Medical voice purification panel lamp integrating automatic dimming and patient monitoring and recognition

Abstract

The invention belongs to the technical field of medical lighting equipment, and discloses a medical voice purification panel lamp integrating automatic dimming and patient monitoring and recognition, which comprises an illumination detection module, a panel lamp control module, a visual detection module and a voice interaction module. The illumination detection module adopts a photosensitive sensor to monitor the environment illumination intensity in real time, the visual detection module collects a patient image through a camera and analyzes the posture and state of a patient in combination with a convolutional neural network algorithm, and the panel light control module automatically adjusts the brightness, the color temperature and the light angle according to a processing result. And the voice interaction module realizes real-time voice interaction between the medical staff and the patient through voice recognition and voice synthesis technologies. The technical problems that a traditional medical panel lamp is single in function and simple in control mode are solved, intelligence and humanization of the medical panel lamp are achieved, the comfort level and safety of the medical environment are effectively improved, and the illumination requirements of different medical scenes are met.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical lighting equipment, and particularly relates to a medical voice-purifying flat lamp integrating automatic dimming and patient monitoring and identification. Background Art

[0002] Intelligent medical flat lamps are a new concept of flat lamp control and management formed by integrating high-tech such as sensor technology, automatic control technology, and image processing technology. They are an inevitable direction for the development of intelligent flat lamps in the digital information age. Through sensors distributed on the flat lamp, they sense the brightness of the hospital environment, the body postures and states of patients, etc. The collected information is calculated, analyzed, and decision-making by the MCU, and relevant instructions are sent according to the decision-making results to adjust the light intensity, color temperature, etc., and perform voice interaction with patients. Intelligent medical flat lamps focus on integrating information resources of various body postures and states of patients, so as to provide intelligent services for hospitals and patients, making them more intelligent, user-friendly, and efficient in daily use.

[0003] In hospitals, places such as wards and operating rooms are generally illuminated by medical flat lamps, and the irradiation targets of medical flat lamps are generally medical staff and patients. The control method of the current medical flat lamp lighting system is usually to operate by manually switching the gear to control the relevant actuators by medical staff or patients. The method is single, the control is not flexible enough, and the medical flat lamp cannot be adaptively adjusted according to the lighting conditions of the hospital scene environment or the body postures and states of patients. Moreover, the medical flat lamp does not have the ability of data collection and calculation, and it is difficult to meet the diverse functional requirements in reality. And specifically, which light intensity, color temperature, and light angle the medical flat lamp is adjusted to be more comfortable needs to be adjusted in real time according to the body postures and states of patients and the actual situation of the scene at that time by medical staff. This process is very dependent on the experience and operation of medical staff.

[0004] The following problems exist in the prior art:

[0005] 1) The control method of traditional medical flat lamps is single. Usually, only the brightness and color temperature can be adjusted by manually switching, lacking intelligent control means.

[0006] 2) Existing medical flat lamps cannot automatically adjust the brightness and color temperature according to the environmental light change, resulting in the light not adapting to the environmental change.

[0007] 3) Traditional medical flat lamps cannot sense the body postures and states of patients and cannot provide personalized lighting services for different patient states.

[0008] 4) Existing medical flat lamps lack a voice interaction function and cannot be controlled and adjusted by voice commands, which is inconvenient to use.

[0009] 5) Traditional medical flat panel lights do not have the ability to collect and analyze data, making it difficult to continuously optimize the lighting effect.

[0010] Currently, the application of artificial intelligence technology and sensor technology is changing the working mode of traditional devices. As a basic public / private service facility, medical flat panel lights must be intelligent. By adopting the design concept of "light detection + vision detection + flat panel light control + voice interaction", this invention makes the medical flat panel lights more intelligent and scientific. Summary of the Invention

[0011] The technical problem to be solved by this invention is to provide a medical voice purification flat panel light that integrates automatic dimming and patient monitoring and recognition to address the defects in the prior art. This flat panel light can automatically adjust the brightness, color temperature, and light angle according to the ambient light intensity, patient body posture, and state, and has a voice interaction function, realizing the intelligence and humanization of medical flat panel lights.

[0012] To achieve the above object, this invention provides a medical voice purification flat panel light that integrates automatic dimming and patient monitoring and recognition, including a light detection module, a vision detection module, a flat panel light control module, and a voice interaction module. The light detection module is used to detect the ambient light in the medical scenario, the vision detection module is used to detect the patient's body posture and state, the medical flat panel light control module is used to execute the MCU command to control the brightness of the flat panel light, and the voice interaction module is used to execute the MCU command for relevant real-time voice communication.

