An intelligent airway training model and its operation evaluation method and system

Abstract

This application relates to the field of medical simulation teaching technology, and more specifically, to an intelligent airway training model and its operational evaluation method and system. The method includes: acquiring raw sensor data of intubation operations based on sensor components in the intelligent airway training model; preprocessing the raw sensor data to obtain operational data; acquiring video data corresponding to the intubation operation using a video acquisition component; evaluating the operational data and video data based on evaluation rules to obtain an operational score; acquiring patient characteristics; acquiring supplementary characteristics based on the operational data, and fusing the supplementary characteristics with the patient characteristics to obtain fused characteristics; acquiring matched cases that match the fused characteristics based on a matching model, and obtaining optimization suggestions for the intubation operation based on the matched cases and supplementary characteristics; and generating an evaluation report based on the operational score and optimization suggestions. This method is used to improve the accuracy of evaluation and the effectiveness of medical teaching for simulated intubation operations based on an airway training model.

Description

Technical Field

[0001] This invention relates to the field of medical simulation teaching technology, and more specifically, to an intelligent airway training model and its operational evaluation method and system. Background Technology

[0002] Difficult airway management refers to a clinical situation where even routinely trained anesthesiologists or emergency responders encounter significant difficulties or even failures during mask ventilation, glottic exposure, or endotracheal intubation. If not managed promptly and effectively, this condition can rapidly lead to hypoxemia, brain injury, and even death. Therefore, it is considered a crucial skill to master in clinical anesthesia, emergency care, and intensive care. Currently, the main methods for managing difficult airways include the use of various visual laryngoscopes, fiberoptic bronchoscopy-guided intubation, supraglottic airway devices, and percutaneous emergency airway establishment. Among these, endotracheal intubation remains the most fundamental and widely used treatment measure.

[0003] In actual intubation procedures, operators often face complex and challenging airways due to a combination of factors, including limited mouth opening, neck movement disorders, anatomical abnormalities or edema in the pharynx and larynx, and tumor lesions. They must make precise judgments and coordinate fine motor skills within a limited timeframe. Skill acquisition under such high pressure relies heavily on repeated simulation training. However, performing difficult airway procedures directly on patients is both ethically unacceptable and carries extremely high medical risks. Therefore, developing and utilizing airway training models that can reproduce challenging clinical scenarios has become an essential choice for medical simulation teaching.

[0004] Current airway training models widely used in medical education still have significant shortcomings. At the equipment level, the various adjustment parameters of existing airway training models are independent of each other, making it impossible to achieve synergistic linkage of multi-anatomical dimensions and realistically simulate complex and difficult airway scenarios. For example, existing airway training models can only adjust the range of motion, such as mouth opening, in a single dimension and lack biomechanical feedback during intubation. At the same time, existing airway training models generally lack real-time data acquisition and quantitative evaluation methods for the entire intubation process, relying mostly on the trainee's subjective observation and scoring, making it difficult to form a closed-loop training mechanism. The teaching effect of using airway training models for simulated operations is poor, and the training effect is difficult to standardize and evaluate. Summary of the Invention

[0005] This invention provides an intelligent airway training model and its operational evaluation method and system, which are used to improve the evaluation accuracy and medical teaching effect of simulated intubation operations based on the airway training model.

[0006] According to a first aspect of this application, an intelligent airway training model is provided, the intelligent airway training model comprising a maxillary and mandibular bone assembly, simulated teeth, a simulated tongue assembly, a cervical spine assembly, a simulated larynx, a simulated trachea, a simulated epiglottic spring, a simulated glottic flap, an epiglottic-glottic bridging ring, an independent adjustment knob, a transmission box, a sensor assembly, and a data processing assembly. The simulated epiglottis spring section is connected to the simulated glottis bending section via the epiglottis glottis bridging ring; The independent adjustment knobs include an epiglottis-glottis angle adjustment knob, an epiglottis lifting force adjustment knob, a mouth opening adjustment knob, and a cervical flexion-extension angle adjustment knob; the epiglottis-glottis angle adjustment knob is mechanically connected to the epiglottis-glottis bridging ring through the transmission box; the epiglottis lifting force adjustment knob is mechanically connected to the simulated epiglottis spring unit through the transmission box; the mouth opening adjustment knob is mechanically connected to the maxilla and mandible assembly and the simulated tongue assembly through the transmission box; and the cervical flexion-extension angle adjustment knob is mechanically connected to the cervical spine assembly through the transmission box. The sensor assembly includes a plurality of sensors, which are disposed in one or more of the simulated teeth, the simulated tongue assembly, the cervical spine assembly, the simulated larynx, the simulated trachea, the simulated epiglottis spring, and the simulated glottis flap. The data processing component is electrically connected to the sensor component.

[0007] Understandably, the intelligent airway training model can accurately simulate the state of various difficult airways by independently adjusting the epiglottis-glottis angle, epiglottis lifting force, mouth opening, and cervical flexion-extension angle; at the same time, it can collect operational data in real time with the help of built-in sensors, providing reliable support for subsequent quantitative assessment and targeted training.

[0008] Optionally, the intelligent airway training model further includes a coordination adjustment knob; the coordination adjustment knob is mechanically connected to the epiglottis glottis bridging ring, the simulated epiglottis spring, the maxillary and mandibular components, the simulated tongue component, and the cervical spine component through the transmission box.

[0009] Understandably, the intelligent airway training model, by adding a collaborative adjustment knob, can adjust the epiglottis-glottis angle, epiglottis lifting force, mouth opening and cervical flexion-extension angle with one click, quickly reproducing typical difficult airway scenarios, significantly improving training efficiency and the convenience of scenario switching.

[0010] Optionally, the coordination adjustment knob is set with several levels, each level being set with a corresponding epiglottis-glottis angle, epiglottis lifting force, mouth opening degree, and cervical flexion-extension angle, so that the coordination adjustment knob can adjust the epiglottis-glottis bridging ring, the simulated epiglottis spring part, the maxillary and mandibular bone assembly, the simulated tongue assembly, and the cervical spine assembly based on the epiglottis-glottis angle, epiglottis lifting force, mouth opening degree, and cervical flexion-extension angle corresponding to the level.

[0011] Understandably, by presetting multiple settings for the coordination adjustment knob, each setting corresponds to a fixed set of parameters for epiglottis-glottis angle, lifting force, mouth opening, and cervical spine angle, it is possible to quickly switch between standardized difficult airway scenarios with one click, ensuring consistency of training conditions and significantly improving scenario configuration efficiency.

[0012] According to a second aspect of this application, a method for evaluating the operation of an intelligent airway training model, the method comprising: Based on the sensor components in the intelligent airway training model according to the first aspect of this application, raw sensing data of the intubation operation is collected at preset time intervals; the raw sensing data at each time interval is preprocessed to obtain corresponding sub-operation data, and all the sub-operation data are summarized to obtain operation data; The video acquisition component is used to acquire the video data corresponding to the intubation operation; The operation data and the video data are evaluated based on preset evaluation rules to obtain an operation score for the intubation operation; Obtain preset patient characteristics; obtain supplementary characteristics of the intubation operation based on the operation data, and fuse the supplementary characteristics with the patient characteristics to obtain fused characteristics; Based on a preset matching model, matching cases that match the fusion features are obtained from a preset case database. Based on the matching cases and the supplementary features, optimization suggestions for the intubation operation are obtained. An evaluation report is generated based on the operation score and the optimization suggestion information.

[0013] Understandably, by integrating operational data and video data, multi-dimensional quantitative scoring of intubation procedures was achieved, enabling objective and precise evaluation of operational quality. Simultaneously, based on the fusion characteristics of patient features and operational data, typical cases were matched and optimization suggestions were generated. Finally, an assessment report containing scores and optimization suggestions was formed, providing trainees with closed-loop and accurate feedback, significantly improving the relevance and teaching effectiveness of airway management training.

