temperature (24.8℃) and oscillation frequency (120→139rpm) changes.

[0053] This example fully demonstrates the entire process from parameter calculation, real-time monitoring, dynamic adjustment to final judgment, fully reflecting the invention's technical creativity and practical value.

[0054] Example 2: The hardware structure of the intelligent drug-loaded microsphere loading device in this embodiment includes: The housing 1 and the second loading chamber 22 are provided. The second loading chamber 22 is made of medical-grade aluminum alloy by CNC machining, and electrode pairs are integrated on its inner wall. The second loading chamber 22 provided in this embodiment is generally a cylindrical cavity used to contain a mixture of drug and microspheres. An injection port 2a is provided on the side wall of the second loading chamber 22. The sealing structure of the injection port 2a can be an independent rubber sealing plug. The sealing plug is a cylindrical structure similar to the stopper of a medical infusion bottle, and its outer diameter is slightly larger than the inner diameter of the injection port 2a. It is tightly embedded in the injection port 2a by pressing. In the non-injection state, the rubber sealing plug fills the entire injection port 2a to achieve a seal. During injection, the syringe needle punctures through the solid part of the rubber sealing plug. After the needle is withdrawn, the puncture channel of the rubber sealing plug will quickly contract and close under its own elasticity, restoring the sealing state.

[0055] This embodiment uses a bioimpedance sensor 32 as the electrical property monitoring module. The environmental control module includes a semiconductor Peltier patch 7 and a high-precision platinum resistance temperature sensor. The oscillation unit 5 includes an oscillation platform 52 driven by a linear motor 51. The control module uses an embedded microprocessor system, and the human-machine interface module includes a touch screen 4 and an audible and visual alarm 6. The data communication interface integrates an Ethernet interface. Optionally, an electrode arrangement scheme for the bioimpedance sensor 32 is provided. (See attached diagram.) Figure 7 As shown, four independent rectangular electrodes are embedded at equal intervals on the circumference of the side wall of the second loading chamber 22, forming two pairs of measuring electrodes. The first electrode E1 and the third electrode E3 constitute the first measuring pair, and the second electrode E2 and the fourth electrode E4 constitute the second measuring pair. The electrodes are made of medical-grade platinum. These electrode materials are chemically stable, have good conductivity, low contact impedance, and are oxidation-resistant. The electrodes are integrated with the metal wall (medical-grade aluminum alloy) of the second loading chamber 22 through micro-embedding injection molding or laser welding processes, ensuring that the measuring surface of the electrode is flush with the inner wall of the chamber, forming a smooth, dead-angle-free flow channel surface to prevent microspheres from being trapped or adhering to the wall. The control module is configured to drive the two pairs of electrodes in a time-division multiplexing manner. At time T1, the first electrode E1 is used as the excitation electrode and the third electrode E3 is used as the receiving electrode to perform an impedance measurement. At time T2, the second electrode E2 is used as the excitation electrode and the fourth electrode E4 is used as the receiving electrode to perform another impedance measurement. The control module takes the average or weighted average of these two (or multiple) measurement results as the effective impedance at that time point. This arrangement can eliminate the measurement deviation caused by the temporary uneven distribution or local aggregation of microspheres, and obtain a global impedance that can represent the state of the mixture in the entire chamber, making the data more reliable.

[0056] The control module's built-in quantitative relationship database establishes an adaptive algorithm model based on drug molecular weight and microsphere charge density. The optimal loading time, temperature control parameters, and oscillation frequency calculation formulas in this model are the same as in Example 1. Furthermore, an aggregation risk threshold Z is added.t calculate, Z0 is the reference impedance, which is the average initial impedance measured after the system has stabilized during the initial loading process. It represents the background electrical characteristics of the drug-microsphere mixture under normal and well-dispersed conditions; K a To determine the aggregation risk factor, extensive preliminary experiments were conducted to analyze the impedance change curves of different drug-microsphere combinations during the loading process. Retrospective analysis revealed that when the impedance increased to 1.4 to 1.6 times the initial value, significant microsphere aggregation was observed under a microscope, and these aggregates posed a risk of clogging microcatheters. Literature (DeBaereT, JVIR 2016) indicated that "overloading with doxorubicin led to a 22% increase in microsphere aggregation rate." This clinical conclusion supports the quantitative relationship between impedance change and aggregation rate, helping to determine a safe warning threshold. To ensure absolute safety, a relatively conservative coefficient with room for error was chosen. Setting it to 1.5 means that the system provides early warning and intervention before aggregation causes irreversible consequences (such as the formation of hard clumps). The reference molecular weight (MW) in the overall formula... ref =600Da, reference charge density CD ref =3.0 μeq / g, K a =1.5.

