Search

CN-122006005-A - ICU intelligent vasoactive drug closed-loop infusion system for cardiac surgery

CN122006005ACN 122006005 ACN122006005 ACN 122006005ACN-122006005-A

Abstract

The invention discloses an intelligent vascular active drug closed-loop infusion system for cardiac surgery ICU, which relates to the technical field of medical equipment and control, in particular to an intelligent vascular active drug closed-loop infusion system for cardiac surgery ICU, comprising a multichannel physiological signal acquisition unit for continuously acquiring blood pressure and heart rate signals of a patient and a central processing and control unit with a built-in drug regulation and control logic module. The regulation and control module automatically generates a delay-free control instruction based on comparison of the real-time physiological signal and a preset target range and directly drives the drug infusion execution unit to adjust the infusion rate. The system also comprises a safety boundary monitoring unit and a nursing record synchronization interface, so that treatment safety and data synchronization are ensured. In addition, the system can integrate data of third-party monitoring equipment, support multi-drug collaborative management and AI-based personalized target dynamic setting, and realize intelligent and automatic closed-loop accurate regulation and control of the infusion of the vasoactive drugs of critical patients.

Inventors

  • ZHANG MIN

Assignees

  • 安徽医科大学第一附属医院

Dates

Publication Date
20260512
Application Date
20260305

Claims (10)

  1. 1. An ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery comprising: The multichannel physiological signal acquisition unit is used for continuously acquiring at least blood pressure and heart rate signals of a patient; The input end of the central processing and control unit is connected with the physiological signal acquisition unit; the control end of the drug infusion execution unit is connected with the output end of the central processing and control unit and is used for driving at least one injection pump to infuse the vasoactive drugs; the central processing and control unit is also connected with a nursing record synchronization interface for synchronizing key detection content with a nursing record list in the running process of the system; the central processing and control unit is internally provided with a medicine regulation logic module, and the module compares the physiological signal value obtained in real time with a target value range preset in the central processing and control unit and automatically generates a control instruction without artificial delay, and directly drives the medicine infusion execution unit to adjust the infusion rate.
  2. 2. The ICU intelligent vasoactive drug closed-loop infusion system for cardiac surgery of claim 1, wherein the drug regulatory logic module comprises a first level regulatory sub-module and a second level regulatory sub-module; The first-stage regulation and control submodule is used for calculating and outputting a preliminary drug infusion rate adjustment quantity according to a pre-stored first-class rule set when the physiological signal value deviates from the target value range; The second-stage regulation and control sub-module receives the preliminary adjustment quantity and the physiological signal change trend data, corrects the preliminary adjustment quantity according to a pre-stored second-class rule set, and generates a final control instruction.
  3. 3. The ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery of claim 2, wherein the correction logic of the second set of rules comprises applying a suppression coefficient to the preliminary adjustment when the rate of change of the physiological signal value is monitored to exceed a preset threshold, and applying a trimming coefficient to the preliminary adjustment when the physiological signal value is monitored to continue to wander at a target value boundary.
  4. 4. The ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery of claim 1, wherein the system further comprises a safety margin monitoring unit; The safety boundary monitoring unit continuously monitors the data flow of the physiological signal acquisition unit and the state of the drug infusion execution unit; When the physiological signal value is detected to exceed a preset safety limit range or at least one abnormal state of signal loss and communication interruption of the injection pump is detected, the safety boundary monitoring unit can immediately trigger the system to enter a preset safety maintenance mode or a safety degradation mode and send out an alarm; the safety boundary monitoring unit can synchronize the event type, the occurrence time and the related physiological data to a nursing record list through the nursing record synchronization interface when an alarm or mode switching is triggered.
  5. 5. The ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery of claim 4, wherein the safe hold mode is to immediately lock the current drug infusion rate and maintain it, and the safe degrade mode is to gradually decrease the drug infusion rate to a basal maintenance level according to a preset stepwise procedure.
  6. 6. The ICU intelligent vasoactive drug closed-loop infusion system for cardiac surgery according to claim 1, wherein the central processing and control unit is further connected with a monitoring data integration interface for receiving at least central venous pressure and blood oxygen saturation data from a third-party monitoring device, and the drug regulation and control logic module takes the central venous pressure and blood oxygen saturation data as auxiliary judgment parameters when generating a control instruction; and the third-party monitoring equipment data received by the monitoring data integration interface, and the infusion rate adjustment instruction and the execution result generated by the drug regulation logic module are synchronized to a nursing record list through the nursing record synchronization interface.
  7. 7. The ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery of claim 1, wherein the system further comprises a simulated deduction unit; the simulation deduction unit can simulate the physiological signal change track under different adjustment strategies in a future period according to the current physiological signal value, the drug infusion rate and the built-in physiological response model of the patient; the output result of the simulation deduction unit can be used for the medicine regulation and control logic module to carry out internal pre-evaluation reference before a control instruction is generated.
  8. 8. The ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery of claim 1, wherein the system comprises a multi-drug co-administration module adapted to infuse two or more vasoactive drugs simultaneously; priority rules and interaction compensation rules among different medicines are pre-stored in the collaborative management module; When the infusion rates of multiple medicines are required to be adjusted, the collaborative management module determines an adjustment sequence according to the priority rule, calculates the influence quantity of one medicine adjustment on the target effect of the other medicine according to the interaction compensation rule and performs compensatory adjustment; The multi-drug co-management module synchronizes the adjustment scheme, the original rate, the target rate and the adjustment cause of each drug to the care record list through the care record synchronization interface when executing any co-adjustment operation involving multiple drugs.
  9. 9. An ICU intelligent vasoactive drug closed loop infusion system for cardiac surgery according to claim 1, wherein: The central processing and control unit further includes an AI-based personalized dynamic targeting module configured to: continuously learning and analyzing a historical response pattern of a current patient's physiological signal to drug infusion rate adjustment during system operation; Dynamically fine-tuning a default target value range pre-stored in the drug regulation logic module based on the learned individual response mode to generate an individual target value range more adapted to the current physiological characteristics of the patient; and synchronizing the adjustment amplitude, the basis and the effective time of the individuation target value range to a nursing record list through the nursing record synchronization interface.
  10. 10. A closed loop infusion system for cardiac surgical ICU intelligent vasoactive drugs according to claim 1 or 9, wherein: The central processing and control unit further comprises a reinforcement learning-based regulation strategy self-optimization module configured to: continuously collecting state data, action data and control effect feedback data of a closed-loop control process; Based on a reinforcement learning algorithm, the parameters of a control rule set pre-stored in the medicine regulation logic module are subjected to iterative optimization by taking maintenance of physiological signals stable and reduction of unnecessary frequent adjustment as optimization targets; The key nodes of policy optimization, parameter comparison before and after optimization and expected effect evaluation are synchronized to a care record list through the care record synchronization interface, The system also comprises a multi-mode man-machine interaction and intention confirming unit which integrates at least one interaction mode of voice recognition, gesture recognition and sight tracking; When the system detects the input of a preset major operation instruction or the active intervention of medical staff, the operation intention can be secondarily confirmed in a multi-mode interaction mode; the man-machine interaction and intention confirming unit records all major interaction events, confirming processes and final instructions and synchronizes to a nursing record list through the nursing record synchronizing interface.

