Artificial intelligence assisted multi-mode small animal anesthesia and consciousness research device and system

Abstract

The invention belongs to the technical field of neurobiology and anesthesiology research, and discloses an artificial intelligence-assisted multi-mode small animal anesthesia and consciousness research device and system, and the system comprises an integrated monitoring cabin which comprises a transparent experiment box and a sound insulation camera obscura; the electroencephalogram / myoelectricity recording module is used for recording cerebral cortex and muscle electrical activities through a multi-channel electrode array; an optical fiber calcium imaging module, an optical fiber probe and a high-sensitivity photoelectric detector; the drug delivery module controls a micro-injection pump in a program mode and supports intravenous, abdominal and suction type drug delivery; the vital sign monitoring module comprises a wireless infrared body temperature detector, a wireless millimeter wave radar detector and a high-precision pressure sensor; the behavior capture module, the infrared high-speed camera and the attitude analysis system are used for monitoring righting reflection and other behaviors; and the AI operation and analysis platform integrates data acquisition, processing and visualization functions. According to the invention, full-process automation from drug administration to data acquisition is realized, manual intervention is not needed, the multi-modal data synchronization precision is improved by more than 90%, and the experiment efficiency is improved by more than 50%.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of neurobiology and anesthesiology research, and particularly relates to an artificial intelligence assisted multi-modal small animal anesthesia and consciousness research device and system. BACKGROUND

[0002] The mechanism of general anesthetics involves complex neurobiological processes, including changes in central nervous system activity during induction, maintenance, and recovery phases of anesthesia. Current techniques for studying anesthetic mechanisms include electroencephalography / muscle electromyography (EEG / EMG), calcium imaging, pharmacological intervention, vital sign monitoring (such as heart rate, respiration, body temperature), and behavioral observation (such as righting reflex). These methods have played an important role in revealing the effects of anesthetic drugs on neuronal activity, neural circuit function, and overall physiological state. However, the limitations of existing research methods and technical equipment significantly constrain experimental efficiency and the integrity of data interpretation.

[0003] Limitations of existing anesthesia research techniques:

[0004] Data collection is scattered: Traditional EEG / EMG recording systems usually operate independently and cannot be synchronized in real time with calcium imaging, vital sign monitoring, or behavioral observation. For example, EEG / EMG recording often requires separate equipment, while calcium signals require additional optical fiber measurement systems, making it difficult to time-align multi-modal data and affecting comprehensive analysis of the dynamic changes of anesthetic state.

[0005] Low degree of automation: In existing systems, anesthetic drug administration (such as intravenous or intracerebroventricular injection) relies heavily on manual operation, and the precise control of drug dosage and administration timing is limited by the experimenter's experience, which is prone to human error. In addition, behavioral detection such as righting reflex requires manual judgment, which is inefficient and highly subjective.

[0006] Single index: Current devices usually focus on a single index (such as EEG power spectrum or heart rate variation), lacking integrated monitoring of neural activity (EEG / calcium signals), physiological state (heart rate, respiration, body temperature), and behavioral response (righting reflex) during the anesthetic process. For example, traditional EEG recording cannot directly reflect the dynamic changes of calcium signals in neuronal populations, and vital sign monitoring is not associated with neural data, leading to a one-sided understanding of anesthetic depth and recovery process.

[0007] Low experimental efficiency: Due to the dispersion of equipment and complex operation, each animal's anesthesia experiment requires multiple instrument switching and repeated calibration, with a long duration (usually > 2 hours) for a single experiment, limiting the feasibility of large sample size research and high-throughput screening.

[0008] Limitations of existing systems:

[0009] Synchronization deficiency: Existing electroencephalogram / electromyogram recording systems (such as Pinnacle or BlackRock) can record neural signals, but cannot integrate fiber-optic calcium imaging or real-time drug delivery modules, resulting in asynchronous data collection. For example, correlation analysis between calcium signals and brain waves requires manual alignment in the later stage, which is prone to introduce time errors.