[0057] Step 1: Parameter Initialization and Dynamic Calculation The control module's built-in quantitative relationship database establishes an adaptive algorithm model based on drug molecular weight and microsphere charge density, with the core parameter relationships being the same as in Example 1. The molecular weight (MW) of doxorubicin is 543.52 Da, the charge density (CD) of the DCBead microspheres is 3.2 μeq / g, and the baseline loading time (T) is... b1 The time is 120 minutes (consistent with the typical loading time of doxorubicin), with a base temperature of T. b2 The temperature is 4℃ (which is consistent with the usual operating temperature of irinotecan), and the fundamental oscillation frequency F b 80 rpm, temperature sensitivity coefficient K t The oscillation optimization coefficient K is 1.25. o The value is 0.9, and the clustering coefficient is 1.5.

[0058] Substitute into the calculation, ; ; .

[0059] Step 2: Reference Impedance Measurement and Environment Setup After receiving the pre-rinse completion confirmation signal, the control module continuously collects impedance data within 5-10 minutes after loading begins, calculating the average value as the reference impedance Z0. At this time, the measured reference impedance Z0 = 1254Ω, and the aggregation risk threshold is set to Z. t =1.5Z0=1881Ω. Then, a command is sent to the temperature control unit with a target temperature of 2.8℃, and a command is sent to the oscillation unit 5 with an oscillation frequency of 72rpm and an amplitude of 2cm. After the environmental parameters stabilize, the real-time monitoring module is started.

[0060] Step 3: Real-time detection and signal processing The control module performs the following calculations once per second: Read the original impedance signal Z r (t), using moving average filtering. ; Z(t) is the impedance value at time t, and Zr(ti) is the impedance value at time ti.

[0061] Real-time impedance ratio R z (t) Calculation, ; When R z (t) < 1.2 indicates the normal loading phase; when 1.2 ≤ R z (t) < 1.5 indicates the early warning stage, when R z (t)≥1.5 indicates the clustering risk stage.

[0062] Step 4: Dynamic Intervention and Parameter Adjustment Eighty-five minutes after loading began, there was a risk of aggregation inside the drug-loaded microspheres. Impedance data for 5 consecutive seconds was Z. r (85:00) = 1820, Z r (85:01) = 1850, Z r (85:02) = 1890, Z r (85:03) = 1920, Z r (85:04) = 1950. The filtered impedance is... ; Impedance ratio calculation Among them, 1.504 > 1.5 triggers a cluster risk warning; The system immediately triggered a strong oscillatory pulse intervention. The strong oscillatory pulse parameters were 200 rpm, 3 cm amplitude, and 15 seconds duration, with 3 cycles of strong oscillation (15 seconds of oscillation, 5 seconds of pause, repeated 3 times). Two minutes after the intervention, Z(87:04) = 1420 Ω, and the intervention effect was evaluated using R. z(87:04)=1420 / 1254=1.132<1.3, risk is eliminated.

[0063] Step 5: Complete the judgment and verification 108 minutes after loading began, the real-time impedance Z(108:00) = 1350Ω, and the impedance ratio R... z (108:00)=1350 / 1250=1.077, cumulative loading time 108 minutes, historical intervention record shows that a strong oscillation intervention was performed at 85 minutes.

[0064] The actual loading time was ≥ 95% of the theoretical optimal time, thus satisfying the condition.

