Noninvasive full-closed-loop artificial pancreas system

Abstract

The invention belongs to the field of medical instruments, and relates to a non-invasive full-closed-loop artificial pancreas system which comprises a glucose optical detection module, a signal processing and analyzing module and a double-channel injection module which are sequentially connected, and the glucose optical detection module is used for emitting laser to a to-be-detected object to excite and generate a Raman spectrum signal; the blood glucose value acquisition module is also used for acquiring an actual blood glucose value of the to-be-detected object according to the Raman spectrum signal and transmitting the actual blood glucose value to the signal processing and analyzing module; the signal processing and analyzing module is used for analyzing the blood glucose change trend in a future preset time period based on the actual blood glucose value so as to generate a blood glucose reference trajectory; the signal processing and analyzing module is further used for controlling two injection channels of the double-channel injection module to be opened according to the blood glucose reference track so as to inject corresponding medicine to the to-be-detected object. According to the application, the functions of noninvasive real-time detection of the blood glucose level and accurate quantitative infusion of the corresponding medicine so as to intelligently control the blood glucose level can be achieved.

Description

Technical Field

[0001] This application belongs to the field of medical devices, and more specifically, relates to a non-invasive fully closed-loop artificial pancreas system. Background Art

[0002] The conventional treatment method for diabetes is insulin injection. However, the traditional insulin injection method usually requires patients to frequently monitor blood glucose levels and manually inject insulin according to the monitoring results, which has problems of complex operation and unstable treatment effect. With the progress of technology, traditional treatment methods have gradually developed towards intelligence and automation. Among them, the artificial pancreas system, as an advanced solution, has gradually become the focus of research.

[0003] Existing artificial pancreas systems are mostly large-scale blood glucose management systems that combine two independent devices, namely a continuous blood glucose monitoring device and an insulin infusion device. Such systems have the following problems: First, blood glucose monitoring usually requires invasive operations, such as using a needle-type blood glucose meter to monitor blood glucose, which poses risks of patient discomfort and easy infection; Second, although existing artificial pancreas systems involve certain closed-loop control algorithms, there are still problems of response delay and insufficient monitoring accuracy when realizing automatic adjustment of insulin dosage; Third, traditional insulin infusion devices are relatively large in size, which causes inconvenience and discomfort to patients during use. Moreover, the operation of blood glucose monitoring devices and insulin infusion devices is complex, and they need to be frequently replaced during long-term use, which affects patient compliance.

[0004] In view of the above technical problems, there is an urgent need for a solution for regulating blood glucose levels that is efficient, comfortable, and low-cost. Summary of the Invention

[0005] Aiming at the defects of the prior art, the purpose of this application is to provide a non-invasive fully closed-loop artificial pancreas system, aiming to solve the problems of insufficient monitoring accuracy and complex devices in existing artificial pancreas systems.

[0006] To achieve the above purpose, this application provides a non-invasive fully closed-loop artificial pancreas system, including a glucose optical detection module, a signal processing and analysis module, and a dual-channel injection module that are connected in sequence. The glucose optical detection module is used to emit laser light towards the object to be measured to excite and generate Raman spectral signals, and is also used to obtain the actual blood glucose value of the object to be measured based on the Raman spectral signals and transmit it to the signal processing and analysis module; the signal processing and analysis module is used to analyze the blood glucose change trend within a preset future time period based on the actual blood glucose value to generate a blood glucose reference trajectory; the signal processing and analysis module is also used to control the two injection channels of the dual-channel injection module to be opened respectively according to the blood glucose reference trajectory to inject corresponding drugs into the object to be measured.

[0007] Further, the glucose optical detection module includes a laser source, a galvanometer, and an input optical module, a static interference unit, an output grating array, a compound parabolic concentrator, an image sensing unit, and a data processing unit arranged in sequence along the optical path; the laser emitted by the laser source is reflected by the galvanometer and then vertically enters the object to be measured to detect blood glucose concentration. The input optical module is used to receive the blood glucose concentration signal emitted from the object to be measured and transmit it to the static interference unit to extract spectral features. The output grating array is used to perform spectral splitting on the composite light with the extracted spectral features. The compound parabolic concentrator is used to focus the optical signal after spectral splitting into the image sensing unit for photoelectric conversion, and the electrical signal after photoelectric conversion is input into the data processing unit for analysis to obtain the glucose concentration value of the object to be measured.

[0008] Furthermore, the laser source uses a dual-wavelength pulsed laser, with wavelengths of 780 nm - 790 nm and 1054 nm - 1074 nm respectively, and uses time-division multiplexing technology for alternating excitation; the pulsed laser emits laser in pulses with a frequency of 10 kHz, and the single-pulse energy ≤ 5 mJ.