[0010] Poor functional expansion: Existing devices are mostly designed for specific purposes, making it difficult to support multiple anesthetic drug delivery methods (such as isoflurane and propofol) or multiple monitoring methods (such as dual recording of body temperature and respiration), limiting the flexibility of research.

[0011] Lack of intelligent analysis: Current systems rely on manual analysis of EEG frequency spectrum or behavior videos, lack of artificial intelligence (AI) assisted multi-modal data fusion and real-time processing capabilities, and cannot quickly identify key turning points of anesthesia status (such as loss or recovery of consciousness). SUMMARY

[0012] To address the problems of existing technologies, the present application provides an artificial intelligence assisted multi-modal small animal anesthesia and consciousness research device and system.

[0013] The present application is implemented as follows: an artificial intelligence assisted multi-modal small animal anesthesia and consciousness research system, which comprises:

[0014] An integrated monitoring cabin containing a transparent experiment box and a soundproof dark box, equipped with a metal grid plate to separate the upper and lower two layers;

[0015] An electroencephalogram / electromyogram recording module with a multi-channel electrode array to record brain cortex and muscle electrical activity;

[0016] A fiber-optic calcium imaging module with a fiber-optic probe and a high-sensitivity photoelectric detector to monitor neuron calcium signals in real time;

[0017] A drug delivery module with a programmable micro-injection pump to support central ventricle and peripheral vein / abdominal cavity drug delivery;

[0018] A vital sign monitoring module including a wireless infrared body temperature detector, a wireless millimeter wave radar detector, and a high-precision pressure sensor;

[0019] A behavior capture module with an infrared high-speed camera and a posture analysis system to monitor behaviors such as righting reflex;

[0020] An AI computing and analysis platform integrating data acquisition, processing, and visualization functions, using artificial intelligence algorithms to analyze multi-modal data.

[0021] Further, the upper layer of the integrated monitoring cabin is a transparent experiment box where experimental animals are placed, and the electroencephalogram / electromyogram electrodes, fiber-optic probes, wireless infrared body temperature detector, wireless millimeter wave radar detector, and infrared high-speed camera are installed.

[0022] The lower layer is a soundproof dark box, and the installed drug delivery module is a micro-injection pump and a mechanical positioning device, which ensures the accuracy of drug delivery and environmental stability.

[0023] The electroencephalogram / electromyogram recording module adopts 6-channel nickel-chromium alloy electrodes with an impedance of 3-4MΩ, records EEG of 0.5-100Hz and EMG of 10-500Hz, and has a sampling rate of 2000Hz, and supports real-time spectrum analysis.

[0024] Further, the optical fiber calcium imaging module has an optical fiber probe with a diameter of 200μm, NA 0.37, and a 470nm LED excitation calcium fluorescence indicator (such as GCaMP6s), records calcium signals, and has a sampling rate of 100Hz and a resolution of ±0.1△F / F0.

[0025] The micro-injection pump in the drug delivery module has a flow rate of 0.1-100μL / min, and realizes intravenous or intraperitoneal drug delivery. In addition, the drug delivery module also supports isoflurane atomization delivery with a concentration of 0.5%-5%.

[0026] Further, the vital sign monitoring module comprises:

[0027] Wireless infrared temperature detector: resolution ±0.1℃, range 30-40℃;

[0028] Wireless millimeter wave radar detector: heart rate 100-1000bpm, error ±2bpm, respiration 30-300 times / minute, error ±2 times / minute;

[0029] High-precision pressure sensor: monitor chest and abdominal movement, sensitivity 0.01mm.

[0030] Further, the infrared high-speed camera in the behavior capture module is 1920x1080 pixels, 120fps, captures positive and negative reflections, limb movement trajectories and head posture, and cooperates with AI posture analysis to have a resolution of ±0.1mm / s.

[0031] Further, the AI operation and analysis platform integrates multi-modal data acquisition with a synchronization accuracy of <10ms, analyzes electroencephalogram spectrum, calcium signal waveform, vital sign trend and behavior characteristics based on CNN and RNN algorithms, and generates an anesthesia state index in real time, such as a loss of consciousness time point.

