Deep Learning-Based Pathogen Recognition and Classification Method and Processing Device

Abstract

The present application discloses a pathogen recognition and classification method and a processing device based on deep learning. The processing device includes an air disinfection processing device and related subsystems. The pathogen recognition and classification method includes: receiving multi-modal sensor data regarding a target space; wherein the multi-modal sensor data includes at least one of environmental data, chemical data, biological data, and optical data; analyzing the multi-modal sensor data to obtain the pathogen type and concentration; and adjusting the working parameters of the air disinfection processing device and related subsystems according to the pathogen type and concentration to perform targeted disinfection on the target space. Through the above method, rapid recognition and response to different pathogens and chemical pollutants are achieved, and the energy usage efficiency of the system is improved.

Description

Technical Field

[0001] This application relates to the field of deep learning recognition and classification, particularly to a method and processing device for pathogen recognition and classification. Background Art

[0002] Traditional air treatment systems have a lagging response, cannot perform precise processing according to different pathogen characteristics, cannot detect biological pathogens and harmful chemical pollutants simultaneously. Existing systems lack intelligent adjustment capabilities, have low energy utilization efficiency, and the air safety monitoring and control systems are fragmented, making it difficult to form a closed loop. Summary of the Invention

[0003] The pathogen recognition and classification method and processing device based on deep learning provided by this application can achieve rapid recognition and response to different pathogens and chemical pollutants, and improve the energy usage efficiency of the system.

[0004] In a first aspect, this application provides a pathogen recognition and classification method based on deep learning. The pathogen recognition and classification method includes: receiving multi-modal sensor data regarding a target space; wherein the multi-modal sensor data includes at least one of environmental data, chemical data, biological data, and optical data; analyzing the multi-modal sensor data to obtain pathogen types and concentrations; and adjusting the operating parameters of an air disinfection treatment device and related subsystems according to the pathogen types and concentrations to perform targeted disinfection on the target space.

[0005] Among them, analyzing the multi-modal sensor data further includes: extracting environmental features, pathogen features, and chemical pollutant features from the multi-modal sensor data; detecting pathogens based on the pathogen features, and detecting chemical pollutants based on the chemical pollutant features; monitoring the environmental status based on the environmental features, monitoring the pathogen status based on the pathogen detection results, and monitoring the chemical pollutant status based on the chemical pollutant detection results; and localizing the pathogen source based on the environmental data and the pathogen detection results, and localizing the chemical pollutant source based on the environmental data and the chemical pollutant detection results; performing an overall risk assessment based on the environmental status monitoring information, pathogen status monitoring information, chemical pollutant status monitoring information, pathogen source localization information, and chemical pollutant source localization information; and formulating a disinfection strategy based on the overall risk assessment results.

[0006] Among them, analyzing the multi-modal sensor data includes: analyzing the multi-modal sensor data using a deep learning model; wherein the deep learning model is trained using distributed federated learning.

[0007] Among them, the deep learning model is trained using distributed federated learning, and the process is as follows: Use reinforcement learning to dynamically adjust the communication frequency with edge computing, optimize the edge computing cluster, and adjust the segmentation strategy of the deep learning model; perform edge forward computing for the adjusted segmentation strategy to generate intermediate representations, and upload the intermediate representations to the regional nodes; perform regional aggregation computing based on the optimized edge computing cluster and intermediate representations, fuse multi-node data, and upload the fusion result to the server; perform central joint modeling training in the server; calculate the global gradient according to the adjusted communication frequency and send it to the regional nodes; update the regional gradient in combination with the local data of the nodes; send the gradient to the edge devices to enable the edge devices to update the local model.

[0008] Among them, the multi-modal sensors are distributedly deployed in the target space to form a distributed sensing network.

[0009] Among them, after receiving the multi-modal sensor data about the target space, it includes: performing data preprocessing on the multi-modal sensor data and selecting a data transmission protocol; distributing tasks based on the preprocessed multi-modal sensor data and selecting corresponding edge computing nodes to process the tasks.

[0010] Among them, a digital twin model is constructed for the air disinfection treatment device and related systems, and the digital twin model is used for simulation and prediction.

[0011] In a second aspect, the present application provides a processing device, which includes an air disinfection treatment device and related subsystems, and is used to implement the deep learning-based pathogen recognition and classification method provided in the first aspect.