[0009] Furthermore, the static interference unit includes a silicon nitride photonic crystal waveguide, a Mach-Zehnder interferometer, and a metasurface grating arranged coaxially in sequence along the optical path, and the surface of the silicon nitride photonic crystal waveguide is modified with a gold nanocone array.

[0010] Furthermore, the input optical module includes an adaptive collimating lens, a beam splitter, and a micro galvanometer arranged in sequence along the optical path. The adaptive collimating lens is used to adjust the focusing depth of the beam emitted from the object to be measured based on the skin color characteristics of the object to be measured, and the beam splitter is used to introduce the beam with the adjusted focusing depth into the micro galvanometer for high-speed collaborative deflection.

[0011] Furthermore, the compound parabolic concentrator is an asymmetric compound parabolic concentrator, and its acceptance angle is not less than 60 °C; and / or, the light incident surface of the compound parabolic concentrator is coated with a gold-titanium dioxide composite film.

[0012] Further, the signal processing and analysis module includes a closed-loop control algorithm unit. The closed-loop control algorithm unit is used to output an updated rate weight of dynamic insulin according to the growth rate and decline rate of the actual blood glucose value in the past T1 time period input; it is also used to output an updated rate weight of basal insulin based on the percentage of the actual blood glucose value higher than the blood glucose set value and the percentage of the actual blood glucose value lower than the blood glucose set value in the past T2 time period input, where T2 > T1; in the above process, the following state space equation is established:

[0013] Wherein,x i-1 represents the actual blood glucose value at the previous moment, x i represents the blood glucose value at the current moment, u i represents the insulin injection amount at that moment, A represents the weight coefficient matrix of the blood glucose value, B represents the weight coefficient matrix of the insulin injection dose, C Y represents the coefficient matrix of the historical blood glucose value, C V represents the blood glucose change rate calculation matrix, y i is the blood glucose historical data after weight correction, v i is the weight of the blood glucose change rate.

[0014] Further, the closed-loop control algorithm unit can perform rolling optimization on the blood glucose reference trajectory based on the actual blood glucose value obtained at the next moment. The cost function used in the rolling optimization is:

[0015] wherein, represents the loss function, N y represents the prediction space, N u represents the control interval, Q ( v k ) represents the blood glucose weight matrix, represents the blood glucose prediction error matrix, represents the insulin weight matrix, represents the blood glucose rate weight matrix, is the insulin injection dose matrix, is the insulin injection rate matrix.

[0016] Further, a temperature sensor and an acceleration sensor are integrated on the glucose optical detection module. The temperature sensor is used to detect the temperature of the object to be measured and transmit it to the signal processing and analysis module. The acceleration sensor is used to detect the exercise intensity of the object to be measured and transmit it to the signal processing and analysis module. The signal processing and analysis module is used to correct the Raman peak position shift based on the temperature, and is also used to adjust the priority of injecting glucagon by the dual-channel injection module (1) according to the exercise intensity.

[0017] Further, the non-invasive fully-closed-loop artificial pancreas system further includes a housing and a hollow microneedle array module. The glucose optical detection module, the signal processing and analysis module, and the dual-channel injection module are sequentially integrated and arranged in the housing. The output port of the dual-channel injection module is communicated with the input port of the hollow microneedle array module, and the output port of the hollow microneedle array module extends through the housing to the outside thereof.

[0018] Generally speaking, compared with the prior art by the above technical solutions conceived in the present application, the following beneficial effects are achieved: (1) Through the glucose optical detection module based on Raman spectroscopy technology, the present application can detect the blood glucose level of patients in real time and accurately without blood extraction, providing a more comfortable monitoring experience. At the same time, based on the monitoring results, the dual-channel injection module precisely delivers insulin or other corresponding drugs in advance, reducing the pain of patients and improving the treatment compliance.

[0019] (2) Combining the closed-loop control algorithm, the present application can automatically analyze the blood glucose level and accurately control insulin infusion according to the monitoring results, can respond to blood glucose changes in real time, automatically adjust the drug injection dose, realize the automated management of diabetic patients, and greatly reduce the burden on patients.

[0020] (3) Each module of the present application is connected by a buckle or a cable, and has a modular and replaceable structure, so that the system has good scalability and maintainability, and can replace or upgrade a single module as needed, extending the service life of the device and reducing the maintenance cost.