[0014] Optionally, the preprocessing of the raw sensing data for each time interval to obtain the corresponding sub-operation data includes: Obtain the current operation data and the corresponding first error; Based on the sub-operational data at the current moment, predict the predicted data for the next moment; The second error corresponding to the predicted score data at the next moment is evaluated based on the first error; and the correction weight is obtained based on the second error. Obtain the raw sensor data for the next moment, and correct the raw sensor data for the next moment based on the predicted data for the next moment and the correction weight to obtain the sub-operation data for the next moment. The second error is updated based on the corrected weights.

[0015] Understandably, by predicting the next moment's data based on the current moment's sub-operational data and errors, and using the corresponding error calculation correction weights to correct the raw sensor data for the next moment in real time, sensor noise and random interference can be effectively reduced, and the continuity and accuracy of operational data can be improved. At the same time, by cyclically updating the error based on the correction weights, the data preprocessing process becomes adaptive, better tracking the dynamic changes in the cannulation operation, and providing high-quality data support for subsequent accurate evaluation.

[0016] Optionally, the preset evaluation rules include one or more of the following: operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules, and the operation standard evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, and the safety risk evaluation rules are preset with corresponding weights; The evaluation of the operation data and the video data based on preset evaluation rules to obtain the operation score of the intubation operation includes: Based on the operation data and the video data, obtain the evaluation data corresponding to the operation specification evaluation rule, the force control accuracy evaluation rule, the operation integrity evaluation rule, and the safety risk evaluation rule; The initial score is obtained based on the operation specification evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, and the safety risk evaluation rules, along with the corresponding evaluation data. The total score is obtained based on the initial score and the corresponding weight.

[0017] Understandably, this assessment method, by pre-setting rules for four dimensions—operational procedures, precision of force control, operational integrity, and safety risks—and assigning weights to each, can comprehensively measure the quality of intubation procedures from multiple perspectives, avoiding the one-sidedness of evaluation based on a single indicator. After acquiring scores for each dimension based on sensor and video data, the total score is calculated by weighting the scores according to their respective weights. This ensures that the assessment results reflect both specialized skill levels and overall safety performance, significantly improving the scientific rigor and objectivity of airway training assessments.

[0018] Optionally, the step of obtaining matching cases that match the fusion features from a preset case database based on a preset matching model includes: The fusion features are normalized to obtain the preprocessed fusion features; The preprocessed fusion features are input into the matching model to obtain the matching degree between each case in the case database and the preprocessed fusion features, and cases with a matching degree greater than a preset threshold are selected as matching cases. And / or, the fusion features include several sub-features; the step of obtaining optimization suggestion information for the intubation operation based on the matched case and the supplementary features includes: Based on the matching model, obtain the preset sub-feature weights corresponding to each sub-feature in the fusion features; Optimization suggestions for the intubation procedure are obtained based on the sub-feature weights, the matched cases, and the supplementary features.

[0019] Understandably, by normalizing the fusion features and inputting them into the matching model, typical cases with a matching degree higher than a preset threshold can be efficiently screened from the case database, significantly improving the accuracy and intelligence of case matching. Simultaneously, based on the preset weights of each sub-feature, matched cases, and supplementary operational features, targeted optimization suggestions can be generated, enabling trainees to clearly identify key deviations between their own operations and standard cases, thereby receiving personalized improvement guidance. This mechanism effectively enhances the accuracy of training feedback and the completeness of the teaching loop.

[0020] Optionally, the method further includes: Adjustment scheme based on the matched case acquisition component; The intelligent airway training model is adjusted according to the component adjustment scheme to simulate the real state corresponding to the matched case; and the state of the components of the intelligent airway training model is obtained during the adjustment; the components of the intelligent airway training model include one or more of the following: epiglottic glottis bridging ring, simulated epiglottic spring, maxillary and mandibular bone components, simulated tongue component and cervical spine component; The status of the component is compared with the security parameters in the preset security threshold library to identify the faulty component; Based on the faulty component, obtain the fault alarm of the intelligent airway training model.

[0021] Understandably, in the process of automatically adjusting the components of the intelligent airway training model based on matched cases to reproduce the real anatomical state, the operating status of each component can be monitored in real time and compared with preset safety parameters. The faulty component can be automatically identified and located, thereby triggering a fault alarm. This not only realizes the intelligent reproduction of the training scenario, but also has the self-diagnosis function of the components, which can effectively prevent training distortion or equipment damage caused by component abnormalities, and significantly improve the reliability, safety and service life of the training model.

[0022] According to a third aspect of this application, an intelligent airway training model operation evaluation system is provided, the system comprising: The data acquisition module is used to acquire raw sensor data of the intubation operation according to a preset time interval based on the sensor components in the intelligent airway training model of the first aspect of this application; preprocess the raw sensor data of each time interval to obtain corresponding sub-operation data, summarize all the sub-operation data to obtain operation data; and acquire video data corresponding to the intubation operation using the video acquisition component. The operation score acquisition module is used to evaluate the operation data and the video data based on preset evaluation rules to obtain the operation score of the intubation operation; The fusion feature acquisition module is used to acquire preset patient features; acquire supplementary features of the intubation operation based on the operation data; and fuse the supplementary features with the patient features to obtain fusion features. The optimization suggestion acquisition module is used to acquire matching cases that match the fusion features in a preset case database based on a preset matching model, and to acquire optimization suggestion information for the intubation operation based on the matching cases and the supplementary features. The evaluation report acquisition module is used to generate an evaluation report based on the operation score and the optimization suggestion information.

[0023] According to a fourth aspect of this application, an electronic device is provided, comprising: Memory, used to store one or more computer programs; A processor, when the one or more computer programs are executed by the processor, implements the intelligent airway training model operation evaluation method described in the second aspect above.

[0024] According to a fifth aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the intelligent airway training model operation evaluation method described in the second aspect above.

[0025] Based on any of the above aspects, the intelligent airway training model and its operational evaluation method and system provided in this application can achieve the following effects: The intelligent airway training model can adjust its components to simulate real-life cases, improving training accuracy and convenience. By independently adjusting knobs or coordinating preset settings, it can quickly simulate complex and challenging airway scenarios with multiple overlapping factors, such as varying mouth opening, cervical spine mobility, and epiglottic elasticity—features characteristic of real clinical cases. Simultaneously, by combining sensor components to collect real-time component status data, it can automatically match and recreate specific case states based on component status. This avoids the cumbersome manual adjustments, replacement of faulty parts, or rough simulations required by traditional models, significantly improving training accuracy and the ease of scenario switching. This allows trainees to conduct efficient and repetitive training on diverse airway anatomical variations.

[0026] To improve training effectiveness, a multi-dimensional scoring system is implemented for training based on an intelligent airway training model. Sensors deployed on multiple components, including simulated teeth, tongue, larynx, epiglottis spring, and glottis flap, collect raw sensor data of the intubation procedure at preset time intervals. This data is then combined with video data for multi-dimensional quantitative evaluation. Pre-defined evaluation rules cover four sub-rules: operational standardization, force control precision, operational completeness, and safety risks. Each rule is assigned an independent weight, and the final weighted score is calculated. This refined multi-dimensional scoring system overcomes the limitations of traditional models that rely solely on subjective or qualitative evaluation. It accurately pinpoints specific errors made by trainees during the procedure, guiding targeted improvements and significantly enhancing the objectivity and learning effectiveness of the training.

[0027] To enhance teaching effectiveness, optimized suggestions are generated by combining matched cases: a matching model is used to fuse supplementary features extracted from trainees' intubation procedures with pre-defined patient characteristics, and then matched with typical cases in a case database to automatically identify the most relevant difficult airway cases. Simultaneously, optimized suggestions are generated based on the matched cases and supplementary features, such as how to adjust tongue position, cervical spine angle, or epiglottic lifting force for specific cases, and an assessment report including procedure scores and improvement strategies is generated. This data-driven, personalized teaching feedback mechanism effectively enhances the relevance and depth of teaching, significantly improving the training outcome for difficult airway management skills. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a schematic diagram of the overall structure of an intelligent airway training model provided in this embodiment.