[0032] Another object of the present application is to provide an artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method based on the artificial intelligence assisted multi-modal small animal anesthesia and consciousness research system, which specifically comprises:

[0033] Step 1: Experimental preparation

[0034] Place the experimental animal on the upper layer of the integrated monitoring capsule, fix the electroencephalogram / electromyogram electrode and the optical fiber probe, and ensure that all sensors and the drug delivery module are in working condition;

[0035] Step 2: Baseline recording

[0036] Record 5-10 minutes of baseline data before anesthesia, including electroencephalogram / electromyogram, optical fiber calcium signal, heart rate, respiration, body temperature and behavioral posture;

[0037] Step 3: Anesthesia induction

[0038] Automatically deliver anesthetic drugs such as propofol 20mg / kg intravenous injection through the drug delivery module, and record multi-modal data synchronously;

[0039] Step 4: Anesthesia maintenance and recovery

[0040] Adjust the drug dosage or concentration, such as propofol 10mg / kg intravenous injection, monitor the maintenance stage and recovery process after drug withdrawal;

[0041] Step 5: Data processing and analysis

[0042] The AI platform performs feature extraction and fusion analysis on multi-modal data to generate anesthesia induction, maintenance and recovery curves, neural activity patterns and behavioral activity time points;

[0043] Step 6: Result storage and visualization

[0044] All data is stored in real time to the database, and visual reports are generated, including EEG curves, time-frequency diagrams, power spectrum diagrams, electromyograms, sleep phase diagrams, calcium signal curves, and vital sign curves.

[0045] Another object of the present application is to provide a computer device, characterized in that the computer device comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method.

[0046] Another object of the present application is to provide a computer readable storage medium storing a computer program, which is executed by a processor to make the processor execute the steps of the artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method.

[0047] Another object of the present application is to provide an information data processing terminal for realizing the artificial intelligence assisted multi-modal small animal anesthesia and consciousness research system.

[0048] In combination with the technical solutions and the technical problems solved above, the technical solutions to be protected by the present application have the following advantages and positive effects:

[0049] First, the system integrates a high-precision drug delivery system, which can automatically complete the operation processes of drug suction, injection, and subsequent cleaning. The positioning error is less than 0.1 mm, which is much better than the precision of traditional manual operation. The entire drug delivery process does not require manual intervention, and can be repeated multiple times according to the preset program to ensure consistency and repeatability of the experiment. At the same time, combined with the experiment scheduling module, the system can run autonomously at night or for a long period of time, greatly saving labor costs and improving overall experimental efficiency by more than 50%, greatly promoting the possibility of high-throughput animal experiments.

[0050] The system innovatively integrates multiple modal data acquisition channels such as electroencephalogram (EEG), electromyogram (EMG), optical fiber calcium imaging, electrocardiogram, respiration, body temperature, and behavior, and realizes millisecond-level time synchronization (error controlled within 10 ms). This enables researchers to dynamically track the state changes during anesthesia from multiple dimensions such as neural electrical signals, physiological indicators, and behavioral performance, far exceeding the analysis dimensions of traditional single parameters such as EEG or MAC value, significantly improving the understanding of anesthesia depth and neural mechanisms.

[0051] The system has a built-in deep learning algorithm based on the fusion architecture of convolutional neural network (CNN) and recurrent neural network (RNN), which can extract and fuse features from the collected multi-modal data, and automatically identify key event nodes such as induction of anesthesia (LOC) and recovery of consciousness (ROC). The model can achieve an identification accuracy of more than 90% after artificial label training, greatly improving the objectivity and consistency of the identification results, avoiding the defects of strong subjectivity and poor repeatability of human judgment, and providing reliable basis for subsequent drug efficacy evaluation and mechanism research.