[0012] Among them, the related subsystems include at least one of an active air purification system, an ultrasonic air purification system, an intelligent fan control system, and a variable rate air supply system.

[0013] Among them, multiple sensors corresponding to the multi-modal sensor data are integrated on a microfluidic chip.

[0014] The beneficial effects of this application are as follows: Different from the prior art, the pathogen recognition and classification method and processing device based on deep learning provided by this application. The processing device includes an air disinfection processing device and related subsystems. The pathogen recognition and classification method includes: receiving multi-modal sensor data regarding a target space; wherein, the multi-modal sensor data includes at least one of environmental data, chemical data, biological data, and optical data; analyzing the multi-modal sensor data to obtain the pathogen type and concentration; adjusting the operating parameters of the air disinfection processing device and related subsystems according to the pathogen type and concentration, and performing targeted disinfection on the target space, capable of realizing the rapid recognition and response to different pathogens and chemical pollutants through the established intelligent air safety prevention and control system, and improving the energy usage efficiency of the system. Brief Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions in the embodiments of this application, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0016] Figure 1 It is a schematic flowchart of an embodiment of the pathogen recognition and classification method and processing device based on deep learning provided by this application;

[0017] Figure 2 It is a schematic flowchart of another embodiment of the pathogen recognition and classification method and processing device based on deep learning provided by this application;

[0018] Figure 3 It is a schematic flowchart of an embodiment of the training of the deep learning model provided by this application;

[0019] Figure 4 It is a schematic structural diagram of an embodiment of the processing device provided by this application. Detailed Embodiments

[0020] The following will clearly and completely describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. It can be understood that the specific embodiments described herein are only used to explain this application, rather than limiting this application. Additionally, it should be noted that for the sake of description, only the parts related to this application are shown in the drawings rather than all the structures. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

[0021] References herein to "embodiments" mean that the particular features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present application. The phrase occurring in various places in the specification is not necessarily referring to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0022] Refer to Figure 1 , Figure 1 FIG. is a schematic flowchart of an embodiment of a pathogen recognition and classification method and a processing device based on deep learning provided by the present application. The device includes an air disinfection processing device and related subsystems. The pathogen recognition and classification method includes:

[0023] Step 11: Receive multimodal sensor data regarding a target space; wherein, the multimodal sensor data includes at least one of environmental data, chemical data, biological data, and optical data.

[0024] In some embodiments, the multimodal sensor data includes at least one of environmental data, chemical data, biological data, and optical data.

[0025] In some embodiments, the multimodal sensor data includes at least two of environmental data, chemical data, biological data, and optical data.

[0026] In some embodiments, the multimodal sensor data includes at least three of environmental data, chemical data, biological data, and optical data.

[0027] In some embodiments, the multimodal sensor data includes environmental data, chemical data, biological data, and optical data.

[0028] Step 12: Analyze the multimodal sensor data to obtain the pathogen type and concentration.

[0029] In some embodiments, a deep learning model can be used to analyze the multimodal sensor data to obtain the pathogen type and concentration.

[0030] In some embodiments, corresponding rules can be used to analyze the multimodal sensor data to obtain the pathogen type and concentration. For example, use an expert system rule base, use a conditional logic tree, and / or a predefined threshold rule to analyze the multimodal sensor data to obtain the pathogen type and concentration.

[0031] Step 13: Adjust the working parameters of the air disinfection processing device and related subsystems according to the pathogen type and concentration, and perform targeted disinfection on the target space.

[0032] In some embodiments, the type, concentration, and treatment time of the disinfectant are dynamically adjusted according to the detection results and environmental conditions to achieve personalized disinfection. The ventilation system is automatically adjusted according to parameters such as indoor air quality and pathogen concentration to optimize the indoor air environment. The air volume of the fan is automatically adjusted according to indoor air quality and pathogen concentration to achieve efficient disinfection. Differential control of different spaces is achieved by adjusting the air supply speed.

[0033] Differential control of different spaces can be achieved by adjusting the air supply speed of the variable-speed air supply system.

[0034] The active air purification system can continuously purify the air by actively generating negative ions, photocatalysis, etc.