[0065] Calculating the impedance fluctuation over the past 10 minutes yields a maximum impedance of 1380Ω (at 98 minutes) and a minimum impedance of 1320Ω (at 105 minutes). The fluctuation rate is (1380-1320) / 1350 = 4.44% < 5%, which meets the requirements. Furthermore, the current impedance ratio R... z =1.077, the system judges that all conditions are met, and the loading of drug-loaded microspheres is complete.

[0066] Step 6: Process Termination and Data Recording After generating the loading completion signal, the oscillation unit 5 and the temperature control unit stop working, the audible and visual alarm 6 is triggered, and a structured operation log is generated and uploaded.

[0067] Operation log content: Drug microsphere assembly: Doxorubicin-DCBead; Theoretical loading time: 112 minutes (calculated based on dynamic model); Actual loading time: 108 minutes; Final loading status: Successful, with one aggregation intervention triggered during the process; Cluster intervention records; Intervention time: 85 minutes; Pre-intervention impedance: 1886Ω; Impedance after intervention: 1420Ω; Intervention effect: Impedance decreased by 24.7%, successfully eliminating the risk of aggregation; The environmental parameter curves fully record the changes in temperature (2.8±0.3℃) and oscillation frequency (72→200→72rpm), while the impedance change curves record the complete trajectory from the reference 1254Ω to the final 1350Ω.

[0068] This embodiment fully demonstrates the unique value of the bioimpedance sensor 32 in monitoring the aggregation state of microspheres, and how the system ensures the safety and reliability of the loading process of easily aggregated drugs such as doxorubicin through active intervention, fully reflecting the technical advantages and creativity of the present invention in solving specific clinical problems.

[0069] The embodiments and / or implementation methods described above are merely preferred embodiments and / or implementation methods for implementing the technology of the present invention, and are not intended to limit the implementation methods of the technology of the present invention in any way. Any person skilled in the art may make some modifications to other equivalent embodiments without departing from the scope of the technical means disclosed in the content of the present invention, but these should still be regarded as the technology or embodiments that are substantially the same as the present invention.

[0070] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are only preferred embodiments of this application. It should be noted that due to the limitations of written expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A smart loading device for drug-loaded microspheres, comprising: The shell (1) has a hatch (11) on its surface. At least one loading chamber (2) is configured to be inserted into or removed from the housing (1) through the openable door (11) for containing a mixture of drug and microspheres. The surface of the loading chamber (2) is provided with an injection port (2a). The injection port (2a) is an elastic seal integrally formed on the surface of the loading chamber (2). The elastic seal closes the injection port (2a) in its natural state to maintain the sealing of the loading chamber (2). It is configured to be pierced by an injection needle and to restore the sealing state by its own elasticity after the needle is pulled out. The feature is that it also includes a real-time monitoring module (3), which is integrated into the shell (1) or loading chamber (2) for non-contact or contact monitoring of signals reflecting the binding state of the drug and microspheres during loading. The real-time monitoring module (3) is an optical property monitoring module or an electrical property monitoring module. The control module, electrically connected to the real-time monitoring module (3), is configured as follows: Access a pre-set quantitative relationship database, which stores at least one target loading parameter determined by an adaptive algorithm model based on drug molecular weight and microsphere charge density. The target loading parameter includes theoretical loading time, target loading rate, and environmental control parameters. Based on the signal output by the real-time monitoring module, the real-time loading status of the drug is calculated or determined. The real-time loading status is compared with the target loading rate obtained from the database. When the real-time loading status reaches or exceeds the target loading rate, a loading completion signal is generated.

2. The intelligent loading device for drug-loaded microspheres according to claim 1, characterized in that, The device further includes an environmental control module, which comprises: A temperature control unit is used to regulate the temperature inside the loading chamber (2); Oscillating unit (5) is used to drive the container to oscillate; The control module is also configured to dynamically control the operating parameters of the temperature control unit and the oscillation unit (5) based on the real-time loading status and / or the preset drug microsphere combination information.