[0052] The system supports automatic control of multiple common anesthetics, including isoflurane, sevoflurane, propofol, cyclopropofol, remazepam, and dexmedetomidine, etc. Both volatile inhaled drugs and intravenous injection drugs can be precisely managed. At the same time, it supports multiple experimental animal models such as mice and rats, and researchers can adjust the experimental process and parameters as needed to adapt to multiple research directions such as neural circuit tracking, anesthesia depth evaluation, and drug screening. The system also reserves interfaces for future expansion of new signal acquisition devices or data processing modules, with good upgrade space and scientific research versatility.

[0053] Second, to provide efficient research tools for neurobiology and anesthesiology research institutions in multiple fields, improve experimental efficiency, and reduce labor costs. Compared with traditional equipment, the experimental efficiency is improved by more than 50%, which can accelerate the progress of scientific research projects, enable scientific research institutions to conduct more experiments and produce more results in the same time, attract scientific research units to purchase, and create considerable sales revenue.

[0054] Assist pharmaceutical companies in screening anesthetic drugs, evaluating drug efficacy, accurately analyzing the effects of drugs on neural activity, physiological state and behavioral response, and improving the success rate and shortening the development cycle of R&D. Shortening the R&D cycle means saving a lot of R&D funds, and pharmaceutical companies are willing to pay a higher fee for cooperation or purchase equipment, bringing substantial commercial returns.

[0055] With the advantage of innovative technology, hold professional training courses and academic seminars, provide technical training and learning exchange platform for scientific researchers, charge training fees, enhance brand awareness and industry influence, and expand business cooperation opportunities.

[0056] The system is an integrated high-end scientific research equipment with high added value, which can generate profits by selling to global scientific research institutions, universities and pharmaceutical companies. Perfect after-sales maintenance and technical upgrade services can charge service fees to form a stable and sustainable source of income.

[0057] Multi-modal data synchronous acquisition and integration: Existing technologies cannot achieve real-time synchronous recording of EEG / EMG, optical fiber calcium signal, vital signs and behavioral data. This invention synchronizes these data for the first time (time error <10ms), providing complete data for comprehensive research on the anesthesia process, and filling the gap in multi-modal data synchronous acquisition and integration.

[0058] Automatic drug delivery and intelligent analysis integration: Traditional devices rely on manual operation for drug delivery and manual data processing, which is inefficient and prone to errors. This invention realizes the automation of the whole process from drug delivery to data acquisition, and uses AI algorithms (CNN+RNN) to automatically identify key nodes of anesthesia (accuracy >90%), leading the way in automatic drug delivery and intelligent analysis integration, and filling the gap in related technologies.

[0059] Solve the problem of multi-modal data time alignment: Multi-modal data time alignment has always been a difficult problem in anesthesia research. Traditional devices operate independently, making it difficult to align multi-modal data, which affects the analysis of the dynamic changes of anesthesia state. This invention solves this problem by integrating design and AI operation and analysis platform, achieving multi-modal data synchronization accuracy <10ms.

[0060] Breakthrough in precise monitoring of anesthesia depth: In the past, there was a lack of integrated monitoring of multiple indicators during anesthesia, making it impossible to accurately monitor anesthesia depth. This invention integrates multiple modules to analyze neural activity, physiological state and behavioral response, and generates anesthesia state index in real time, such as the time points of loss and recovery of consciousness, achieving precise monitoring of anesthesia depth.

[0061] Break the limitation of device function bias: in the past, it is believed that the function of scientific research equipment should focus on single index monitoring or specific operation, which leads to poor expansibility of equipment function. The invention breaks this bias and designs a system with diverse functions and scalability, supporting multiple anesthesia drug administration methods and multiple monitoring methods, and compatible with different research needs, providing a new direction for scientific research equipment design.