[0035] In other embodiments, the indicators monitored by the system may include at least one of: TEMP (temperature), HUMI (humidity), NH3 (ammonia), CO (carbon monoxide), CH2O (formaldehyde), O2 (oxygen), CO2 (carbon dioxide), O3 (ozone), SO2 (sulfur dioxide), C6H6 (benzene), C7H8 (toluene), C8H 10 (xylene), RN (radon), TVOC (total volatile organic compounds), PM0.3 (inhalable particulate matter), PM0.5 (fine particulate matter), PM1.0 (fine particulate matter), PM2.5 (fine particulate matter), PM10 (inhalable particulate matter), Beta-ray (β-ray), Gamma-ray (γ-ray), and X-ray (X-ray).

[0036] In this embodiment, sensor data collected by a variety of sensors such as spectroscopy, electrochemistry, mass, biology, and gas sensing can be combined to construct a multi-modal fusion model to achieve precise identification of different types of pathogens and chemical pollutants, while realizing the functions of pollution source localization and pollution status monitoring.

[0037] In this embodiment, the processing device uses multiple sensors to continuously monitor parameters such as indoor air quality and pathogen concentration. Through multi-sensor fusion pattern recognition, precise identification of different types of pathogens and chemical pollutants is achieved, while realizing the functions of pollution source localization and pollution status monitoring. And according to the analysis results, the system automatically adjusts the working parameters of the ventilation system and the disinfection device to achieve targeted disinfection of different types of pathogens. Through the established intelligent air safety prevention and control system, rapid identification and response to different pathogens and chemical pollutants can be achieved, improving the energy use efficiency of the system.

[0038] Refer to Figure 2 , Figure 2 which is a schematic flowchart of another embodiment of the pathogen recognition and classification method and processing device based on deep learning provided by this application.

[0039] Step 21: Receive multimodal sensor data regarding the target space; wherein, the multimodal sensor data includes at least one of environmental data, chemical data, biological data, and optical data.

[0040] In some embodiments, the received multimodal sensor data can be subjected to multimodal fusion to obtain multimodal fusion data.

[0041] Step 22: Extract environmental features, pathogen features, and chemical pollutant features from the multimodal sensor data.

[0042] In some embodiments, environmental features, pathogen features, and chemical pollutant features can be extracted from the multimodal fusion data.

[0043] Step 23: Perform pathogen detection based on the pathogen features and perform chemical pollutant detection based on the chemical pollutant features.

[0044] In some embodiments, biological data can be collected by a biosensor. A biosensor is a device that combines a biorecognition element and sensor technology, and converts target molecules (such as pathogenic microorganisms) into measurable signals through a specific biorecognition process. Its working principle can be divided into immunosensing, cell sensing, and enzyme sensing.

[0045] In some embodiments, chemical data can be collected by a chemical sensor. Its principle is to pre-select appropriate chemical substances so that when pollutants come into contact with the chemical materials, reactions occur, thereby changing the electrical properties of the sensor, and then realizing the detection of pollutants. For example, monitoring the concentrations of pollutants such as PM2.5, PM10, and VOCs in the air.

[0046] Step 24: Monitor the environmental status based on the environmental features, monitor the pathogen status based on the pathogen detection results, and monitor the chemical pollutant status based on the chemical pollutant detection results.

[0047] In some embodiments, for environmental parameter analysis, environmental status monitoring is achieved to obtain environmental status information.

[0048] In some embodiments, for environmental data and pathogen detection results, pathogen dynamic monitoring based on time series analysis is adopted to obtain pathogen status information.

[0049] In some embodiments, for environmental data and chemical pollutant detection results, chemical pollutant dynamic monitoring based on time series analysis is adopted to obtain chemical pollutant status information.

[0050] Step 25: Locate the pathogen source based on the environmental data and pathogen detection results, and locate the chemical pollutant source based on the environmental data and chemical pollutant detection results.

[0051] In some embodiments, for the environmental data and pathogen detection results, a pathogen source location method based on triangulation and diffusion model is adopted to obtain the pathogen release position and achieve pathogen source location.

[0052] In some embodiments, for the environmental data and chemical pollutant detection results, a chemical pollutant source location method based on diffusion model is adopted to obtain the chemical pollutant release position and achieve chemical pollutant source location.

[0053] Step 26: Conduct an overall risk assessment based on the environmental status monitoring information, pathogen status monitoring information, chemical pollutant status monitoring information, pathogen source location information, and chemical pollutant source location information.

[0054] Step 27: Develop a disinfection strategy based on the overall risk assessment results.