3. The intelligent loading device for drug-loaded microspheres according to claim 2, characterized in that, The temperature control unit includes several semiconductor PAL patches (7) and a temperature sensor. The semiconductor PAL patches (7) are attached to the outer wall of the loading chamber (2) with thermally conductive silicone grease. The temperature sensor is embedded in the loading chamber (2) and is used to directly monitor the temperature of the drug and microsphere mixture. The temperature sensor is a high-precision platinum resistance sensor. The oscillation unit (5) includes a linear motor (51) and an oscillation platform (52). The oscillation platform (52) is provided with an electric locking block (52b) to fix the loading chamber (2). The linear motor (51) is controlled by the control module to generate a specific frequency and amplitude to drive the oscillation platform (52) to oscillate in the vertical direction.

4. The intelligent loading device for drug-loaded microspheres according to claim 2, characterized in that, The loading chamber (2) includes a first loading chamber (21), and a transparent observation window (21a) is provided on the top of the first loading chamber (21). The transparent observation window (21a) is made of quartz glass and its inner surface is coated with a hydrophobic coating. The optical property monitoring module is a fluorescence spectral sensor (31). The fluorescence spectral sensor (31) is equipped with an optical probe (31a). The optical probe (31a) faces the transparent observation window (21a) and illuminates the inside of the first loading chamber (21).

5. The intelligent loading device for drug-loaded microspheres according to claim 4, characterized in that, The control module is specifically configured to: calculate the filtered fluorescence intensity signal in real time; compare the fluorescence intensity signal with the target fluorescence intensity threshold calculated based on the initial fluorescence intensity and the target loading rate; when the fluorescence intensity signal is continuously lower than the target fluorescence intensity threshold and the real-time calculated loading rate is lower than the preset minimum loading rate threshold, control the environmental control module to perform one or more intervention actions, the intervention actions including increasing the frequency of the oscillation unit (5) or adjusting the temperature setting value of the temperature control unit.

6. The intelligent loading device for drug-loaded microspheres according to claim 1, characterized in that, The loading chamber (2) includes a second loading chamber (22), and the electrical property monitoring module is a bioimpedance sensor (32). The electrodes of the bioimpedance sensor (32) are embedded in the side wall of the second loading chamber (22).

7. The intelligent loading device for drug-loaded microspheres according to claim 6, characterized in that, The control module is specifically configured to: calculate in real time the impedance value representing the overall state of the mixture and the impedance ratio relative to the reference impedance; compare the impedance ratio with a preset aggregation risk threshold; and when the impedance ratio is detected to reach or exceed the aggregation risk threshold, control the oscillation unit (5) to execute one or more strong oscillation pulses to disperse the microspheres.

8. The intelligent loading device for drug-loaded microspheres according to claim 1, characterized in that, The device also includes a human-computer interaction module, which includes a touch screen (4) for displaying a drug microsphere combination selection menu, a real-time loading rate curve and remaining estimated time. An audible and visual alarm (6) is used to provide an audible and visual prompt when the loading completion signal is generated.

9. The intelligent loading device for drug-loaded microspheres according to claim 1, characterized in that, The device also includes a data communication interface. The control module is configured to generate a structured operation log and upload the log to an external system through the data communication interface. The operation log includes at least: drug microsphere combination, theoretical loading time, actual loading time, final loading status determination result, and environmental parameter curves during the loading process.

10. A method for intelligent loading and control of drug-loaded microspheres, characterized in that, The method includes the following steps: Receive instructions for selecting drug microsphere combinations; Based on a pre-set quantitative relational database, the target loading parameters corresponding to the combination are determined. The target loading parameters include theoretical loading time, target loading rate, and environmental control parameters. The environmental control module controls the environmental parameters of the loading chamber (2) containing the drug and microsphere mixture to reach the environmental control parameters. The real-time monitoring module is activated to acquire optical or electrical signals reflecting the binding state of the drug and microspheres in a non-contact or contact manner. The real-time loading status of the drug is calculated based on the optical or electrical signals. The real-time loading status is compared with the target loading rate. When the real-time loading status reaches or exceeds the target loading rate, a loading completion signal is generated.

Images