[0030] Figure 2 This is a schematic diagram showing the detailed structure of the simulated head of an intelligent airway training model provided in this embodiment.

[0031] Figure 3 This is a schematic diagram of the simulated trachea detailed structure of an intelligent airway training model provided in this embodiment.

[0032] Figure 4 This is a schematic diagram showing the detailed structure of the epiglottis in a simulated intelligent airway training model provided in this embodiment.

[0033] Figure 5 This is a flowchart of an operational evaluation method for an intelligent airway training model provided in this embodiment.

[0034] Figure 6 This is a flowchart for preprocessing raw sensor data provided in this embodiment.

[0035] Figure 7 This is a flowchart for scoring the cannulation operation provided in this embodiment.

[0036] Figure 8 This is a flowchart for generating fault alarms provided in this embodiment.

[0037] Figure 9 This is a schematic diagram of the functional modules of an intelligent airway training model operation evaluation system provided in this embodiment.

[0038] Figure 10 This is a schematic diagram of the structure of the electronic device provided in this embodiment.

[0039] Reference numerals: 11-Independent adjustment knob, 111-Collaborative adjustment knob, 112-Epiglottis-Glottis angle adjustment knob, 113-Epiglottis lifting force adjustment knob, 114-Mouth opening adjustment knob, 115-Cervical spine flexion-extension angle adjustment knob, 12-Transmission box, 13-Data processing component, 14-Display screen, 15-Voice broadcast component, 21-Simulated skull, 22-Simulated skin, 23-Simulated tongue component, 231-Simulated tongue body, 232-Connecting tube, 233-Valve device, 24-Maxillary and mandibular bone components, 241-Temporomandibular joint component, 242-Simulated teeth, 243-Maxilla, 244-Mandible, 25-Side cover, 3-Cervical spine component, 41-Simulated larynx, 42-Simulated trachea, 431-Simulated epiglottis spring, 432-Simulated glottis flap, 433-Epiglottis-glottis bridging ring. Detailed Implementation The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this application. To better illustrate the following embodiments, some components in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product; it is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0040] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0041] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0042] Existing airway training models have independent adjustments to various anatomical parameters, making it impossible to coordinate and link them to realistically simulate complex and difficult airway scenarios with multiple overlapping factors. At the same time, they generally lack real-time data acquisition and quantitative evaluation methods for the entire intubation process, relying on subjective observation and scoring, which makes it difficult to form a closed-loop training mechanism, resulting in poor teaching effectiveness and the inability to standardize the evaluation of training quality.

[0043] This embodiment provides a technical solution that can solve the above problems. The specific implementation of this application will be described in detail below with reference to the accompanying drawings.

[0044] like Figure 1 As shown, an intelligent airway training model is provided. The intelligent airway training model includes a maxillary and mandibular bone assembly 24, simulated teeth 242, simulated tongue assembly 23, cervical spine assembly 3, simulated larynx 41, simulated trachea 42, simulated epiglottic spring 431, simulated glottic flap 432, epiglottic-glottic bridging ring 433, independent adjustment knob 11, transmission box 12, sensor assembly, and data processing assembly 13; The simulated epiglottis spring part 431 is connected to the simulated glottis bent plate part 432 through the epiglottis glottis bridging ring 433; The independent adjustment knob 11 includes an epiglottis-glottis angle adjustment knob 112, an epiglottis lifting force adjustment knob 113, a mouth opening adjustment knob 114, and a cervical flexion-extension angle adjustment knob 115. The epiglottis-glottis angle adjustment knob 112 is mechanically connected to the epiglottis-glottis bridging ring 433 through the transmission box 12. The epiglottis lifting force adjustment knob 113 is mechanically connected to the simulated epiglottis spring part 431 through the transmission box 12. The mouth opening adjustment knob 114 is mechanically connected to the maxilla and mandible assembly 24 and the simulated tongue assembly 23 through the transmission box 12. The cervical flexion-extension angle adjustment knob 115 is mechanically connected to the cervical spine assembly 3 through the transmission box 12. The sensor assembly includes a plurality of sensors, which are disposed in one or more of the simulated teeth 242, the simulated tongue assembly 23, the cervical spine assembly 3, the simulated laryngeal cavity 41, the simulated trachea 42, the simulated epiglottic spring part 431, and the simulated glottic flap part 432. The data processing component 13 is electrically connected to the sensor component.

[0045] In this embodiment, the simulated tongue assembly 23 is a detachable assembly. Preferably, the simulated tongue assembly 23 includes a simulated tongue body 231, a connecting pipe 232, and a valve device 233. The simulated tongue body 231 is connected to the valve device 233 via the connecting pipe 232. The simulated tongue body 231 is an inflatable structure. Gas is input or released through the connecting pipe 232 and the valve device 233 to adjust the volume of the simulated tongue body 231. For example, the volume of the simulated tongue body 231 without gas injection is 60 cm³. If gas is injected into the simulated tongue body 231, its volume is the sum of the injected air volume and the base volume.

[0046] Preferably, the intelligent airway training model further includes a side cover 25 for concealing internal components to simulate a normal patient state. It is understood that the side cover 25 can be disassembled to allow the simulated tongue assembly 23 to be removed via the corresponding location on the disassembled side cover 25.

[0047] Preferably, the maxillary and mandibular components 24 in the intelligent airway training model include a temporomandibular joint component 241, a maxilla 243, and a mandible 244, wherein the maxilla 243 is connected to the mandible 244 by controlling the temporomandibular joint component 241.

[0048] Preferably, the maxillary and mandibular bone assembly 24 is an adjustable assembly, which, in conjunction with the simulated tongue assembly 23, can adjust the mouth opening of the intelligent airway training model. Preferably, the mouth opening can be adjusted by adjusting the relative movement of the maxilla 243, mandible 244, and simulated tongue 231 in the maxillary and mandibular bone assembly 24. Exemplarily, the mouth opening can be dynamically adjusted from 0 cm to 6 cm by adjusting the relative movement of the maxilla 243, mandible 244, and simulated tongue assembly 23 in the maxillary and mandibular bone assembly 24.

[0049] In this embodiment, the epiglottis-glottis angle adjustment knob 112, the epiglottis lifting force adjustment knob 113, the mouth opening adjustment knob 114, and the cervical flexion-extension angle adjustment knob 115 of the independent adjustment knob 11 are each equipped with a spring-type locking member, and are respectively meshed with the driven gear in the transmission box 12. For example, the module of the driven gear is 0.5.

[0050] For example, the epiglottis-glottis angle adjustment knob 112, the epiglottis lifting force adjustment knob 113, the mouth opening adjustment knob 114, and the cervical flexion-extension angle adjustment knob 115 of the independent adjustment knob 11 are respectively connected to the corresponding driven gears via high-strength nylon transmission ropes with a diameter of 1 mm and a tensile strength ≥50 N; the rotation angle range of the epiglottis-glottis angle adjustment knob 112, the epiglottis lifting force adjustment knob 113, the mouth opening adjustment knob 114, and the cervical flexion-extension angle adjustment knob 115 is... 180°~+180°, supporting fine-tuning of small angles. When it is set to 0°, the corresponding epiglottis glottis bridging ring 433, the simulated epiglottis spring part 431, the maxillary and mandibular bone assembly 24, the simulated tongue assembly 23 and the cervical spine assembly 3 are in the reference initial position.