[0062] Change the inherent concept of manual analysis of data: the traditional concept believes that manual analysis of data is more reliable, ignoring the problems of low efficiency and strong subjectivity. The invention introduces AI intelligent analysis based on deep learning algorithm to automatically process multi-modal data, improve the objectivity and accuracy of analysis results, and overcome the technical bias of excessive dependence on manual analysis of data. BRIEF DESCRIPTION OF DRAWINGS

[0063] Figure 1 is the structure diagram of the multi-modal small animal anesthesia and consciousness research device and system assisted by artificial intelligence provided by the embodiment of the invention;

[0064] Figure 2 is the structure diagram of the vital sign monitoring module provided by the embodiment of the invention;

[0065] Figure 3 is the flow chart of the multi-modal small animal anesthesia and consciousness test method assisted by artificial intelligence provided by the embodiment of the invention;

[0066] Figure 4 is the representative synchronous recording result of EEG / EMG, calcium signal, heart rate, body temperature, respiration and behavior posture provided by the embodiment of the invention;

[0067] Figure 5 is the column chart of the influence of different anesthetic drugs on anesthesia state index provided by the embodiment of the invention;

[0068] Figure 6 is the multi-modal data trend chart in the anesthesia process provided by the embodiment of the invention;

[0069] Figure 7 is the behavior recovery time comparison box plot provided by the embodiment of the invention;

[0070] Figure 8 is the effect diagram of synchronous recording of each module index of the animal in the awake state provided by the embodiment of the invention;

[0071] Figure 9 is the effect diagram of synchronous recording of each module index of the animal in the anesthesia state provided by the embodiment of the invention;

[0072] In the figure: 1, wireless infrared temperature detector; 2, wireless millimeter wave radar detector; 3, high-precision pressure sensor. DETAILED DESCRIPTION

[0073] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0074] like Figure 1 As shown, this embodiment of the invention provides an artificial intelligence-assisted multimodal small animal anesthesia and consciousness research system, the system comprising:

[0075] An integrated monitoring chamber, comprising a transparent experimental chamber and a soundproof dark chamber;

[0076] EEG / EMG recording module, multi-channel electrode array, to record electrical activity of the cerebral cortex and muscles;

[0077] Fiber optic calcium imaging module, consisting of a fiber optic probe and a high-sensitivity photodetector, for real-time monitoring of neuronal calcium signals;

[0078] The drug delivery module features a programmable microinfusion pump that supports intravenous / peritoneal and nasal mask drug delivery.

[0079] The vital signs monitoring module includes a wireless infrared body temperature detector, a wireless millimeter-wave radar detector, and a high-precision pressure sensor;

[0080] The behavior capture module, infrared high-speed camera, and attitude analysis system monitor behaviors such as righting reflections;

[0081] The AI ​​computing and analysis platform integrates data acquisition, processing, and visualization functions, and uses artificial intelligence algorithms to analyze multimodal data.

[0082] The upper layer of the integrated monitoring cabin is a transparent experimental box for placing experimental animals and installing EEG / EMG electrodes, fiber optic probes, wireless infrared body temperature detectors, wireless millimeter-wave radar detectors, and infrared high-speed cameras.

[0083] The lower layer is a soundproof dark box, where a micro-injection pump is installed to ensure accurate drug delivery and environmental stability;

[0084] The EEG / EMG recording module uses 6-channel nickel-chromium alloy electrodes with an impedance of 3-4 MΩ. It records EEG at 0.5-100 Hz and EMG at 10-500 Hz, with a sampling rate of 2000 Hz, and supports real-time spectrum analysis.

[0085] The fiber optic calcium imaging module has a fiber optic probe with a diameter of 200 μm, NA of 0.37, connected to a GCaMP6s fluorescent indicator, excited by a 470 nm LED, recording calcium signals, with a sampling rate of 100 Hz and a resolution of ±0.1 ΔF / F.

[0086] The micro-injection pump in the administration module realizes intravenous or intraperitoneal administration at a flow rate of 0.1-100 μL / min, supports isoflurane atomization delivery, and has a concentration of 0.5%-5%.

[0087] As shown in Figure 2 The vital sign monitoring module comprises:

[0088] The wireless infrared temperature detector 1 has a resolution of ±0.1℃ and a range of 30-40℃.

[0089] The wireless millimeter wave radar detector 2 has a heart rate of 100-1000 bpm, an error of ±2 bpm, and a respiration of 30-300 times / minute, an error of ±2 times / minute.