[0013] Specifically, the technical solution of this invention includes:

[0014] Light detection module: A photosensitive sensor is used to continuously monitor the ambient light intensity, and the collected light data is transmitted to the MCU for processing. The photosensitive sensor adopts a voltage division circuit method, whose resistance decreases with the increase of light intensity and increases with the decrease of light intensity. By monitoring the change of the voltage value in the voltage division circuit, the change of the ambient light intensity in the current environment is reflected.

[0015] Vision detection module: The patient's image is collected by a camera, and after image preprocessing, it is input into a convolutional neural network for patient body posture and state recognition. The convolutional neural network adopts a deep learning algorithm and is trained through a labeled patient body posture and state image data set to achieve accurate recognition of the patient's body posture and state.

[0016] Flat panel light control module: Adjust the brightness, color temperature, and light angle of the flat panel light according to the control instruction of the MCU. This module adopts a circuit isolation and two-way power supply design method. The circuit is separated into two parts through an optoelectronic isolator to achieve electrical isolation of the control circuit and the execution circuit.

[0017] Voice interaction module: It includes a voice collection unit, a voice recognition unit, a voice synthesis unit, a noise purification unit and a speaker, and realizes real-time voice interaction between medical staff and patients. It controls the working state of the flat lamp through voice commands and provides voice reminder and feedback functions.

[0018] MCU control unit: As the core processing unit of the system, it receives the signals from the light detection module and the vision detection module, analyzes and processes them, and then sends control instructions to the flat lamp control module and the voice interaction module to realize the intelligent control of the medical flat lamp.

[0019] The sensor system is an important part of the flat lamp system and also a necessary functional module for constructing the entire medical flat lamp control and management system. Its main function is to connect the sensor system formed by the flat lamp to the MCU to complete information collection and processing. The core of the entire flat lamp control system is the data processing function of the MCU and the control of each functional module.

[0020] The flat lamp sensor system is composed of multiple photosensitive sensors, cameras, etc. Each sensor and functional module in this system is equipped with a node control unit necessary for node control. These functional modules realize the state control of a single module through corresponding driver programs, and at the same time, the MCU processes the information collected by the sensors, and finally realizes the control of the flat lamp illumination and the recognition of the patient's body posture and state.

[0021] In order to realize the functions of the flat lamp control system such as patient information collection and flat lamp state control, a sensor network needs to be constructed. For the flat lamp, corresponding sensor drive circuits, control circuits, etc. need to be designed. The main functions realized in this part include: patient data collection, illumination information collection of medical staff and the patient's surrounding environment, flat lamp brightness control, voice interaction, data analysis and processing, and other expandable functions.

[0022] To achieve the recognition of the patient's body posture and status, a convolutional neural network is used to directly learn features from the video images of the patient's daily behaviors in the hospital. Batch normalization and residual modules are added to the network to accelerate network convergence, and features of different depths are fused. Three size feature maps are used for prediction to improve the network's detection ability for multi-size targets. The convolutional neural network mainly consists of a convolutional layer, a Relu activation function, a pooling layer, a fully connected layer, and a dropout layer. First, the picture dataset containing the patient's body posture and status is manually annotated, and the pictures and label files of the patient's body posture and status are sent into the convolutional neural network for training and learning. The specific learning process is as follows: First, a part of the features of the user's working and learning status is extracted through the convolutional layer, but the convolutional result cannot be directly used as a feature map. Then, through the action of the Relu activation function, the function output result is used as the feature map to endow the network with non-linear fitting ability. Next, feature sampling is performed through the pooling layer to compress the features. The pooling operation not only reduces the size of the feature map but also makes the features extracted by the network have translational invariance. Finally, the probability that the image is predicted as each category is calculated through the activation function in the fully connected layer, and the category corresponding to the maximum probability is taken as the category of this prediction (i.e., the target patient). During the training process, the activation function is used as an index of detection accuracy. First, the value of the loss is calculated forward, and then this error is backpropagated to the dropout layer, and the weight of each neuron is used to take the partial derivative of the error. Through SGD (stochastic gradient descent method), this derivative tends to zero, so the value of the loss will decrease. This constitutes a complete training iteration process. After two cycles, the prediction accuracy of recognizing the patient's body posture and status can reach a relatively high level. After the training is completed, in the test stage, the convolutional neural network can frame the target patient with a rectangular bounding box and obtain the planar coordinates of the target patient, providing data for the subsequent control decision of the medical flat lamp.