Description

ICU intelligent vasoactive drug closed-loop infusion system for cardiac surgery Technical Field The invention relates to the technical field of medical equipment and control, in particular to an ICU intelligent vasoactive drug closed-loop infusion system for cardiac surgery. Background In cardiac surgical post-operative management, maintaining patient hemodynamic stability is critical. Vasoactive drugs (e.g., inotropic drugs, vasoconstrictors, vasodilators, etc.) are key therapeutic tools for regulating and controlling blood pressure, cardiac output, organ perfusion, etc. core vital signs. Currently, in cardiac surgical ICU, the infusion of vasoactive drugs is commonly performed in an open loop, manual regulation mode. I.e. medical staff manually set and adjust the infusion rate of the injection pump according to physiological parameters such as arterial blood pressure, heart rate, central venous pressure and the like of a patient which are intermittently acquired and by combining clinical experience. This traditional model has a series of inherent drawbacks and potential risks, and has become a bottleneck for refined and efficient post-operative management. First, significant delay and hysteresis are major issues. The physiological state of a patient, especially early post-operative, is in dynamic rapid changes. From the time the parameter change is identified, to the healthcare worker evaluating the decision, and then manually adjusting the infusion pump, there is a manual delay of several minutes or even longer in the overall process. During this time, the patient may already be in a dangerous state of hypotension or hypertension, increasing the risk of organ hypoperfusion or cardiac overload. This lag in regulation not only affects the optimal timing of the treatment, but may also lead to fluctuations in physiological parameters up and down around the target values, which are detrimental to stable homeostasis. Second, high-intensity human dependence brings uncertainty to individual experience differences. Closed loop adjustment requires the healthcare worker to maintain a high level of attention for a long period of time, frequently viewing the monitor and calculating the adjusted dose. In a busy ICU environment, this greatly increases the workload of the healthcare worker and is also prone to overlooking from fatigue. Meanwhile, the adjustment of the medicine dosage is highly dependent on personal experience and instant judgment of medical staff, and different operators possibly have policy and scale differences, so that standardized and homogeneous accurate treatment is difficult to realize, and the controllability of the overall medical quality is affected. Furthermore, the processing power for complex dynamic processes is limited. The regulation of hemodynamics is a complex process of multivariate, nonlinear, and reciprocal. For example, adjusting a drug may affect both blood pressure and heart rate, and there are peaks of effects with different time constants. When multiple vasoactive drugs are used in combination, the interaction (synergy or antagonism) between them makes manual adjustment more complex. The traditional manual mode is difficult to analyze and coordinate the complex relationships quantitatively in real time, and is generally difficult to realize optimal control based on multi-parameter fusion by adopting a relatively conservative or simplified processing mode. In addition, safety guardianship has a blind area. In the traditional mode, the monitoring of the state of the infusion system itself (such as the blockage of an infusion line, the communication failure of an injection pump, the impending completion of the infusion of the drug, etc.) is relatively separated from the monitoring of the physiological state of the patient. When an anomaly occurs, the early warning and intervention also rely on manual discovery, possibly creating a delay in treatment. In extreme cases, such as sudden loss of signal or equipment failure, the lack of an automatic safety redundancy mechanism may directly jeopardize patient safety. To address the above challenges, the field of medical automation has begun exploring closed-loop infusion techniques. However, most of the existing researches or product schemes focus on single-drug control of single parameters (such as blood pressure only), and the control strategy is relatively simple (such as a classical PID controller), so that the method has insufficient adaptability to complex and multi-stage pathophysiological characteristics of patients after cardiac surgery and clinical practical requirements of multi-drug combination. These systems often lack advanced intelligent decision logic to cope with rapid fluctuations in physiological signals, multiparameter coupling, multi-drug synergy, and differential targeting at different clinical stages, as well as deep, prospective safety boundaries and abnormal state handling mechanisms. Accordingly, a smart vasoact