[0051] For example, the rotation angles of the epiglottis-glottis angle adjustment knob 112, the epiglottis lifting force adjustment knob 113, the mouth opening adjustment knob 114, and the cervical flexion-extension angle adjustment knob 115 have the following preset mapping relationship with the corresponding controlled component parameters: When the rotation angle of the epiglottis-glottis angle adjustment knob 112 is 0°, the reference angle between the simulated epiglottis spring part 431 and the simulated glottis bend part 432 adjusted by the epiglottis-glottis bridging ring 433 is 30°, and the effective adjustment angle range of the simulated epiglottis spring part 431 and the simulated glottis bend part 432 is 15° to 45°. For every 60° rotation of the epiglottis-glottis angle adjustment knob 112, the angle change of the simulated epiglottis spring part 431 and the simulated glottis bend part 432 is 5°.

[0052] When the rotation angle of the epiglottis lifting force adjustment knob 113 is 0°, the reference force on the simulated epiglottis spring part 431 is 20N, and the effective adjustment range of the force on the simulated epiglottis spring part 431 is 5N to 35N; every 60° rotation of the epiglottis lifting force adjustment knob 113 corresponds to a change of 5N in the force on the simulated epiglottis spring part 431.

[0053] The mouth opening adjustment knob 114 has a rotation angle of 0° corresponding to a mouth opening of 3cm, and the effective adjustment range of the mouth opening is 0cm to 6cm. Furthermore, the mouth opening adjustment knob 114 changes by 1cm for every 60° rotation.

[0054] The cervical flexion-extension angle adjustment knob 115, when rotated to 0°, corresponds to a 0° bending angle for the cervical spine component 3, and the effective adjustment range of the bending angle of the cervical spine component 3 is as follows: 45°~+45°, where 45° represents the bending angle of the cervical spine component 3 when it is flexed forward by 45° relative to its normal, unbent state, and 45° represents the bending angle of the cervical spine component 3 when it is tilted backward by 45° relative to its normal, unbent state. Furthermore, every 60° rotation of the cervical spine flexion / extension angle adjustment knob 115 corresponds to a 15° change in the bending angle of the cervical spine component 3.

[0055] In this embodiment, a power interface is provided on the back of the intelligent airway training model base, with an input voltage of 220 V and an output voltage of 12 V.

[0056] In this embodiment, the data processing component 13 may employ an STM32F407 main control chip and integrate a wireless communication component and a data storage unit. The wireless communication component supports Wi-Fi and / or Bluetooth transmission, and the data storage unit has a capacity of ≥16 GB. The data storage unit can store several typical clinical cases and corresponding model parameters, and can also store the sensor parameters corresponding to the standard intubation procedures of the typical clinical cases. Preferably, the data processing component 13 can also be electrically connected to the transmission box 12.

[0057] Preferably, the intelligent airway training model further includes a display screen 14, electrically connected to the data processing component 13. Preferably, the display screen can be a touch screen, enabling full-process touch operation. Exemplarily, the touch screen can be a 7.7-inch multi-touch screen with a resolution of 1280×720, communicatively connected to the data processing component 13 via a ribbon cable. The display screen 14 is used to display visual content such as parameters of each component of the intelligent airway training model and evaluation reports, and can provide independent adjustment knobs and / or collaborative adjustment knobs for virtual touch, allowing the data processing component 13 to adjust the adjustable components of the intelligent airway training model, providing another adjustment method for the adjustable components. The adjustable components can be the epiglottic-glottic bridging ring 433, the simulated epiglottic spring part 431, the maxillary and mandibular bone components 24, the simulated tongue component 23, and the cervical spine component 3 of the intelligent airway training model. Preferably, the touch display screen 14 may also be equipped with an Artificial Intelligence (AI) scene mode and a custom scene mode: in the AI ​​scene mode, it supports mode switching for different case scenarios based on artificial intelligence, and the AI-based scene matching response time is ≤1 s. In the custom scene mode, the adjustable components of the intelligent airway training model can be adjusted via independent adjustment knobs and / or collaborative adjustment knobs via virtual touch, with an adjustment accuracy of ±0.1 cm / ±1°.

[0058] Preferably, the intelligent airway training model further includes a voice broadcast component 15, which is used to broadcast evaluation information and prompts for the intubation operation before, during, and after the intubation operation. The voice broadcast component 15 is electrically connected to the data processing component 13. For example, the voice broadcast component 15 has a speaker power of 2W, supports switching between Chinese and English, and has a voice broadcast delay of ≤0.3 s.

[0059] Preferably, the intelligent airway training model further includes a simulated skull 21 and simulated skin 22, which together with other components of the intelligent airway training model form a realistic simulated human head model, enabling trainees to perform intubation operations in a more realistic clinical setting.

[0060] Specifically, the sensor assembly includes one or more of the following: a first sensor disposed on the simulated tongue assembly 23; a second sensor disposed on the simulated laryngeal cavity 41; a third sensor disposed on the back of the simulated glottis flap portion 432; a fourth sensor disposed on the simulated teeth 242; a fifth sensor disposed on the simulated epiglottic spring portion 431; a sixth sensor disposed at the end of the simulated trachea 42; and a seventh sensor disposed on the cervical spine assembly 3. Preferably, the first sensor is a pressure sensor, which can be disposed on the simulated tongue 231 in the simulated tongue assembly 23, for detecting the pressure of the simulated tongue. The second sensor is a pressure sensor, disposed on the simulated laryngeal cavity 41, for detecting the pressure of the simulated laryngeal cavity 41. The third sensor is a pressure sensor, for detecting the pressure of the simulated glottis flap 432. The fourth sensor is a pressure sensor, for detecting the pressure of the simulated teeth 242. The fifth sensor is a pressure sensor, for detecting the pressure of the simulated epiglottis spring 431. The sixth sensor is a touch sensor, for sensing the tracheal insertion status at the end of the simulated trachea 42. The seventh sensor is a curvature sensor, for obtaining the specific degree of curvature of the cervical spine assembly 3.

[0061] Preferably, the intelligent airway training model may further include a simulated mouth, and the simulated mouth may also be equipped with an eighth sensor for detecting the pressure value of the mouth.

[0062] Preferably, the sampling accuracy of the sensor assembly is displacement ±0.01 cm and angle ±0.1°.

[0063] Specifically, the intelligent airway training model also includes a coordination adjustment knob 111; the coordination adjustment knob is mechanically connected to the epiglottis glottis bridging ring 433, the simulated epiglottis spring part 431, the maxillary and mandibular bone assembly 24, the simulated tongue assembly 23 and the cervical spine assembly 3 through the transmission box 12.

[0064] Specifically, the coordinated adjustment knob 111 is set with several gear positions, each gear position being set with a corresponding epiglottis-glottis angle, epiglottis lifting force, mouth opening degree, and cervical flexion-extension angle, so that the coordinated adjustment knob can adjust the epiglottis-glottis bridging ring 433, the simulated epiglottis spring part 431, the maxillary and mandibular bone assembly 24, the simulated tongue assembly 23, and the cervical spine assembly 3 based on the epiglottis-glottis angle, epiglottis lifting force, mouth opening degree, and cervical flexion-extension angle corresponding to the gear position.

[0065] For example, the coordinated adjustment knob is set with 0 to 4 difficulty levels for intubation operations, where level 0 is the initial position, and levels 1 to 4 correspond to the non-difficult, mildly difficult, moderately difficult, and severely difficult levels, respectively. Each level has a corresponding epiglottic-glottic angle, epiglottic lifting force, mouth opening, and cervical flexion-extension angle, and these parameters can be set and changed in the corresponding parameter boxes on the display screen 14. Preferably, it can also be further coordinated with the independent adjustment knob 11. It is understood that the adjustment method and logic of the coordinated adjustment knob on the epiglottis-glottis bridging ring 433, the simulated epiglottis spring part 431, the maxillary and mandibular bone assembly 24, the simulated tongue assembly 23 and the cervical spine assembly 3 are similar to those of the independent adjustment knob on the epiglottis-glottis bridging ring 433, the simulated epiglottis spring part 431, the maxillary and mandibular bone assembly 24, the simulated tongue assembly 23 and the cervical spine assembly 3, and can be referred to the description of the independent adjustment knob 11.