[0055] In some embodiments, a deep learning model can be used to develop a disinfection strategy based on the overall risk assessment results. Further, during the disinfection process, the latest multi-modal sensor data can be collected in real time, and the disinfection strategy can be updated according to the latest multi-modal sensor data according to the above control logic.

[0056] In some embodiments, a deep learning model can be used to analyze the multi-modal sensor data; among them, the deep learning model is trained using distributed federated learning.

[0057] In some embodiments, the features for training can be obtained from the multi-modal sensor data. For example, the received multi-modal sensor data can be subjected to multi-modal fusion to obtain multi-modal fusion data. The multi-modal fusion data is subjected to high-dimensional feature representation, and then feature dimension reduction, feature enhancement, and semantic association modeling are performed to obtain a high-dimensional feature representation. And it is classified and labeled, and the classified and labeled features are used as training features. Among them, classification can be based on rules and machine learning.

[0058] Refer to Figure 3 , the process of training the deep learning model using distributed federated learning is as follows:

[0059] Step 31: Dynamically adjust the communication frequency with edge computing, optimize the edge computing cluster, and adjust the segmentation strategy of the deep learning model using reinforcement learning.

[0060] Step 32: Perform edge forward calculation for the adjusted segmentation strategy to generate an intermediate representation, and upload the intermediate representation to the regional nodes.

[0061] Step 33: Perform regional aggregation calculation based on the optimized edge computing cluster and intermediate representation, fuse multi-node data, and upload the fusion result to the server.

[0062] Step 34: Conduct central joint modeling training in the server.

[0063] Step 35: Calculate the global gradient according to the adjusted communication frequency and distribute it to the regional nodes.

[0064] Step 36: Update the regional gradient by combining the local data of the nodes.

[0065] Step 37: Distribute the gradient to the edge devices to enable the edge devices to update the local model.

[0066] After completing one round of training, determine whether the inference and fusion requirements of the edge model (local model) are met. If so, perform multi-modal model fusion to improve the generalization ability.

[0067] In some implementations, multi-modal sensors are distributedly deployed in the target space to form a distributed sensing network, which can achieve real-time monitoring and dynamic response to the environment within the range of a single device and between multiple devices.

[0068] After forming the distributed sensing network, after receiving the multi-modal sensor data about the target space, perform data preprocessing on the multi-modal sensor data and select a data transmission protocol; distribute tasks based on the preprocessed multi-modal sensor data and select corresponding edge computing nodes to process the tasks.

[0069] For example, for the distributed sensing network, in terms of data acquisition, it involves multi-modal sensor access, data acquisition frequency and synchronization settings, modal data alignment and timestamp synchronization, and raw data quality assessment and calibration, and then data preprocessing and compression are performed at the edge node, as well as the selection of data transmission protocol and QOS calculation.

[0070] Among them, edge data preprocessing mainly performs operations such as noise filtering and smoothing on the data, then performs data standardization / normalization operations, and further performs data fusion, such as early fusion / late fusion. And store the preprocessed data at the edge, such as storing according to the corresponding cache strategy, and perform local data backup to generate corresponding logs.

[0071] After data preprocessing, cluster management and monitoring are carried out. Then, task distribution and load balancing processing are performed based on the preprocessed data, and the corresponding tasks are assigned to the corresponding edge computing nodes, such as Edge Computing Node 1, Edge Computing Node 2, and Edge Computing Node 3. In addition, the edge model can be updated and iterated using the preprocessed data. Moreover, the preprocessed data can be used for environmental perception by the edge model to obtain corresponding intelligent decisions. In addition, the environmental perception system also needs to set corresponding API interfaces, such as between the edge and the cloud server, and between the edge and the device, and security authentication needs to be performed on the API interfaces.

[0072] Furthermore, the cloud server can perform data aggregation and compression on the corresponding intelligent decisions and preprocessed data, perform cloud data storage / analysis, and evaluate the corresponding analysis results using the first evaluation network. Further, the edge model can be updated and iterated using the analysis results.

[0073] Among them, the relevant subsystems include at least one of an active air purification system, an ultrasonic air purification system, an intelligent fan control system, and a variable rate air supply system.

[0074] Among them, a digital twin model is constructed for the air disinfection treatment device and related systems, and the digital twin model is used for simulation and prediction.