[0023] The medical voice purification flat lamp integrating automatic dimming and patient monitoring and recognition of the present invention has the following beneficial effects:

[0024] 1. The present invention realizes the intelligent control of the medical flat lamp, automatically adjusts the brightness, color temperature, and light angle according to the environmental light intensity and the patient's body posture and status, and improves the comfort and safety of the medical environment;

[0025] 2. The present invention realizes the accurate recognition of the patient's body posture and status through a convolutional neural network, provides data support for the intelligent control of the flat lamp, and makes the lighting effect more user-friendly;

[0026] 3. The present invention integrates functions of light detection, visual detection, flat lamp control, and voice interaction, realizes multi-modal interaction, and meets the lighting needs of different medical scenarios;

[0027] 4. The present invention has a voice interaction function, and controls the working state of the flat lamp through voice commands, improving the convenience of use for medical staff and patients;

[0028] 5. The present invention adopts a design method of circuit isolation and dual power supply, ensuring the safety and stability of the system;

[0029] 6. The present invention has a data storage and analysis function, providing data support for the continuous optimization of the flat lamp control strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0031] Figure 1 It is a functional module planning diagram of a medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification of the present invention;

[0032] Figure 2 It is a state control scheme diagram of a medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification of the present invention;

[0033] Figure 3 It is a flowchart of the convolutional neural network training process of a medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0035] Please refer to Figure 1 As shown, the medical flat lamp control program runs on the MCU of the flat lamp control unit. The functional modules formed by combining the peripheral hardware circuits can realize data acquisition and processing and the control of the flat lamp state. The main functions that need to be realized in this part according to requirements include: vision detection module, light detection module, flat lamp control module, voice interaction module.

[0036] Embodiment 1:

[0037] Please refer to Figure 1As shown in the figure, the medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification includes a light detection module, a vision detection module, a flat lamp control module, a voice interaction module, and an MCU control unit.

[0038] The light detection module uses a photosensitive sensor to monitor the ambient light intensity in real time. The system can adjust and control the flat lamp according to the brightness information of the medical staff or the patient's surrounding environment, which is achieved through the light detection module. A photoresistor is used as the sensor for collecting brightness information. According to the basic working principle of the photoresistor, its resistance decreases with the increase of light intensity and increases with the decrease of light intensity. Usually, after the photoresistor is connected to a voltage dividing circuit, the change of the voltage value in the voltage dividing circuit can reflect the change of the light intensity in the current environment. Using this principle, the circuit of the light detection module is designed.

[0039] The vision detection module collects patient images through a camera, and inputs them into a convolutional neural network for patient posture and state recognition after image preprocessing. For the structural design of the convolutional neural network, please refer to Figure 3 As shown in the figure, it includes a convolutional layer, a Relu activation function layer, a pooling layer, a fully connected layer, and a dropout layer. This network is trained on a labeled dataset of patient posture and state images to achieve accurate recognition of patient posture and state.

[0040] The flat lamp control module adjusts the brightness, color temperature, and light angle of the flat lamp according to the control instructions of the MCU. Each functional module of the medical flat lamp sensor system mainly provides decision-making parameters for the state control of the flat lamp (illuminance, color temperature, light angle). The circuit of this module needs to support AC power supply of 100 - 240V. Therefore, the circuit of this module needs to support a large voltage and current range. However, usually, the MCU cannot directly provide such a large current and voltage control. Considering this point, it is an inevitable choice to control a large voltage and current through a small voltage and current. In the design of this circuit, the circuit is divided into two parts by an opto-isolator as a switch. The opto-isolator circuit generally consists of three parts: light emission, reception, and signal amplification. The unidirectional transmission of signals can effectively achieve electrical isolation between the input and output ends. The flat lamp switch circuit is mainly composed of an electromagnetic relay, which uses the electromagnetic principle to achieve the switch control function. It can control the on and off of a 240V AC circuit through a 5V voltage, thereby realizing the control of the flat lamp illumination.

[0041] The voice interaction module includes a voice collection unit, a voice recognition unit, a voice synthesis unit, a noise purification unit, and a speaker, realizing real-time voice interaction between medical staff and patients. The working state of the flat lamp is controlled through voice commands, providing voice reminder and feedback functions. This module uses digital signal processing technology to achieve the collection, recognition, synthesis, and playback of voice signals, improving the interactivity and convenience of the system.