[0066] Preferably, when the coordination adjustment knob is set to level 1 (non-difficult level), the angle between the simulated epiglottic spring 431 and the simulated glottis flap 432 is set to 45°, the force on the simulated epiglottic spring 431 is 15 N + mouth opening 4 cm + cervical spine component tilt angle 30° + tongue volume 80 cm³. When the coordination adjustment knob is set to level 2 (slightly difficult level), the angle between the simulated epiglottic spring 431 and the simulated glottis flap 432 is set to 35°, the force on the simulated epiglottic spring 431 is 20 N + mouth opening 3 cm + cervical spine component tilt angle 15° + tongue volume 90 cm³. When the coordination adjustment knob is set to level 3, i.e., the moderate difficulty level, the angle between the simulated epiglottic spring part 431 and the simulated glottis flap part 432 is set to 25°, the force on the simulated epiglottic spring part 431 is 25N + mouth opening 3cm + cervical spine component 0° backward tilt angle and tongue volume 100 cm³. When the coordination adjustment knob is set to level 4, i.e., the severe difficulty level, the angle between the simulated epiglottic spring part 431 and the simulated glottis flap part 432 is set to 15°, the force on the simulated epiglottic spring part 431 is 30N + mouth opening 2cm + cervical spine component 15° forward flexion angle and tongue volume 110 cm³.

[0067] like Figure 5 As shown, this embodiment provides an operational evaluation method for an intelligent airway training model, which can be further divided into the following steps: S100. Based on the sensor components in the above-mentioned intelligent airway training model, the raw sensing data of the intubation operation is collected at a preset time interval; the raw sensing data of each time interval is preprocessed to obtain the corresponding sub-operation data, and all the sub-operation data are summarized to obtain the operation data; the video acquisition component is used to acquire the video data corresponding to the intubation operation. In this embodiment, the sensor components in the above-mentioned intelligent airway training model collect raw sensing data of the intubation operation at preset time intervals. The raw sensing data may include one or more of the following: tongue pressure value obtained by the first sensor, pharyngeal pressure value and mucosal pressure value obtained by the second sensor, epiglottic pressure value obtained by the third and fifth sensors, tooth pressure value obtained by the fourth sensor, tracheal end sensing value obtained by the sixth sensor, cervical spine curvature angle value obtained by the seventh sensor, and mouth corner pressure value obtained by the eighth sensor.

[0068] In this embodiment, the data collected by the sensor is collected at preset time intervals, enabling real-time acquisition of the intubation operation status. After acquiring the raw sensor data at the current time, the raw sensor data at the current time is preprocessed based on the raw sensor data at the previous time to obtain the sub-operation data at the current time. After the intubation operation is completed, all the obtained sub-operation data are summarized to obtain the operation data of the current intubation operation.

[0069] In this embodiment, it is necessary not only to acquire operational data based on the intelligent airway training model, but also to acquire video data corresponding to the intubation operation using a video acquisition component. It is understood that in this embodiment, the method further includes an external video acquisition component, which may include a video recording unit for filming the intubation operation of the trainee and / or a laryngoscope video recording unit. Specifically, the video recording unit of the video acquisition component records the pre-intubation operation of the trainee to obtain pre-intubation video data; the laryngoscope video recording unit of the video acquisition component records the laryngoscope content corresponding to the laryngoscope used for intubation to obtain laryngoscope video data. The video acquisition component is communicatively connected to the data processing unit in the intelligent airway training model, enabling the pre-intubation video data and laryngoscope video data to be sent to the data processing unit. In the data processing unit, the pre-intubation video data and laryngoscope video data are used as video data, and this video data serves as the data basis for subsequent evaluation.

[0070] In this embodiment, the Kalman filter algorithm can be used to preprocess the data for the division operation.

[0071] Specifically, such as Figure 6As shown, the preprocessing of the raw sensor data for each time interval to obtain the corresponding sub-operation data may include the following steps: S110. Obtain the sub-operation data at the current moment and the corresponding first error; S120. Based on the sub-operation data at the current moment, predict the prediction data for the next moment; Specifically, the prediction data for the next moment It can be: in, This represents the preset state transition matrix. This represents the sub-operation data at the current moment. This represents the preset control input matrix. Preset control input vector.

[0072] S130. Evaluate the second error corresponding to the predicted data at the next moment based on the first error; and obtain the correction weight based on the second error; Specifically, the second error corresponding to the predicted score data at the next moment. for: in This represents the first error corresponding to the sub-operation data at the current moment. This represents a fixed parameter indicating the inherent noise of the corresponding sensor component; The correction weight at the next time step : in, This represents the preset observation matrix. This indicates the fixed parameters for the accuracy calibration of the corresponding sensor component.

[0073] S140. Obtain the raw sensing data at the next moment, and correct the raw sensing data at the next moment based on the predicted data at the next moment and the correction weight to obtain the sub-operation data at the next moment. Specifically, the sub-operation data at the next moment for: in, This represents the raw sensor data at the next moment. This indicates the preset Kalman gain.

[0074] S150, Update the second error based on the corrected weights.

[0075] Specifically, the updated second error for: S200. Evaluate the operation data and the video data based on preset evaluation rules to obtain the operation score of the intubation operation; Specifically, the preset evaluation rules include one or more of the following: operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules, and the operation standard evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, and the safety risk evaluation rules are preset with corresponding weights; Preferably, in this embodiment, the preset evaluation rules include four types: operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules, and the weights of the operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules are 40%, 30%, 15%, and 15%, respectively.

[0076] Specifically, if the total score for the trainee's intubation procedure is 100 points, the evaluation rules can be set as follows: (1) Operational procedure assessment rules (40 points) (2) Rules for evaluating the accuracy of force control (30 points) (3) Complete operation assessment rules (15 points) (4) Safety Risk Assessment Rules (15 points) Preferably, in the scoring rules, the cumulative deduction for each sub-item does not exceed its own weight, and the deduction stops when the weight is exhausted. It is understood that this embodiment may also include rules for serious intubation errors. Serious errors directly leading to unqualified intubation include: based on laryngoscope video data and pharyngeal pressure values, a peak laryngoscope pressure >20 N or sustained above 15 N for >0.5 s; based on cervical spine flexion angle values, an absolute value of the cervical spine flexion-extension angle >45° or sustained deviation >1 s; based on tracheal end sensor values, failure to initiate correction for >30 s after the catheter is mistakenly inserted into the esophagus; based on laryngoscope video data and relevant laryngoscope parameters, a cuff inflation pressure >20 cmH2O for >1 s.

[0077] like Figure 7 As shown, the process of evaluating the operation data and the video data based on preset evaluation rules to obtain an operation score for the intubation operation may include the following steps: S210. Based on the operation data and the video data, obtain the evaluation data corresponding to the operation specification evaluation rule, the force control accuracy evaluation rule, the operation integrity evaluation rule, and the safety risk evaluation rule; In this embodiment, different evaluation rules require scoring based on different evaluation data. Therefore, it is necessary to obtain the corresponding evaluation data based on the scoring rules corresponding to the operation specification evaluation rule, the force control accuracy evaluation rule, the operation integrity evaluation rule, and the safety risk evaluation rule described above.

[0078] S220. Obtain the corresponding initial score according to the operation specification evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, the safety risk evaluation rules, and the corresponding evaluation data; S230. Obtain the total score based on the initial score and the corresponding weight.

[0079] In this embodiment, the corresponding initial scores are obtained from the operation specification evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, and the safety risk evaluation rules, and then the total score of the intubation operation is obtained by weighting them according to the corresponding weights.