[0021] (4) The signal processing and analysis module of the present application can integrate various physiological information sensors such as an acceleration sensor and a temperature sensor, not only can monitor the blood glucose level, but also can detect information such as the daily activities and fall risk of patients in real time, providing more comprehensive health management and further improving the intelligent level of the system. Description of the Drawings

[0022] Figure 1 is a schematic diagram of the non-invasive fully-closed-loop artificial pancreas system provided by the embodiment of the present application; Figure 2 is a schematic diagram of the glucose optical detection module provided by the embodiment of the present application; Figure 3 is a schematic diagram of the structure of the coupled optical module provided by the embodiment of the present application; Figure 4 is a schematic diagram of the structure of the static interference unit provided by the embodiment of the present application; Figure 5 is a frame diagram of the closed-loop control algorithm model of the non-invasive fully-closed-loop artificial pancreas system provided by the embodiment of the present application; Figure 6It is a schematic diagram showing the blood glucose changes of a diabetic patient in a day provided by an embodiment of the present application; Figure 7 It is a schematic diagram showing the change of blood glucose density within a day provided by an embodiment of the present application; Figure 8 It is a structural diagram of a micro wearable artificial pancreas system provided by an embodiment of the present application; Figure 9 It is an exploded structural schematic diagram of a micro wearable artificial pancreas system provided by an embodiment of the present application; Figure 10 It is a structural schematic diagram of a hollow microneedle array module provided by an embodiment of the present application; Figure 11 It is a structural schematic diagram of a dual-channel injection module provided by an embodiment of the present application; Figure 12 It is a structural schematic diagram of a dual-channel injection module with a replaceable medicine cartridge provided by an embodiment of the present application.

[0023] In all the drawings, the same reference numerals are used to represent the same elements or structures, where: 1 - Dual-channel injection module, 11 - Injection outlet, 12 - Catheter, 13 - Medicine cartridge, 14 - Micro pump, 15 - Drug filling channel, 2 - Signal processing and analysis module, 3 - Glucose optical detection module, 31 - Laser source, 32 - Galvanometer, 33 - Object to be measured; 34 - Coupling optical module, 341 - Adaptive collimating lens, 342 - Bias beam splitter, 343 - Micro galvanometer; 35 - Static interference unit, 351 - Silicon nitride photonic crystal waveguide, 352 - Mach-Zehnder interferometer, 353 - Metasurface grating; 36 - Coupling out grating array, 37 - Compound parabolic concentrator, 38 - Image sensor, 39 - Data processing unit, 4 - Outer shell, 5 - Hollow microneedle array module, 51 - Drug inlet. Detailed implementation manners

[0024] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0025] The term "and / or" in this article is a relationship describing associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The symbol " / " in this article represents an "or" relationship between associated objects, for example, A / B represents A or B.

[0026] The terms "first" and "second" in the specification and claims herein are used to distinguish different objects rather than to describe a specific order of the objects. For example, a first response message and a second response message are used to distinguish different response messages rather than to describe a specific order of the response messages.

[0027] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0028] In the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. For example, multiple processing units refer to two or more processing units, etc.; multiple elements refer to two or more elements, etc.

[0029] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0030] The present embodiment provides a non-invasive fully closed-loop artificial pancreas system, comprising a glucose optical detection module, a signal processing and analysis module, and a dual-channel injection module connected in sequence, wherein the glucose optical detection module is used to emit a laser toward an object to be tested to excite and generate a Raman spectral signal, and is also used to obtain the actual blood glucose value of the object to be tested according to the Raman spectral signal and transmit it to the signal processing and analysis module; the signal processing and analysis module is used to analyze the blood glucose change trend within a preset time period in the future based on the actual blood glucose value to generate a blood glucose reference trajectory; the signal processing and analysis module is also used to control the two injection channels of the dual-channel injection module to be opened respectively according to the blood glucose reference trajectory to inject insulin or glucagon into the object to be tested.

[0031] Specifically, Figure 1 As shown, it is a schematic diagram of the non-invasive fully closed-loop artificial pancreas system provided in this embodiment (hereinafter referred to as the artificial pancreas system), and the artificial pancreas system includes a glucose optical detection module based on an optical detection method, a signal processing and analysis module, and a dual-channel injection module. Glucose information in body fluids is detected by the glucose optical detection module, and the glucose optical detection module sends the Raman spectrum signal carrying the blood glucose concentration to the signal processing and analysis module in the form of a wire or an electrode contact. The signal processing and analysis module analyzes and calculates the Raman spectrum signal to obtain the drug dose required to regulate blood glucose, and transmits the drug dose information to the dual-channel injection module through the electrode contacts. The dual-channel injection module includes a micropump and a hollow microneedle array. When the micropump corresponding to each channel is turned on, a certain dose of the corresponding drug calculated is injected into the hollow microneedle array.