[0090] The high-precision pressure sensor 3 monitors chest and abdominal movements and has a sensitivity of 0.01 mm.

[0091] The infrared high-speed camera in the behavior capture module has a resolution of 1920x1080 pixels, 120 fps, and captures positive and negative reflections, limb movement trajectories, and head posture, and cooperates with AI posture analysis to have a resolution of ±0.1 mm / s.

[0092] The AI operation and analysis platform integrates multi-modal data acquisition, has a synchronization accuracy of <10 ms, analyzes electroencephalogram spectrum, calcium signal waveform, vital sign trend and behavior characteristics based on CNN and RNN algorithms, and generates an anesthesia state index in real time, such as a loss of consciousness time point.

[0093] As shown in Figure 3 The embodiment of the present application provides a kind of artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method based on the artificial intelligence assisted multi-modal small animal anesthesia and consciousness research device and system of the present application, which specifically comprises:

[0094] S1: experimental preparation

[0095] Place the experimental animal on the upper layer of the integrated monitoring cabin, fix the electroencephalogram / electromyogram electrode and the optical fiber probe, and ensure that all sensors and administration modules are in working condition;

[0096] S2: baseline recording

[0097] Record baseline data for 5-10 minutes before anesthesia, including electroencephalogram / electromyogram, optical fiber calcium signal, heart rate, respiration, body temperature and behavior posture;

[0098] S3: anesthesia induction

[0099] Automatically deliver anesthetic drugs through the administration module, such as intravenous injection of propofol 20 mg / kg, and simultaneously record multi-modal data;

[0100] S4: anesthesia maintenance and recovery

[0101] Adjust the drug dose or concentration, such as propofol 10mg / kg intravenous injection, monitor the maintenance phase and recovery process after drug withdrawal;

[0102] S5: Data processing and analysis

[0103] The AI platform extracts and analyzes the features of the multi-modal data, generates anesthesia induction, maintenance and recovery curves, neural activity patterns and behavior activity time points;

[0104] S6: Result storage and visualization

[0105] All data is stored in real time in the database, and visual reports are generated, including EEG curves, time-frequency diagrams, power spectrum diagrams, electromyography, sleep phase diagrams, calcium signal curves, and vital sign curves.

[0106] The embodiment of the present application provides a computer device, characterized in that the computer device comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method.

[0107] The embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to make the processor execute the steps of the artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method.

[0108] The embodiment of the present application provides an information data processing terminal for realizing the artificial intelligence assisted multi-modal small animal anesthesia and consciousness research device and system.

[0109] The embodiment of the present application obtains the related evidence of the technical effect.

[0110] Embodiment: Evaluation of new anesthetic drug research and development

[0111] The artificial intelligence assisted multi-modal small animal anesthesia and consciousness research device and system of the present application are used to evaluate the efficacy of a new intravenous anesthetic drug.

[0112] Experimental preparation: 50 healthy mice are selected and placed on the upper layer of the integrated monitoring cabin, the electroencephalogram / electromyogram electrode and optical fiber probe are fixed, and it is ensured that all sensors and drug delivery modules work normally.

[0113] Baseline recording: Before anesthesia, the system records 8 minutes of baseline data, covering electroencephalogram / electromyogram, optical fiber calcium signal, heart rate, respiration, body temperature and behavior posture, etc.

[0114] Anesthesia induction and maintenance: through the administration module, the new anesthetic drug is automatically delivered at a specific rate, while adjusting the dosage for anesthesia maintenance, and the multi-modal data is recorded synchronously throughout the process. Figure 4

[0115] Figure 5 is a bar chart of the influence of different anesthetic drugs on the anesthesia state index provided by the embodiments of the present application;

[0116] Figure 6 is a multi-modal data trend chart during anesthesia provided by the embodiments of the present application;

[0117] Figure 7 is a behavior recovery time comparison box plot provided by the embodiments of the present application;