[0080] S300: Obtain preset patient characteristics; obtain supplementary characteristics of the intubation operation based on the operation data, and fuse the supplementary characteristics with the patient characteristics to obtain fused characteristics; In this embodiment, each intubation procedure corresponds to a set of patient characteristics, used to simulate the recording of the patient's basic characteristics during clinical intubation. Exemplarily, these patient characteristics may include the patient's anatomical features, physical signs, medical history, and auxiliary characteristics. The anatomical features may include the patient's initial mouth opening and initial cervical flexion-extension angle, etc.; the physical signs may include body mass index (BMI), patient age, etc.; the medical history may include whether the patient has hypertension and / or diabetes, etc.; and the auxiliary characteristics may include the patient's MacPherson strut classification, etc.

[0081] In this embodiment, after the trainee performs the intubation operation, the corresponding operation data can be obtained. The operation data is then used to extract features to obtain corresponding supplementary features. The supplementary features and the patient features are then fused to obtain fused features, providing a data foundation for subsequent matching work.

[0082] S400. Based on a preset matching model, obtain matching cases that match the fusion features from a preset case database, and obtain optimization suggestion information for the intubation operation based on the matching cases and the supplementary features. In this embodiment, the preset matching model can be a random forest model, and the matching model can be integrated into the data processing unit of the intelligent airway training model. Preferably, the matching model needs to be trained to obtain a trained matching model, and matched cases are obtained based on the trained matching model.

[0083] Selected from the case database M The cases were divided into training sets at a ratio of 70%–80% and 20%–30%. and test set Each case corresponds to a historical fusion feature. For example, the training set... The test set includes E cases. This includes F cases. It is understood that supplementary historical features of each case and corresponding historical patient features can be obtained and fused to obtain historical fused features. It is understood that the historical fused features may include... Each feature, the first The historical fusion features corresponding to each case can be: Furthermore, each training sample consists of historical fusion features and corresponding difficult airway labels, which can represent... ( , ,in For the first The difficult airway label corresponds to the historical fusion characteristics of each case.

[0084] The features in the training samples are classified: hierarchical features are converted to continuous values, binarized features are set to 1 / 0, and continuous features retain the measured values. Then, the min-max standardization formula is applied. All feature values ​​are normalized to the [0, 1] interval to eliminate dimensional differences, resulting in preprocessed historical fusion features.

[0085] Using the Bootstrap sampling method with replacement, from The training set for each decision tree in the random forest model is constructed by extracting samples from the middle. The samples that were not selected (outside the bag) are used to evaluate the generalization ability of the random forest model.

[0086] During the training of the random forest model, each time a node splits, it is from all... Randomly selected from historical fusion features Each feature is analyzed and rounded to avoid a single feature dominating the split. The optimal splitting pair is selected based on Gini impurity. ,in This represents the optimal splitting characteristic. This represents the optimal splitting threshold. And the node... The impurity of the gin is: in, This represents the Gini impurity function. This indicates the preset number of categories. For the node The proportion of class samples.

[0087] The conditional Gini impurity after splitting is: in, Indicates the feature used in the current split. This indicates the threshold used for the current split. Represents a node The left node after the split, Represents a node The right node after the split, Represents a node The number of samples included Indicates the left node The number of samples included Indicates the right node The number of samples included.

[0088] Recursively split nodes until the number of samples in a node is ≤ 5 or the Gini impurity is ≤ 0.01, at which point splitting stops and pruning is performed, ultimately resulting in a single decision tree. Build After planting trees, the final label is output through hard voting: in To match the prediction function of the model, This is an indicator function; it takes the value 1 if the condition is true, and 0 otherwise. For fusion features At that time, the first The fusion features of the trees on the input are: The predicted output; Labeled as a difficult airway.

[0089] The matching model is trained and tested using the above method to obtain a matching model that has completed training and testing.

[0090] In this embodiment, the preset case database can be stored in the intelligent airway training model. For example, the case database contains more than 1200 real case parameter combinations, covering ankylosing spondylitis, cervical spine injury, lower respiratory tract tumors, and sleep apnea syndrome. Each case type has multiple clinical cases corresponding to different patient types and clinical scenarios. The intelligent airway training model can also be modified by trainees or instructors to obtain actual parameters from real cases and manually adjust its components, enabling the creation of a new case and its storage in the case database for subsequent training or matching.

[0091] Specifically, the step of obtaining matching cases that match the fusion features from a preset case database based on a preset matching model includes: The fusion features are normalized to obtain the preprocessed fusion features; The preprocessed fusion features are input into the matching model to obtain the matching degree between each case in the case database and the preprocessed fusion features, and cases with a matching degree greater than a preset threshold are selected as matching cases.

[0092] Specifically, the fusion features include several sub-features; the step of obtaining optimization suggestion information for the intubation operation based on the matched cases and the supplementary features includes: Based on the matching model, obtain the preset sub-feature weights corresponding to each sub-feature in the fusion features; Optimization suggestions for the intubation procedure are obtained based on the sub-feature weights, the matched cases, and the supplementary features.

[0093] For example, in a trained matching model, by Obtain the preset sub-feature weights corresponding to each sub-feature, where This represents the total number of decision trees in the matching model. Indicates the first A tree.

[0094] pass Get matching degree ,in S ∈ [0,1], the closer to 1, the higher the matching degree.

[0095] Understandably, based on the obtained matched cases as a reference, and by comparing and calculating the supplementary features according to the weighted features, optimization suggestions for the supplementary features are obtained. For example, in the matched case, the force applied to the epiglottis during intubation in standard operation is a first force value, while the supplementary feature indicates a second force value, which is greater than the first force value. This supplementary feature allows for successful intubation in the matched case, but allows for more appropriate intubation procedures for the patient. Therefore, optimization suggestions can be displayed on the screen of the intelligent airway training model or announced via a voice broadcast component, such as the aforementioned optimization suggestion of appropriately reducing the force applied to the epiglottis during intubation.

[0096] S500: Generate an evaluation report based on the operation score and the optimization suggestion information.

[0097] In this embodiment, based on the operation score and optimization suggestion information obtained from this intubation operation, an evaluation report of the intubation operation is generated, and the optimization suggestions are displayed on the screen of the intelligent airway training model or broadcast through the voice broadcast component to provide real-time guidance and feedback for the trainee's intubation operation.

[0098] Specifically, such as Figure 8 As shown, the method further includes the following steps: S610. Adjustment scheme based on the matched case acquisition component; In this embodiment, based on the pathological parameter data corresponding to the matched cases, such as cervical stiffness, tongue hypertrophy, and abnormal epiglottic movement, adjustment parameter schemes for each component in each of the intelligent airway training models are automatically generated so as to subsequently adjust the intelligent airway training models to more realistically simulate the intubation conditions of the matched cases.

[0099] S620. Adjust the intelligent airway training model according to the component adjustment scheme to simulate the real state corresponding to the matched case; and obtain the state of the components of the intelligent airway training model during the adjustment; the components of the intelligent airway training model include one or more of the following: epiglottic glottis bridging ring, simulated epiglottic spring, maxillary and mandibular bone components, simulated tongue component and cervical spine component; In this embodiment, the components of the intelligent airway training model are adjusted based on the component adjustment scheme to make the intelligent airway training model present a simulated state similar to a matched case. The components of the intelligent airway training model include one or more of the following: an epiglottic-glottic bridging ring, a simulated epiglottic spring, a maxillary and mandibular bone assembly, a simulated tongue assembly, and a cervical spine assembly. Preferably, the adjustable components of the intelligent airway training model are adjusted according to the coordinated adjustment knob.

[0100] In this embodiment, it is also necessary to obtain the status of the components of the intelligent airway training model during the adjustment process, so as to provide a data basis for subsequent acquisition of faulty components.