[0032] In this embodiment, as Figure 2 shown, the aforementioned glucose optical detection module includes a laser source 31, a galvanometer 32, and an input optical module 34, a static interference unit 35, an output grating array 36, a compound parabolic concentrator 37, an image sensing unit 38, and a data processing unit 39 arranged in sequence along the optical path. An object to be measured 33 is located on the optical path between the galvanometer 32 and the input optical module 34; the laser emitted by the laser source 31 is reflected by the galvanometer 32 and then perpendicularly enters the object to be measured 33 to detect glucose molecules. The input optical module 34 is used to receive the pulse signal emitted from the object to be measured 33 and transmit it to the static interference unit 35 to extract spectral features. The output grating array 36 is used to perform spectral splitting on the composite light with the extracted spectral features. The compound parabolic concentrator 37 is used to focus the optical signal after spectral splitting into the image sensing unit 38 for photoelectric conversion, and the electrical signal after photoelectric conversion is input into the data processing unit 39 to analyze the glucose concentration value of the object to be measured.

[0033] In this embodiment, the aforementioned laser source uses a dual-wavelength pulsed laser, with wavelengths of 780 nm to 790 nm and 1054 nm to 1074 nm respectively. For example, a pulsed laser with two wavelengths of 785 nm and 1064 nm is used, and the object to be measured (i.e., skin tissue) is alternately excited through time-division multiplexing technology. The 785-nm laser is targeted at a depth of 0.1 mm to 0.3 mm in the epidermis layer, which can excite glucose molecules in the epidermal interstitial fluid and is less interfered by the stratum corneum of the skin. The 1064-nm laser can penetrate to a depth of 0.5 mm to 1.2 mm in the dermis layer and can detect the glucose concentration in the tissue fluid around capillaries, reducing the interference of melanin absorption.

[0034] Furthermore, the aforementioned pulsed laser emits laser in pulses with a frequency of 10 kHz, and the single-pulse energy ≤ 5 mJ. While avoiding thermal damage, the signal is extracted through a lock-in amplifier to suppress ambient light noise.

[0035] As Figure 3As shown in the figure, the aforementioned coupled-in optical module 34 includes an adaptive collimating lens 341, a polarization beam splitter 342, and a micro galvanometer 343 arranged in sequence along the optical path. The collimated beam emitted by the laser light source first passes through the adaptive collimating lens 341, which dynamically adjusts the focusing depth (±0.2 mm) according to the user's skin color characteristics to compensate for optical scattering caused by skin color differences. Subsequently, the polarization beam splitter 342 guides the beam into the micro galvanometer 343, and through the high-speed cooperative deflection of the micro galvanometer 343, two-dimensional scanning of the laser spot in a 5 mm × 5 mm area on the skin surface is achieved, covering heterogeneous structures such as sweat glands and hair follicles. The external controller integrally and real-time combines the skin color sensor and focus feedback data, synchronously optimizing the focusing position of the adaptive collimating lens 341 and the scanning path of the micro galvanometer 343 to ensure the uniformity and representativeness of the light spot in the detection area, ultimately improving the matching accuracy and detection efficiency of the skin optical characteristics of users with different skin colors.

[0036] As Figure 4 shown in the figure, the aforementioned static interference unit 35 includes a silicon nitride photonic crystal waveguide 351, a Mach-Zehnder interferometer 352 (MZI), and a metasurface grating 353 arranged in an array in sequence along the optical path. The silicon nitride photonic crystal waveguide 351 is designed with a bandgap matching the Raman characteristic peak of glucose (1120 cm⁻¹) to suppress background noise (such as the 1450 cm⁻¹ peak of collagen). Moreover, the surface of the silicon nitride photonic crystal waveguide is modified with an array of gold nanocones. For example, the array of gold nanocones is deposited by methods such as physical adsorption, chemical deposition, sol-gel method, and electrochemical deposition. The array of gold nanocones enhances the Raman spectral signal through the local surface plasmon resonance effect (LSPR), enhancing the Raman signal by at least 10³~10 4 times.

[0037] The aforementioned Mach-Zehnder interferometer 352 finely tunes the phase difference of the interference arms through the thermo-optic effect to compensate for the wavelength shift caused by temperature drift in real time (with an accuracy of ±0.01 nm). Moreover, the output end of the Mach-Zehnder interferometer is connected to a metasurface grating 353, which maps the scattered light according to the wavelength-space distribution to generate a Raman spectral image with a high resolution (0.5 cm - ⁻¹).