[0075] Refer to Figure 4 , Figure 4 FIG. is a schematic structural diagram of an embodiment of a processing device provided in the present application. The processing device 100 includes an air disinfection treatment device 10 and related subsystems 20, and is used to implement the control of the pathogen recognition and classification method based on deep learning as described in any of the above embodiments. The related subsystems 20 include at least one of an active air purification system, an ultrasonic air purification system, an intelligent fan control system, and a variable rate air supply system.

[0076] In some embodiments, the related subsystems 20 include at least one of an active air purification system, an ultrasonic air purification system, an intelligent fan control system, and a variable rate air supply system.

[0077] Among them, multiple sensors corresponding to multi-modal sensor data are integrated on a microfluidic chip. Integrating multiple sensors onto a microfluidic chip can achieve miniaturized, portable, and high-throughput detection.

[0078] In an application scenario, the air disinfection treatment device 10 can be modularly designed. For example, scene customization, modular design for scenarios such as hospitals, offices, and homes. And standardize the corresponding general interfaces to achieve quick plugging and unplugging.

[0079] The processing device 100 has a corresponding environmental perception system, which can perform multi-physical field simulations. For example, thermodynamic simulations can optimize the evaporation and diffusion of disinfectants. For example, CFD optimization can analyze air flow and pollutant distribution, as well as experimental verification and simulation feedback to dynamically adjust design parameters.

[0080] The environmental perception system can make intelligent decisions based on sensor data, evaluate the intelligent decisions, and optimize according to the evaluation. For example, optimize the process of data fusion and analysis, such as optimizing the process of edge computing and cloud computing. And optimize the corresponding disinfection strategy according to the evaluation. Then, the strategy is sent down according to the disinfection strategy. For example, sent to the active air purification system, ultrasonic air purification system, intelligent fan control system, variable rate air supply system, so that the active air purification system, ultrasonic air purification system, intelligent fan control system, variable rate air supply system perform corresponding operations. Further, a second evaluation can be carried out in combination with the disinfection strategy, and the parameters in the disinfection strategy are adjusted using the corresponding disinfection strategy loss function.

[0081] Furthermore, the processing device 100 also has an intelligent integration platform, which can interact with the active air purification system, ultrasonic air purification system, intelligent fan control system, variable rate air supply system, sensor module, and air disinfection processing device to achieve remote visual control.

[0082] Furthermore, the processing device 100 can be divided into an intelligent sensing fusion system layer, an intelligent control layer, and an execution system layer.

[0083] The intelligent sensing fusion system layer can be divided into a sampling unit and a detection unit.

[0084] The sampling unit can include multi-channel sampling, preconcentration device, and automatic sampler.

[0085] The detection unit can include a spectral detection module, an electrochemical sensing module, and a mass sensor.

[0086] The intelligent control layer can involve signal processing and AI analysis.

[0087] Signal processing can include multi-source data fusion feature extraction and pattern recognition.

[0088] AI analysis can use deep learning models for analysis and cross-validation for real-time warning.

[0089] The execution system layer can include a modular disinfection processing unit, an intelligent ventilation control system, a variable frequency ultrasonic collaborative air treatment system, etc.

[0090] Specifically, the intelligent ventilation control system based on environmental perception can automatically adjust the ventilation system according to parameters such as indoor air quality and pathogen concentration to optimize the indoor air environment.

[0091] The variable frequency ultrasonic wave collaborative air purification system can improve the air purification efficiency and reduce energy consumption.

[0092] The modular design can facilitate customization according to different scenarios and requirements.

[0093] The variable frequency ultrasonic wave system can generate cavitation effects through ultrasonic waves to destroy the cell structures of pathogens.

[0094] The intelligent fan control system can automatically adjust the fan air volume according to the indoor air quality and pathogen concentration to achieve efficient disinfection.

[0095] The variable speed air supply system can achieve differential control of different spaces by adjusting the air supply speed.

[0096] The adaptive disinfection strategy can dynamically adjust the type, concentration and treatment time of disinfectants according to the detection results and environmental conditions to achieve personalized disinfection.

[0097] The collaborative multi-physical field simulation can combine CFD simulation, thermodynamic simulation, etc. to optimize the air flow and disinfectant distribution and improve the disinfection efficiency.

[0098] Active air purification can continuously purify the air by actively generating negative ions, photocatalysis, etc.