[0042] The MCU control unit, as the core processing unit of the system, receives the signals from the light detection module and the vision detection module. After analysis and processing, it sends control instructions to the flat lamp control module and the voice interaction module to achieve intelligent control of the medical flat lamp. The control strategies of the MCU include an automatic dimming strategy based on the ambient light intensity, a flat lamp control strategy based on the patient's body posture and status, a control strategy based on the voice commands of medical staff, and a voice reminder strategy based on the change of the patient's status.

[0043] Embodiment 2:

[0044] Please refer to Figure 2 As shown, the flat lamp control solution of the present invention includes the following steps:

[0045] Step S1: Collect ambient light intensity data through the light detection module;

[0046] Step S2: Collect the patient's image through the vision detection module and perform patient body posture and status recognition;

[0047] Step S3: The MCU analyzes and processes the collected data to generate control instructions;

[0048] Step S4: The flat lamp control module adjusts the brightness, color temperature, and light angle of the flat lamp according to the control instructions;

[0049] Step S5: The voice interaction module performs voice reminders and feedback according to the control instructions.

[0050] In this embodiment, the control strategies of the MCU specifically include:

[0051] Automatic dimming strategy based on ambient light intensity: When the ambient light intensity is lower than the preset threshold, automatically increase the brightness of the flat lamp; when the ambient light intensity is higher than the preset threshold, automatically decrease the brightness of the flat lamp; set different light intensity thresholds according to different time periods and different medical scenarios.

[0052] Flat lamp control strategy based on patient body posture and status: When it is recognized that the patient is in a sleeping state, automatically decrease the brightness of the flat lamp and adjust it to a warm color temperature; when it is recognized that the patient is in a reading state, automatically adjust the brightness and light angle of the flat lamp to provide appropriate reading light; when it is recognized that the patient is in a treatment state, automatically adjust the brightness and color temperature of the flat lamp to meet medical needs.

[0053] Control strategy based on the voice commands of medical staff: Receive the voice commands of medical staff, such as "increase brightness", "decrease color temperature", etc.; parse the voice commands to generate corresponding control commands; execute the control commands to adjust the working state of the flat lamp.

[0054] Voice reminder strategy based on patient status changes: Monitor changes in the patient's status, such as transitioning from a sleeping state to a waking state; generate corresponding voice reminders based on the status changes; play the voice reminders through a speaker, such as "Good morning. The lighting has been adjusted to be suitable for you."

[0055] Embodiment 3:

[0056] Please refer to Figure 3 As shown, the convolutional neural network training process of the present invention includes the following steps:

[0057] Step S1: Manually annotate the picture dataset containing the patient's body posture and status;

[0058] Step S2: Send the pictures and label files of the patient's body posture and status into the convolutional neural network for training;

[0059] Step S3: Extract the features of the patient's body posture and status through the convolutional layer;

[0060] Step S4: Through the action of the Relu activation function, take the function output result as the feature map;

[0061] Step S5: Perform feature sampling through the pooling layer to compress the features;

[0062] Step S6: Calculate the probability that the image is predicted to be each category through the activation function in the fully connected layer;

[0063] Step S7: Take the category corresponding to the maximum probability as the category of this prediction;

[0064] Step S8: Calculate the loss value and update the network parameters through backpropagation;

[0065] Step S9: Repeat steps S3 to S8 until the network converges and the training is completed.

[0066] In this embodiment, the feature extraction structure of the convolutional neural network specifically includes:

[0067] Backbone network, used to extract basic features from the image; this backbone network adopts a pre-trained deep convolutional neural network, which can effectively extract low-level features in the image, such as edge, texture and other information, laying a foundation for subsequent feature extraction and fusion.

[0068] Multi-layer convolutional module, used to further extract the basic features; this module consists of multiple convolutional layers, and each convolutional layer is followed by a batch normalization layer and a ReLU activation function. Through layer-by-layer feature extraction, more abstract and semantic feature representations can be obtained.

[0069] The reverse fusion network module is used to fuse features of different depths. This module adopts a feature pyramid structure to fuse deep features with shallow features through upsampling, realizing the complementarity of features at different levels and enhancing the network's recognition ability of the patient's body posture and state.