[0038] Furthermore, the aforementioned compound parabolic concentrator 37 is selected as an asymmetric compound parabolic concentrator (CPC), whose acceptance angle ≥ 60°, and the collection efficiency is increased by 3 times compared with traditional lenses. Moreover, a gold-titanium dioxide composite film is coated on the light incident surface of the compound parabolic concentrator 37, with a reflectivity > 98% in the 500 - 1200 nm band to reduce stray light loss. Or the compound parabolic concentrator 37 is selected as an asymmetric compound parabolic concentrator, whose acceptance angle is not less than 60°, and there may be no gold-titanium dioxide composite film on its light incident surface.

[0039] In this embodiment, a scientific-grade CMOS sensor (quantum efficiency > 80%, wavelength of 1064 nm) is used, in conjunction with a 16-bit ADC (analog-to-digital converter), to digitize weak signals.

[0040] In this embodiment, the aforementioned signal processing and analysis module includes a closed-loop control algorithm unit. The closed-loop control algorithm unit is used to output an updated rate weight of dynamic insulin according to the growth rate and decline rate of the actual blood glucose value within the past T1 time period of the input; it is also used to output an updated rate weight of basal insulin based on the percentage of the actual blood glucose value higher than the blood glucose set value and the percentage of the actual blood glucose value lower than the blood glucose set value within the past T2 time period of the input, where T2 is greater than T 1。

[0041] A model framework diagram of the human microcirculation is built in Matlab using a closed-loop control algorithm and docked with the FDA-certified diabetes physiological model T1DMS. The input of the model is the blood glucose value, and the outputs are the basal insulin rate and the dynamic insulin rate. Specifically, for the blood glucose historical data within the past 15 minutes, the closed-loop control algorithm can update the rate weight of dynamic insulin according to the growth or decline rate of blood glucose; for the blood glucose historical data within the past 24 hours, the closed-loop control algorithm can update the basal insulin rate weight according to the percentage of blood glucose higher than the set value and the percentage lower than the set value within the past 24 hours.

[0042] In a preferred embodiment, the closed-loop control algorithm unit can perform rolling optimization on the blood glucose reference trajectory based on the actual blood glucose value obtained at the next moment.

[0043] As Figure 5 shown, it is the closed-loop control algorithm model framework diagram (i.e., the closed-loop control algorithm unit) of the non-invasive fully closed-loop artificial pancreas system provided in this embodiment. The core algorithm of the closed-loop control algorithm unit adopts the model predictive control algorithm (MPC, model predictive control). When the real-time blood glucose value X d (i.e., the set value in the figure) is input into the model, the MPC algorithm will predict the blood glucose trend (blood glucose change trend) within a future period of time based on the current blood glucose value and generate a reference trajectory X r (k + j). Based on the predicted future blood glucose trend, the dual-channel injection module is controlled to inject the corresponding drugs in advance for blood glucose regulation, such as injecting insulin and glucagon.

[0044] The process of the MPC algorithm modifying the predicted blood glucose trend based on the actual blood glucose value detected at the next moment is called rolling optimization. The idea of rolling optimization is to continue to obtain the blood glucose value x(k) and predict the future blood glucose trend X according to the prediction model m(k + j), and generate a new trajectory X after online calibration p (k + j), and the following state space equations need to be established during this process:

[0045] When the closed-loop control algorithm unit performs rolling optimization on the blood glucose reference trajectory based on the actual blood glucose value obtained at the next moment, the cost function used is:

[0046] Among them, x i-1 represents the actual blood glucose value at the previous moment, x i represents the blood glucose value at the current moment, u i represents the insulin injection amount at this moment, A represents the weight coefficient matrix of the blood glucose value, B represents the weight coefficient matrix of the insulin injection dose, C Y represents the coefficient matrix of the historical blood glucose value, C V represents the blood glucose change rate calculation matrix, represents the loss function, N y represents the prediction space, N u represents the control interval, Qv k represents the blood glucose weight matrix, z k represents the blood glucose prediction error matrix, represents the insulin weight matrix, represents the blood glucose rate weight matrix, y i is the blood glucose historical data after weight correction, v i is the blood glucose change rate weight, is the insulin injection dose matrix, is the insulin injection rate matrix.

[0047] As Figure 6 shown, it is a schematic diagram of the blood glucose change of a diabetic patient in a day under closed-loop control. It can be seen that the blood glucose rises significantly after eating, and the closed-loop control algorithm immediately adjusts the insulin dose, and the blood glucose returns to the normal physiological range in a very short time. The blood glucose range in a day is as Figure 7As shown, it can be seen that the time of hypoglycemia occurrence is 0%, the time within the normal physiological range is 84%, and only 16% of the blood glucose values exceed the normal range. In fact, for normal individuals, transient hyperglycemia events also occur after eating. For diabetic patients, on the basis of avoiding hypoglycemia, blood glucose should be controlled within the normal physiological range as much as possible. By Figure 7 It can be seen that the control effect of this algorithm is better.