[0099] In summary, the processing device 100 monitors parameters such as indoor air quality and pathogen concentration in real time through multiple sensors. While accurately identifying different types of pathogens and chemical pollutants through the multi-sensor fusion pattern recognition, it also realizes the functions of pollution source location and pollution status monitoring.

[0100] The processing device 100 analyzes the sensor data using a deep learning model to accurately identify the pathogen type and concentration.

[0101] The processing device 100 automatically adjusts the working parameters of the ventilation system and disinfection device according to the analysis results to achieve targeted disinfection of different types of pathogens. On the premise of protecting data privacy, the federated learning technology is used to jointly model the data from different places to improve the generalization ability of the model.

[0102] The processing device 100 continuously adjusts the control strategy according to the disinfection effect to achieve dynamic optimization of the indoor air quality.

[0103] The processing device 100 deploys sensor nodes in different areas to form a distributed sensing network to achieve zonal monitoring and dynamic response to the environment within the range of a single device and between multiple devices.

[0104] In several embodiments provided by the present application, it should be understood that the disclosed deep learning-based pathogen recognition and classification method and processing device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0105] If the integrated unit in the above-mentioned other embodiments is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the deep learning-based pathogen recognition and classification method described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0106] The above is only the embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A pathogen identification and classification method based on deep learning, characterized in that: The pathogen identification and classification method comprises: Receiving multimodal sensor data about a target space; wherein the multimodal sensor data includes environmental data, chemical data, and biological data; analyzing the multimodal sensor data to obtain pathogen type and concentration; Adjusting the operating parameters of the air disinfection treatment device and related subsystems according to the type and concentration of the pathogens to perform targeted disinfection of the target space; The analyzing the multimodal sensor data further includes: extracting environmental signatures, pathogen signatures, and chemical pollutant signatures from the multimodal sensor data; Performing pathogen detection based on the pathogen characteristics, and performing chemical pollutant detection based on the chemical pollutant characteristics; Conducting environmental status monitoring based on the environmental characteristics, pathogen status monitoring based on pathogen detection results, and chemical pollutant status monitoring based on chemical pollutant detection results; and locating the source of pathogens based on the environmental data and the pathogen detection results, and locating the source of chemical pollutants based on the environmental data and the chemical pollutant detection results; Conduct an overall risk assessment based on environmental status monitoring information, pathogen status monitoring information, chemical pollutant status monitoring information, pathogen source location information, and chemical pollutant source location information; Develop disinfection strategies based on overall risk assessment results; Analyzing the multimodal sensor data using a deep learning model; Wherein, the deep learning model is trained using distributed federated learning; The deep learning model is trained using distributed federated learning as follows: Use reinforcement learning to dynamically adjust the communication frequency with edge computing, optimize edge computing clusters, and adjust the segmentation strategy of deep learning models; Perform edge forward calculations on the adjusted segmentation strategy, generate intermediate representations, and upload the intermediate representations to the regional nodes; Perform regional aggregation calculations based on the optimized edge computing cluster and the intermediate representation, fuse multi-node data, and upload the fusion results to the server; Conduct central joint modeling training in the server; Based on the adjusted communication frequency, the global gradient is calculated and sent to the regional nodes. After completing a round of training, it is determined whether the edge model inference and fusion requirements are met. If so, multimodal model fusion is performed. Combine the local data of the node to update the regional gradient; Send gradients to edge devices so that they can update their local models. Multimodal sensors are distributedly deployed in the target space to form a distributed sensing network; After receiving the multimodal sensor data about the target space, the method further includes: Performing data preprocessing on the multimodal sensor data and selecting a data transmission protocol; Tasks are distributed based on the pre-processed multimodal sensor data, and the corresponding edge computing nodes are selected to process the tasks.

2. The pathogen identification and classification method according to claim 1, characterized in that: The related subsystems include: at least one of an active air purification system, an ultrasonic air purification system, an intelligent fan control system, and a variable rate air supply system.

3. The pathogen identification and classification method according to claim 2, characterized in that: A digital twin model is constructed for the air disinfection treatment device and related systems, and the digital twin model is used for simulation and prediction.

4. A processing device, characterized in that: The processing device includes an air disinfection processing device and related subsystems, which are used to implement the pathogen identification and classification method according to any one of claims 1 to 3.

5. The processing device according to claim 4, characterized in that The multiple sensors corresponding to the multimodal sensor data are integrated on a microfluidic chip.

Images