[0101] S630. Compare the status of the component with the security parameters in the preset security threshold library to obtain the faulty component; In this embodiment, the intelligent airway training model further includes a preset safety threshold library, which records the range of safety parameters of the components in the intelligent airway training model.

[0102] In this embodiment, if the state of a component is detected to exceed the range of safety parameters in the safety threshold library, it indicates that the component is faulty or abnormally regulated, and it is identified as a faulty component.

[0103] S640. Obtain a fault alarm for the intelligent airway training model based on the faulty component.

[0104] In this embodiment, based on the faulty component and its deviation from safety parameters, targeted fault alarm information is generated and displayed on the screen of the intelligent airway training model or broadcast via a voice broadcast component to remind the user to maintain or readjust the intelligent airway training model in a timely manner.

[0105] like Figure 9 As shown in the illustration, this application also provides an intelligent airway training model operation evaluation system. Optionally, the system includes: Data acquisition module 711, operation score acquisition module 712, fusion feature acquisition module 713, optimization suggestion acquisition module 714, evaluation report acquisition module 715, wherein: The data acquisition module 711 is used to collect raw sensor data of the intubation operation according to a preset time interval based on the sensor components in the above-mentioned intelligent airway training model; preprocess the raw sensor data of each time interval to obtain the corresponding sub-operation data; summarize all the sub-operation data to obtain the operation data; and use the video acquisition component to acquire the video data corresponding to the intubation operation. In this embodiment, the data acquisition module 711 can be used to perform... Figure 5 For a detailed description of the data acquisition module 711, please refer to the description of step S100 shown.

[0106] Specifically, the data acquisition module 711 is further configured to acquire the sub-operation data at the current moment and the corresponding first error; predict the prediction data at the next moment based on the sub-operation data at the current moment; evaluate the second error corresponding to the predicted sub-operation data at the next moment based on the first error; and obtain a correction weight based on the second error; acquire the original sensing data at the next moment; correct the original sensing data at the next moment based on the prediction data at the next moment and the correction weight to obtain the sub-operation data at the next moment; and update the second error based on the correction weight.

[0107] In this embodiment, the data acquisition module 711 can also be used to perform... Figure 6 For a detailed description of the data acquisition module 711, see steps S110-S150 shown below. For further details on steps S110-S150, please refer to the description of steps S110-S150.

[0108] The operation score acquisition module 712 is used to evaluate the operation data and the video data based on preset evaluation rules to obtain the operation score of the intubation operation; In this embodiment, the operation scoring acquisition module 712 can be used to perform... Figure 5 For a detailed description of the operation scoring acquisition module 712, please refer to the description of step S200 shown.

[0109] Specifically, the preset evaluation rules in the operation score acquisition module 712 include one or more of operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules, and the operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules are preset with corresponding weights; the operation score acquisition module 712 is also used to acquire evaluation data corresponding to the operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules based on the operation data and the video data; acquire corresponding initial scores according to the operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules and the corresponding evaluation data; and acquire a total score based on the initial scores and the corresponding weights.

[0110] In this embodiment, the operation scoring acquisition module 712 can also be used to perform... Figure 7 For a detailed description of the operation scoring acquisition module 712, see steps S210-S230 shown below. For further details on steps S210-S230, please refer to the description of steps S210-S230.

[0111] The fusion feature acquisition module 713 is used to acquire preset patient features; acquire supplementary features of the intubation operation based on the operation data; and fuse the supplementary features with the patient features to obtain fusion features. In this embodiment, the fusion feature acquisition module 713 can be used to perform... Figure 5 For a detailed description of the fusion feature acquisition module 713, see step S300 shown below. For a detailed description of step S300, please refer to the description of step S300.

[0112] The optimization suggestion acquisition module 714 is used to acquire matching cases that match the fusion features in a preset case database based on a preset matching model, and to acquire optimization suggestion information for the intubation operation based on the matching cases and the supplementary features. In this embodiment, the optimization suggestion acquisition module 714 can be used to perform... Figure 5 For a detailed description of the optimization suggestion acquisition module 714 shown in step S400, please refer to the description of step S400.

[0113] The evaluation report acquisition module 715 is used to generate an evaluation report based on the operation score and the optimization suggestion information.

[0114] In this embodiment, the evaluation report acquisition module 715 can be used to perform... Figure 5 For a detailed description of the evaluation report acquisition module 715, please refer to the description of step S500 shown.

[0115] Specifically, the intelligent airway training model operation evaluation system further includes a fault detection module. The fault detection module is used to obtain a component adjustment scheme based on the matched case; adjust the intelligent airway training model according to the component adjustment scheme to simulate the real state corresponding to the matched case; and obtain the state of the components of the intelligent airway training model during adjustment. The components of the intelligent airway training model include one or more of the following: an epiglottic-glottic bridging ring, a simulated epiglottic spring, a maxillary and mandibular bone assembly, a simulated tongue assembly, and a cervical spine assembly; compare the state of the components with safety parameters in a preset safety threshold library to obtain faulty components; and obtain fault alarms for the intelligent airway training model based on the faulty components.

[0116] In this embodiment, the fault detection module can be used to perform... Figure 8 For a detailed description of the fault detection module, please refer to the description of steps S610-S640 shown.

[0117] This application also provides an electronic device, the structure of which is as follows: Figure 10As shown, the electronic device includes a memory 811, a processor 812, a communication module 813, and an input / output interface 814, etc. Optionally, the memory 811, the processor 812, the communication module 813, and the input / output interface 814 can be connected and communicate with each other through a bus 815.

[0118] The memory 811 is used to store one or more computer programs and to transfer the code of the computer programs to the processor 812; when the one or more computer programs are executed by the processor 812, an intelligent airway training model operation evaluation method is implemented in the embodiments of this application.

[0119] Optionally, the electronic device can be connected to a network via the communication module 813, and can communicate with other devices (such as terminals or servers) via the network to achieve data interaction. The electronic device can be various forms of digital computers, such as desktop computers, servers, workbenches, mainframe computers, or other types of computers. The electronic device can also be various forms of mobile terminals, such as smartphones, tablets, wearable devices (such as helmets, glasses, watches, etc.) and other similar mobile terminals.

[0120] Optionally, the electronic device can connect to required input / output devices, such as a keyboard or display device, via the input / output interface 814. The electronic device itself may have a display device, and other display devices can also be connected externally via the input / output interface 814. Optionally, a storage device, such as a hard disk, can also be connected via the input / output interface 814 to store data from the electronic device, read data from the storage device, or store data from the storage device in the memory 811. It is understood that the input / output interface 814 can be a wired interface or a wireless interface. Depending on the actual application scenario, the device connected to the input / output interface 814 can be a component of the electronic device or an external device connected to the electronic device when needed.

[0121] Optionally, the memory 811 may be a volatile memory and / or a non-volatile memory. The volatile memory may be a random access memory, etc., and the non-volatile memory may be a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, or a flash memory, etc.

[0122] Optionally, the computer program stored in the memory 811 can be divided into one or more modules, which are stored in the memory 811 and executed by the processor 812 to perform the method provided in this embodiment. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the electronic device.

[0123] Optionally, the processor 812 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 812 include, but are not limited to, a central processing unit, a graphics processing unit, a digital signal processor, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, and can also be any suitable controller, microcontroller, processor, etc. The processor 812 executes the various methods and processes of this embodiment, exemplified by, a smart airway training model operation evaluation method according to an embodiment of this application.

[0124] Optionally, the bus 815 may include a path for transmitting information. Depending on its function, the bus 815 may be divided into an address bus, a data bus, a control bus, etc.

[0125] In an optional implementation, this application embodiment also provides a computer storage medium storing a computer program thereon. When executed by a computer, the computer program enables the computer to perform the methods described in the above-described method embodiments. Part or all of the computer program can be loaded and / or installed on the memory 811 of an electronic device. When the computer program is executed by the processor 812, one or more steps of a smart airway training model operation evaluation method according to this application embodiment can be performed.