[0118] Figure 8 is an effect diagram of synchronous recording of each module index under the animal wakeful state provided by the embodiments of the present application;

[0119] Figure 9 is an effect diagram of synchronous recording of each module index under the animal anesthesia state provided by the embodiments of the present application;

[0120] Data processing and analysis: the AI operation and analysis platform performs deep processing and fusion analysis on the collected multi-modal data, generating detailed anesthesia depth curves, neural activity patterns, and behavior recovery time points. Based on these data, researchers clearly understand the specific effects of the new drug on mouse neural activity, physiological state, and behavioral response. For example, it is found that the drug can quickly reduce the frequency and amplitude of brain electrical signals during the induction stage, and cause muscle relaxation; during the maintenance stage, by monitoring vital signs, a suitable drug concentration range is determined, which can ensure anesthesia effect and maintain stable physiological state of mice.

[0121] Result evaluation: through comprehensive analysis of multi-modal data, researchers find that the new anesthetic drug performs well in anesthesia induction speed and maintenance effect, but some mice show short-term behavioral abnormalities during the recovery stage. Based on these results, the drug development team adjusts the drug formula accordingly, greatly improving the research and development efficiency, avoiding blind attempts, and saving a lot of time and cost.

[0122] ​It should be noted that embodiments of the present application can be realized by hardware, software, or a combination of software and hardware. The hardware portion can be realized by application-specific logic; application-specific logic refers to hardware circuits that are specifically designed and optimized for a particular application or function. Unlike general-purpose processors, which have a wide range of applicability, application-specific logic is tailored for a specific class of computational tasks or functions, resulting in higher performance, lower power consumption, and smaller area occupation. The software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a specially designed hardware. Those skilled in the art can understand that the above-mentioned devices and methods can be realized using computer executable instructions and / or included in processor control code, such as provided on a carrier medium, such as a USB flash disk, a solid state disk (SSD), a programmable memory, such as a read-only memory (firmware), or a data carrier, such as an optical or electronic signal carrier. The devices of the present application and their modules can be realized by hardware circuits, such as very large scale integrated circuits or gate arrays, semiconductors, such as logic chips, transistors, etc., or programmable hardware devices, such as field programmable gate arrays, programmable logic devices, etc., by software executed by various types of processors, or by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0123] The above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any modification, equivalent replacement, and improvement made by those skilled in the art within the technical scope disclosed by the present application, as long as it is within the spirit and principle of the present application, should be covered within the protection scope of the present application.

Claims

1. An artificial intelligence-assisted multi-modal small animal anesthesia and consciousness study system, characterized in that, The system comprises: An integrated monitoring cabin, including a transparent experimental box and a soundproof dark box; An electroencephalogram / electromyogram recording module, a multi-channel electrode array, recording the electrical activity of the cerebral cortex and muscles; A fiber-optic calcium imaging module, a fiber-optic probe and a high-sensitivity photodetector, real-time monitoring of neuronal calcium signals; A drug delivery module, a programmable micro-syringe pump, supporting peripheral intravenous, intraperitoneal and inhalation drug delivery; A vital sign monitoring module, including a wireless infrared temperature probe, a wireless millimeter wave radar probe and a high-precision pressure sensor; A behavior capture module, an infrared high-speed camera and a posture analysis system, monitoring behaviors such as righting reflex; An AI computing and analysis platform, integrating data acquisition, processing and visualization functions, using artificial intelligence algorithms to analyze multi-modal data.

2. The artificial intelligence-assisted multi-modal small animal anesthesia and conscious study system according to claim 1, wherein, The upper layer of the integrated monitoring cabin is a transparent experimental box, where experimental animals are placed, and electroencephalogram / electromyogram electrodes, fiber-optic probes, wireless infrared temperature probes, wireless millimeter wave radar probes and infrared high-speed cameras are installed; The lower layer is a soundproof dark box, where a micro-syringe pump is installed as a drug delivery module, ensuring the accuracy of drug delivery and environmental stability; The electroencephalogram / electromyogram recording module uses 6-channel nichrome electrodes with an impedance of 3-4 MΩ, recording EEG at 0.5-100 Hz and EMG at 10-500 Hz, with a sampling rate of 2000 Hz, supporting real-time spectral analysis.