[0070] The feature fusion network module is used to aggregate feature maps according to size, generating three feature maps of different sizes containing features of different depths. Through feature aggregation operations, this module integrates features from different levels to form a multi-scale feature representation, effectively improving the network's detection ability for targets of different sizes.

[0071] The prediction module is used to predict the position and category of the target bounding box. This module processes the fused feature map, predicts the position coordinates and size of the target through regression, and simultaneously predicts the category of the target through classification, achieving precise recognition of the patient's body posture and state.

[0072] Example Four:

[0073] The present invention further includes a data storage and analysis module, which is used to record and analyze the usage data of the flat lamp, the patient state change data, and the environmental light data, providing data support for optimizing the flat lamp control strategy. Specifically, this module includes:

[0074] The data acquisition unit is used to acquire the working state data of the flat lamp, the patient state change data, and the environmental light data;

[0075] The data storage unit is used to store and backup the acquired data;

[0076] The data analysis unit is used to statistically analyze the stored data and mine the laws and trends in the data;

[0077] The strategy optimization unit is used to optimize the flat lamp control strategy according to the data analysis results, improving the intelligence level of the system.

[0078] Through the data storage and analysis module, the system can achieve self-learning and optimization. As the usage time increases, the control strategy of the flat lamp will more conform to the actual needs of medical staff and patients, improving the user experience and satisfaction.

[0079] Example Five:

[0080] In this embodiment, the voice interaction module adopts noise purification technology, which can effectively filter the background noise in the medical environment and improve the quality of speech recognition and speech synthesis. This noise purification technology includes:

[0081] The adaptive noise cancellation algorithm automatically adjusts the filtering parameters according to the characteristics of the environmental noise;

[0082] Frequency domain suppression technology suppresses noise components in the frequency domain and retains the speech signal;

[0083] Deep learning speech enhancement method enhances the speech signal through a neural network model;

[0084] Multi-channel beamforming technology uses a microphone array to improve the signal-to-noise ratio of the target speech.

[0085] Through these noise purification technologies, even in a noisy medical environment, the system can accurately recognize the speech commands of medical staff and provide clear voice feedback, greatly improving the convenience and reliability of voice interaction.

[0086] Example Six:

[0087] In this embodiment, the automatic adjustment strategy of the flat lamp control module also takes into account the special requirements of different lighting scenarios:

[0088] Diagnostic lighting mode: When it is recognized that medical staff are performing patient diagnosis, it is automatically adjusted to a high-brightness, neutral color temperature lighting effect to ensure visual clarity during the diagnosis process;

[0089] Treatment lighting mode: When it is recognized that a treatment operation is in progress, the lighting angle and range are automatically adjusted according to the specific treatment type to provide the best lighting conditions for the treatment;

[0090] Rest lighting mode: When it is recognized that the patient is in a resting state, it is automatically adjusted to a low-brightness, warm color temperature lighting effect to create a comfortable resting environment;

[0091] Emergency lighting mode: When it is judged to be an emergency situation through voice commands or visual recognition, it is immediately adjusted to the highest brightness to ensure clear visibility of emergency operations;

[0092] Night lighting mode: During the night period, the brightness is automatically reduced and adjusted to a warm tone to reduce interference with the patient's sleep.

[0093] Through these preset lighting modes and automatic recognition switching mechanisms, the system can provide the most suitable lighting effects for different medical scenarios, improve medical work efficiency and enhance patient comfort.

[0094] The medical voice purification flat lamp integrating automatic dimming and patient monitoring recognition of the present invention realizes the intelligent control of the medical flat lamp through the organic combination of a light detection module, a visual detection module, a flat lamp control module and a voice interaction module. The system can automatically adjust the brightness, color temperature and light angle of the flat lamp according to the environmental light intensity and the patient's body posture and state, and provides a convenient control method and user-friendly service through the voice interaction function.

[0095] The system solves the technical problems of the traditional medical flat lamp, such as single function and simple control method. Through deep learning technology, it realizes the accurate recognition of the patient's body posture and state, providing data support for the intelligent control of the flat lamp. At the same time, the system has data storage and analysis functions, which can continuously optimize the control strategy and improve the system performance.

[0096] The medical voice purification flat lamp of the present invention not only improves the comfort and safety of the medical environment, but also provides more convenient working conditions for medical staff, and has broad application prospects and market value.