[0048] An acceleration sensor and a temperature sensor are also integrated on the aforementioned glucose optical detection module. The acceleration sensor is used to detect the exercise intensity of the object to be measured. For example, when it detects that the user is exercising vigorously (acceleration > 2g), it automatically raises the priority of glucagon infusion to prevent hypoglycemia; the aforementioned temperature sensor is used to detect the temperature of the object to be measured, and the signal processing and analysis module is used to regulate the priority of injecting glucagon by the dual-channel injection module 1 according to the exercise intensity, and is also used to correct the Raman peak position shift (-0.15 cm⁻¹ / ℃) according to the temperature (accuracy of ±0.1℃), so as to achieve adaptive calibration.

[0049] In the preferred embodiment, the closed-loop control algorithm unit can also perform a self-check of the laser power-signal response curve every 24 hours. If the deviation > 10%, it will trigger an alarm and switch to the backup light source of this application. At the same time, it also supports the dual-spectrum verification mode. For example, when comparing the detection results of 785nm and 1064nm, when the detection difference > 15%, it will start re-calibration.

[0050] In the preferred embodiment, the epidermal water content can also be measured by integrating an impedance sensor of 10kHz~100kHz, and the laser power is adjusted within 1mW~10mW to compensate for the change in light flux.

[0051] In the preferred embodiment, a personalized database of the user can also be established. When used for the first time, a multi-group of Raman spectrum data is collected through a 15-minute adaptive calibration mode to construct the user's exclusive "skin optical fingerprint", including parameters such as stratum corneum thickness and melanin distribution. These parameter data are encrypted and stored in the edge computing unit, which is integrated in the signal processing and analysis module and supports continuous monitoring with an error rate < 5% in the offline mode.

[0052] As Figure 8 shown, it is the structural diagram of the micro wearable artificial pancreas system provided by this embodiment. Figure 9Explosion structure diagram of the miniature wearable artificial pancreas system provided in this embodiment. The miniature wearable non-invasive fully closed-loop artificial pancreas system further includes a housing 4 and a hollow microneedle array module 5. The aforementioned dual-channel injection module 1, signal processing and analysis module 2, and glucose optical detection module 3 are sequentially integrated and arranged in the housing 4. The housing 4 specifically includes an upper housing, a middle housing, and a lower housing, and the respective housing parts are detachably connected in sequence through a snap structure. The replaceable hollow microneedle array module 5 includes microneedles arranged in an array. The microneedle material is a biocompatible polymer, ceramic, or metal. The microneedle length is 0.5 mm to 1.2 mm, the inner diameter is ≤50 μm, and it is mechanically connected to the corresponding micropump through a snap structure to achieve non-invasive subcutaneous drug delivery. The housing 4 is made of flexible medical silicone material, and multiple isolation chambers can be provided inside. Each module of the artificial pancreas system is installed in the corresponding chamber, and rapid disassembly and assembly are achieved through a cable and a magnetic interface, and the blood glucose data and infusion records can be synchronized with the smart terminal in real time through low-power Bluetooth.

[0053] The aforementioned dual-channel injection module 1 is detachably embedded on the upper housing. The middle housing has an installation cavity, and the signal processing and analysis module 2 is integrally fixed in the installation cavity of the middle housing. The upper and lower surfaces of the signal processing and analysis module 2 are electrically connected to the dual-channel injection module 1 and the glucose optical detection module 3 respectively. Electrical connection can be specifically achieved by electrode connection and wire connection methods. The detection end of the glucose optical detection module 3 passes through the hollow microneedle array module 5 and is embedded in the lower housing, and its detection end has the same orientation as the output port of the hollow microneedle array module 5. The output port of the dual-channel injection module 1 is communicated with the input port of the hollow microneedle array module 5 through a catheter 11. The output port of the hollow microneedle array module 5 extends to the outside of the lower housing for inputting the corresponding blood glucose regulating drug to the object to be measured. The artificial pancreas system of this embodiment can replace the detachable dual-channel injection module 1 and hollow microneedle array module 5 according to the service life, and other modules can be recycled.

[0054] Figure 10 Structure diagram of the hollow microneedle array module of this embodiment. The hollow microneedle array module is used to cooperate with the dual-channel injection module 1 to complete non-invasive injection. It can have one drug inlet 51, or it can also have two inlets at the same time to complete the injection of two different drugs. Each drug inlet forms a flow channel connection with different microneedles, and drug injection is completed through the respective microneedles.