3. The artificial intelligence-assisted multi-modal small animal anesthesia and conscious study system of claim 1, wherein, In the fiber-optic calcium imaging module, the fiber-optic probe has a diameter of 200 μm and a NA of 0.37, and a 470 nm LED is used to excite calcium fluorescence indicators such as GCaMP6s to record calcium signals at a sampling rate of 100 Hz and a resolution of ±0.1△F / F0; In the drug delivery module, the micro-syringe pump has a flow rate of 0.1-100 μL / min, enabling intravenous or intraperitoneal drug delivery, and supporting isoflurane aerosol delivery at a concentration of 0.5%-5%.

4. The artificial intelligence-assisted multi-modal small animal anesthesia and conscious study system of claim 1, wherein, The vital sign monitoring module includes: Wireless infrared temperature probe: resolution ±0.1℃, range 30-40℃; Wireless millimeter wave radar probe: heart rate 100-1000bpm, error ±2bpm, respiration 30-300 times / minute, error ±2 times / minute; High-precision pressure sensor: monitors chest and abdominal movements, sensitivity 0.01mm.

5. The artificial intelligence-assisted multi-modal small animal anesthesia and conscious study system of claim 1, wherein, In the behavior capture module, the infrared high-speed camera has a resolution of 1920x1080 pixels at 120 fps, capturing righting reflex, limb movement trajectory and head posture, and cooperating with AI posture analysis with a resolution of ±0.1mm / s.

6. The artificial intelligence-assisted multi-modal small animal anesthesia and conscious study system of claim 1, wherein, The AI computing and analysis platform integrates multi-modal data acquisition with a synchronization accuracy of <10ms, and uses CNN and RNN algorithms to analyze electroencephalogram frequency spectrum, power spectrum, sleep phase, electromyogram waveform, calcium signal waveform, vital sign trends and behavior characteristics, and generates an anesthesia state index in real time, such as the time point of loss of consciousness.

7. An artificial intelligence assisted multi-modal small animal anesthesia and consciousness testing method based on the artificial intelligence assisted multi-modal small animal anesthesia and consciousness research system of claims 1-6. The method specifically includes: Step 1: Experimental preparation Place the experimental animal on the upper layer of the integrated monitoring cabin, fix the electroencephalogram / electromyogram electrodes and fiber-optic probes, and ensure that all sensors and drug delivery modules are in working condition; Step 2: Baseline recording Record baseline data for 5-10 minutes before anesthesia, including electroencephalogram / electromyogram, fiber-optic calcium signal, heart rate, respiration, body temperature and behavior posture; Step 3: Anesthesia induction Automatically deliver anesthetic drugs, such as propofol 20 mg / kg intravenous injection, simultaneously record multi-modal data through the administration module; Step 4: Anesthesia maintenance and recovery Adjust the drug dosage or concentration, such as propofol 10 mg / kg intravenous injection, monitor the maintenance phase and recovery process after drug withdrawal; Step 5: Data processing and analysis AI platform for multi-modal data feature extraction and fusion analysis, generate anesthesia induction, maintenance and recovery curve, neural activity pattern and behavior activity time point; Step 6: Result storage and visualization All data are stored in real time to the database, generate visualization report, EEG curve, time-frequency graph, power spectrum graph, electromyogram, sleep phase graph, calcium signal curve, vital signs curve, etc.

8. A computer device, comprising: The computer device comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method according to any one of claim 7.

9. A computer readable storage medium, storing a computer program, the computer program is executed by a processor to make the processor execute the steps of the artificial intelligence assisted multi-modal small animal anesthesia and consciousness test method according to any one of claim 7.

10. An information data processing terminal, characterized by The information data processing terminal is used to realize the artificial intelligence assisted multi-modal small animal anesthesia and consciousness research system according to any one of claim 1-6.

Images