[0097] Although the specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are only examples. Without departing from the principle and essence of the present invention, those skilled in the art can make various omissions, substitutions and changes to the details of the above methods and systems. For example, combining the above method steps, thus performing substantially the same function in a substantially the same way to achieve substantially the same result, belongs to the scope of the present invention. Therefore, the scope of the present invention is only defined by the appended claims.

Claims

1. A medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification, characterized in that Including: A light detection module, which is used to collect the ambient light intensity data of the medical scenario; A vision detection module, which is used to collect patient images and identify the patient's body posture and status; A flat lamp control module, which is used to adjust the brightness, color temperature and light angle of the flat lamp according to the processing results; A voice interaction module, which is used to realize the real-time voice interaction between medical staff and patients; The system takes the MCU as the core processing unit, receives the signals of the light detection module and the vision detection module, analyzes and processes them, and sends control instructions to the flat lamp control module and the voice interaction module to realize the intelligent control of the medical flat lamp.

2. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 1, characterized in that, The light detection module includes: a photosensitive sensor, which is used to detect the ambient light intensity; a signal conditioning circuit, which is used to convert the electrical signal output by the photosensitive sensor into a digital signal that can be processed by the MCU; The photosensitive sensor adopts a voltage division circuit method, and its resistance value decreases with the increase of light intensity and increases with the decrease of light intensity. By monitoring the change of the voltage value in the voltage division circuit, the change of the light intensity in the current environment is reflected.

3. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 1, characterized in that The vision detection module includes: a camera, which is used to collect patient images; an image preprocessing unit, which is used to preprocess the collected images; a convolutional neural network, which is used to identify the patient's body posture and status for the preprocessed images; The convolutional neural network includes a convolutional layer, a Relu activation function layer, a pooling layer, a fully connected layer and a dropout layer, and realizes the identification of the patient's body posture and status by training on the labeled patient body posture and status image data set.

4. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 3, characterized in that The feature extraction structure of the convolutional neural network includes: a backbone network, which is used to extract basic features from the image; a multi-layer convolutional module, which is used to further extract the basic features; a reverse fusion network module, which is used to fuse features of different depths; a feature fusion network module, which is used to aggregate feature maps according to the size, and generate three feature maps of different sizes and containing different depths of features; a prediction module, which is used to predict the position and category of the target bounding box.

5. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 1, characterized in that The flat lamp control module includes: a control signal processing unit, which is used to receive the control instructions sent by the MCU; an optoelectronic isolator, which is used to realize the electrical isolation between the control circuit and the execution circuit; a flat lamp drive circuit, which is used to drive the brightness, color temperature and light angle adjustment device of the flat lamp; The flat lamp control circuit adopts a circuit isolation and two-way power supply design method, where the input end receives the control signal output by the controller, and the output end is connected to the medical flat lamp supplementary light switch circuit and the voice information control circuit.

6. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 1, characterized in that, The voice interaction module includes: a voice collection unit, which is used to collect the voice signals of medical staff and patients; a voice recognition unit, which is used to convert the voice signals into text information; a voice synthesis unit, which is used to convert the text information into voice signals; a noise purification unit, which is used to filter out environmental noise and improve the quality of voice signals; a speaker, which is used to play the synthesized voice signals.

7. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 1, wherein The control strategy of the MCU includes: an automatic dimming strategy based on the ambient light intensity; a flat lamp control strategy based on the patient's body posture and status; a control strategy based on the voice instructions of medical staff; a voice reminder strategy based on the change of the patient's status.

8. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 7, wherein The automatic dimming strategy based on ambient light intensity is as follows: when the ambient light intensity is lower than the preset threshold, automatically increase the brightness of the panel light; when the ambient light intensity is higher than the preset threshold, automatically decrease the brightness of the panel light; set different light intensity thresholds according to different time periods and different medical scenarios.

9. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 7, characterized in that, The control strategy of the panel light based on the patient's body posture and status is as follows: when it is recognized that the patient is in a sleeping state, automatically decrease the brightness of the panel light and adjust it to a warm color temperature; when it is recognized that the patient is in a reading state, automatically adjust the brightness and light angle of the panel light to provide appropriate reading illumination; when it is recognized that the patient is in a treatment state, automatically adjust the brightness and color temperature of the panel light to meet medical needs.

10. The medical voice purification flat lamp integrating automatic dimming and patient monitoring and identification according to claim 1, characterized in that, The system further includes a data storage and analysis module, which is used to record and analyze the usage data of the panel light, the patient status change data, and the ambient light data, so as to provide data support for optimizing the panel light control strategy.

Images