[0055] Figure 11Schematic diagram of the dual-channel injection module provided by this application. Among them, the dual-channel injection module 1 is integrally fixed inside the upper housing of the housing 4, and it includes a medicine storage 13 placed inside the housing. A medicine filling channel 15 is provided on the medicine storage 13, and the medicine filling channel 15 extends outside the upper housing of the housing 4, and medicine can be manually filled into the medicine storage 13. The medicine storage 13 and the micropump 14 are connected by a catheter 12. The injection unit composed of the medicine storage 13 and the micropump 14 includes a pair, and each is connected to the microneedle array through a corresponding injection outlet 11.

[0056] Figure 12 Schematic diagram of a dual-channel injection module with a replaceable medicine storage. The medicine storage 13 is embedded in the upper housing of the housing 4 and is fixedly connected to the upper housing by a snap-fastening method. The micropump 14 is located between the medicine storage 13 and the signal processing and analysis module 2 (specifically an integrated circuit board). After the medicine inside is used up, the medicine storage 13 can be directly removed from the outside of the housing, and then other medicine storages filled with medicine can be replaced. One or more medicine storages (not shown in the figure) can be embedded in the position where the upper housing embeds the medicine storage, and each medicine storage is connected to a corresponding micropump and is connected to the microneedle array module to achieve the independent medicine delivery function.

[0057] It should be understood that expressions such as "including" and "may include" that can be used in this application indicate the existence of the disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements. In this application, terms such as "including" and / or "having" can be interpreted as indicating a specific characteristic, number, operation, constituent element, component, or a combination thereof, but cannot be interpreted as excluding the existence or the possibility of adding one or more other characteristics, numbers, operations, constituent elements, components, or a combination thereof.

[0058] In addition, in this application, the expression "and / or" includes any and all combinations of the associated listed words. For example, the expression "A and / or B" can include A, can include B, or can include both A and B.

[0059] In the description of the embodiments of the present application, it should be noted that unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense. For example, "connection" can be a detachable connection or a non-detachable connection; it can be a direct connection or an indirect connection through an intermediate medium. Among them, "fixed connection" means that the two are connected and the relative positional relationship after connection remains unchanged. "Rotational connection" means that the two are connected and can rotate relative to each other after connection. "Sliding connection" means that the two are connected and can slide relative to each other after connection. The orientation terms mentioned in the embodiments of the present application, such as "top", "bottom", "inner", "outer", "left", "right", etc., are only references to the direction of the accompanying drawings. Therefore, the orientation terms used are for better and clearer description and understanding of the embodiments of the present application, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the embodiments of the present application.

[0060] In addition, in the embodiments of the present application, mathematical concepts such as symmetry, equality, parallelism, and perpendicularity are mentioned. These limitations are all in view of the current technological level, rather than absolute strict definitions in the mathematical sense. A small deviation is allowed, and approximations such as approximate symmetry, approximate equality, approximate parallelism, and approximate perpendicularity are all acceptable. For example, when it is said that A is parallel to B, it means that A is parallel to B or approximately parallel to B, and the included angle between A and B can be between 0 degrees and 10 degrees. When it is said that A is perpendicular to B, it means that A is perpendicular to B or approximately perpendicular to B, and the included angle between A and B can be between 80 degrees and 100 degrees.

[0061] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A non-invasive fully closed-loop artificial pancreas system, characterized in that, It includes a glucose optical detection module (3), a signal processing and analysis module (2), and a dual-channel injection module (1) connected in sequence. The glucose optical detection module (3) is used to emit laser towards the object to be measured to excite and generate Raman spectral signals, and is also used to obtain the actual blood glucose value of the object to be measured based on the Raman spectral signals and transmit it to the signal processing and analysis module (2); the signal processing and analysis module (2) is used to analyze the blood glucose change trend within a preset future time period based on the actual blood glucose value to generate a blood glucose reference trajectory; the signal processing and analysis module (2) is also used to control the two injection channels of the dual-channel injection module (1) to be opened respectively according to the blood glucose reference trajectory to inject corresponding drugs into the object to be measured.

2. The non-invasive fully-closed-loop artificial pancreas system according to claim 1, wherein, The glucose optical detection module (3) includes a laser source (31), a galvanometer (32), and an input optical module (34), a static interference unit (35), an output grating array (36), a compound parabolic concentrator (37), an image sensing unit (38), and a data processing unit (39) arranged in sequence along the optical path; the laser emitted by the laser source (31) is reflected by the galvanometer (32) and then vertically enters the object to be measured to detect the blood glucose concentration. The input optical module (34) is used to receive the blood glucose concentration signal emitted from the object to be measured and transmit it to the static interference unit (35) to extract spectral features. The output grating array (36) is used to perform spectral splitting on the composite light with the extracted spectral features. The compound parabolic concentrator (37) is used to focus the optical signal after spectral splitting into the image sensing unit (38) for photoelectric conversion, and the electrical signal after photoelectric conversion is input into the data processing unit (39) for analysis to obtain the glucose concentration value of the object to be measured.