[0126] Optionally, the computer-readable storage medium may be a random access memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, etc.

[0127] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the technical solution of the present invention, and are not intended to limit the specific implementation of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the claims of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. An intelligent airway training model, comprising: The intelligent airway training model includes a maxillary and mandibular bone assembly, simulated teeth, a simulated tongue assembly, a cervical spine assembly, a simulated larynx, a simulated trachea, a simulated epiglottic spring assembly, a simulated glottic flap assembly, an epiglottic-glottic bridging ring, an independent adjustment knob, a transmission box, a sensor assembly, and a data processing assembly. The simulated epiglottis spring section is connected to the simulated glottis bending section via the epiglottis glottis bridging ring; The independent adjustment knobs include an epiglottis-glottis angle adjustment knob, an epiglottis lifting force adjustment knob, a mouth opening adjustment knob, and a cervical flexion-extension angle adjustment knob; the epiglottis-glottis angle adjustment knob is mechanically connected to the epiglottis-glottis bridging ring through the transmission box; the epiglottis lifting force adjustment knob is mechanically connected to the simulated epiglottis spring unit through the transmission box; the mouth opening adjustment knob is mechanically connected to the maxilla and mandible assembly and the simulated tongue assembly through the transmission box; and the cervical flexion-extension angle adjustment knob is mechanically connected to the cervical spine assembly through the transmission box. The sensor assembly includes a plurality of sensors, which are disposed in one or more of the simulated teeth, the simulated tongue assembly, the cervical spine assembly, the simulated larynx, the simulated trachea, the simulated epiglottis spring, and the simulated glottis flap. The data processing component is electrically connected to the sensor component.

2. The intelligent airway training model of claim 1, wherein, The intelligent airway training model also includes a coordination adjustment knob; the coordination adjustment knob is mechanically connected to the epiglottis glottis bridging ring, the simulated epiglottis spring, the maxillary and mandibular components, the simulated tongue component, and the cervical spine component through the transmission box.

3. The intelligent airway training model of any one of claims 1 or 2, wherein, The coordinated adjustment knob is set with several positions, each position having a corresponding epiglottis-glottis angle, epiglottis lifting force, mouth opening degree, and cervical flexion-extension angle. This allows the coordinated adjustment knob to adjust the epiglottis-glottis bridging ring, the simulated epiglottis spring, the maxillary and mandibular components, the simulated tongue component, and the cervical spine component based on the corresponding epiglottis-glottis angle, epiglottis lifting force, mouth opening degree, and cervical flexion-extension angle of the position.

4. A method for evaluating the operation of an intelligent airway training model, characterized in that, The method includes: The sensor component in the intelligent airway training model according to any one of claims 1 to 3 collects raw sensing data of the intubation operation at a preset time interval; the raw sensing data at each time interval is preprocessed to obtain corresponding sub-operation data, and all the sub-operation data are summarized to obtain operation data. The video acquisition component is used to acquire the video data corresponding to the intubation operation; The operation data and the video data are evaluated based on preset evaluation rules to obtain an operation score for the intubation operation; Obtain preset patient characteristics; obtain supplementary characteristics of the intubation operation based on the operation data, and fuse the supplementary characteristics with the patient characteristics to obtain fused characteristics; Based on a preset matching model, matching cases that match the fusion features are obtained from a preset case database. Based on the matching cases and the supplementary features, optimization suggestions for the intubation operation are obtained. An evaluation report is generated based on the operation score and the optimization suggestion information.

5. The method for evaluating the operation of an intelligent airway training model according to claim 4, characterized in that, The preprocessing of the raw sensor data for each time interval to obtain the corresponding sub-operation data includes: Obtain the current operation data and the corresponding first error; Based on the sub-operational data at the current moment, predict the predicted data for the next moment; The second error corresponding to the predicted score data at the next moment is evaluated based on the first error; and the correction weight is obtained based on the second error. Obtain the raw sensor data for the next moment, and correct the raw sensor data for the next moment based on the predicted data for the next moment and the correction weight to obtain the sub-operation data for the next moment. The second error is updated based on the corrected weights.

6. The method for evaluating the operation of an intelligent airway training model according to claim 4, characterized in that, The preset evaluation rules include one or more of the following: operation standard evaluation rules, force control accuracy evaluation rules, operation integrity evaluation rules, and safety risk evaluation rules. The operation standard evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, and the safety risk evaluation rules are preset with corresponding weights. The evaluation of the operation data and the video data based on preset evaluation rules to obtain the operation score of the intubation operation includes: Based on the operation data and the video data, obtain the evaluation data corresponding to the operation specification evaluation rule, the force control accuracy evaluation rule, the operation integrity evaluation rule, and the safety risk evaluation rule; The initial score is obtained based on the operation specification evaluation rules, the force control accuracy evaluation rules, the operation integrity evaluation rules, and the safety risk evaluation rules, along with the corresponding evaluation data. The total score is obtained based on the initial score and the corresponding weight.

7. The method for evaluating the operation of an intelligent airway training model according to claim 4, characterized in that, The step of obtaining matching cases that match the fusion features from a preset case database based on a preset matching model includes: The fusion features are normalized to obtain the preprocessed fusion features; The preprocessed fusion features are input into the matching model to obtain the matching degree between each case in the case database and the preprocessed fusion features, and cases with a matching degree greater than a preset threshold are selected as matching cases. And / or, the fusion features include several sub-features; the step of obtaining optimization suggestion information for the intubation operation based on the matched case and the supplementary features includes: Based on the matching model, obtain the preset sub-feature weights corresponding to each sub-feature in the fusion features; Optimization suggestions for the intubation procedure are obtained based on the sub-feature weights, the matched cases, and the supplementary features.

8. The method for evaluating the operation of an intelligent airway training model according to claim 4, characterized in that, The method further includes: Adjustment scheme based on the matched case acquisition component; The intelligent airway training model is adjusted according to the component adjustment scheme to simulate the real state corresponding to the matched case; and the state of the components of the intelligent airway training model is obtained during the adjustment; the components of the intelligent airway training model include one or more of the epiglottic glottis bridging ring, the simulated epiglottic spring, the maxilla and mandible components, the simulated tongue component, and the cervical spine component; The status of the component is compared with the security parameters in the preset security threshold library to identify the faulty component; Based on the faulty component, obtain the fault alarm of the intelligent airway training model.

9. An intelligent airway training model operation evaluation system, characterized in that, The system includes: The data acquisition module is used to acquire raw sensor data of the intubation operation according to a preset time interval based on the sensor component in the intelligent airway training model of any one of claims 1 to 3; preprocess the raw sensor data of each time interval to obtain corresponding sub-operation data; summarize all the sub-operation data to obtain operation data; and acquire video data corresponding to the intubation operation using the video acquisition component. The operation score acquisition module is used to evaluate the operation data and the video data based on preset evaluation rules to obtain the operation score of the intubation operation; The fusion feature acquisition module is used to acquire preset patient features; acquire supplementary features of the intubation operation based on the operation data; and fuse the supplementary features with the patient features to obtain fusion features. The optimization suggestion acquisition module is used to acquire matching cases that match the fusion features in a preset case database based on a preset matching model, and to acquire optimization suggestion information for the intubation operation based on the matching cases and the supplementary features. The evaluation report acquisition module is used to generate an evaluation report based on the operation score and the optimization suggestion information.

10. An electronic device, characterized in that, include: Memory, used to store one or more computer programs; A processor, when the one or more computer programs are executed by the processor, implements the intelligent airway training model operation evaluation method as described in any one of claims 4 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute and implement the intelligent airway training model operation evaluation method as described in any one of claims 4 to 8.

Images