3. The non-invasive fully closed-loop artificial pancreas system according to claim 2, wherein The laser source uses a dual-wavelength pulsed laser with wavelengths of 780 nm - 790 nm and 1054 nm - 1074 nm respectively, and uses time-division multiplexing technology for alternating excitation; the pulsed laser emits laser in pulses with a frequency of 10 kHz, and the single-pulse energy ≤ 5 mJ.

4. The non-invasive fully-closed-loop artificial pancreas system according to claim 2, wherein The static interference unit (35) includes a silicon nitride photonic crystal waveguide (351), a Mach-Zehnder interferometer (352), and a metasurface grating (353) arranged coaxially in sequence along the optical path, and the surface of the silicon nitride photonic crystal waveguide (351) is modified with a gold nanocone array.

5. The non-invasive fully closed-loop artificial pancreas system according to claim 2, characterized in that, The input optical module (34) includes an adaptive collimating lens (341), a biased beam splitter (342), and a micro galvanometer (343) arranged in sequence along the optical path. The adaptive collimating lens (341) is used to adjust the focusing depth of the beam emitted from the object to be measured based on the skin color characteristics of the object to be measured. The biased beam splitter (342) is used to introduce the beam with the adjusted focusing depth into the micro galvanometer (343) for high-speed collaborative deflection.

6. The non-invasive fully-closed artificial pancreas system according to claim 2, wherein The compound parabolic concentrator (37) is an asymmetric compound parabolic concentrator, and its acceptance angle is not less than 60 °C; and / or, the incident light surface of the compound parabolic concentrator (37) is coated with a gold-titanium dioxide composite film.

7. The non-invasive fully-closed artificial pancreas system according to claim 1, wherein, The signal processing and analysis module (2) includes a closed-loop control algorithm unit, which is used to output an updated rate weight of dynamic insulin according to the growth rate and decline rate of the actual blood glucose value in the past T1 time period of the input; and is also used to output an updated rate weight of basal insulin based on the percentage of the actual blood glucose value higher than the blood glucose set value and the percentage of the actual blood glucose value lower than the blood glucose set value in the past T2 time period of the input, where T2 is greater than T1; in the foregoing process, the following state space equation is established: Among them, x i-1 represents the actual blood glucose value at the previous moment, x i represents the blood glucose value at the current moment, u i represents the insulin injection amount at this moment, A represents the weight coefficient matrix of the blood glucose value, B represents the weight coefficient matrix of the insulin injection dose, C Y represents the coefficient matrix of the historical blood glucose value, C V represents the blood glucose change rate calculation matrix, y i is the historical blood glucose data after weight correction, v i is the weight of the blood glucose change rate.

8. The non-invasive fully-closed artificial pancreas system according to claim 7, wherein, The closed-loop control algorithm unit can perform rolling optimization on the blood glucose reference trajectory based on the actual blood glucose value obtained at the next moment, and the cost function used in the rolling optimization is: Among them, represents the loss function, N y represents the prediction space, N u represents the control interval, Q ( v k ) represents the blood glucose weight matrix, represents the blood glucose prediction error matrix, represents the insulin weight matrix, represents the blood glucose rate weight matrix, is the insulin injection dose matrix, is the insulin injection rate matrix.

9. The non-invasive fully-closed artificial pancreas system according to claim 1, characterized in that, The glucose optical detection module (3) is integrated with a temperature sensor and an acceleration sensor. The temperature sensor is used to detect the temperature of the object to be measured and transmit it to the signal processing and analysis module (2), and the acceleration sensor is used to detect the exercise intensity of the object to be measured and transmit it to the signal processing and analysis module (2). The signal processing and analysis module (2) is used to correct the Raman peak position shift based on the temperature, and is also used to adjust the priority of injecting glucagon by the dual-channel injection module (1) according to the exercise intensity.

10. The non-invasive fully closed-loop artificial pancreas system according to claim 1, characterized in that, The non-invasive fully closed-loop artificial pancreas system further includes a housing and a hollow microneedle array module. The glucose optical detection module (3), the signal processing and analysis module (2), and the dual-channel injection module (1) are sequentially integrated and arranged in the housing. The output port of the dual-channel injection module (1) is communicated with the input port of the hollow microneedle array module, and the output port of the hollow microneedle array module extends outside the housing through the housing.

Images