Intelligent Smoke Removal Device with AI Smoke Recognition Function

Abstract

The present invention discloses an intelligent smoke removal device with AI smoke recognition function, which relates to the technical field of air purification. It includes a smoke removal action control unit, a smoke intelligent detection unit, a negative pressure smoke removal module and a foot pedal module. The smoke removal action control unit includes a touch screen, a smoke action execution module, a filter element management module and a power plug. The power plug is used to supply power to the smoke removal device. The smoke intelligent detection unit includes a data processing sub-unit and an AI smoke recognition sub-unit. The AI smoke recognition sub-unit is used to receive the data detected by the smoke detector and the air quality sensor, and perform identification and classification through machine learning algorithms to judge whether there is smoke and particulate matter. The present invention uses an AI processor for deep learning to achieve accurate recognition of smoke, avoid misjudgment and missed judgment. Doctors can intuitively understand the smoke concentration in the operating room and the operation status of the smoke removal system, which is convenient for adjustment and optimization, and improves the success rate of the operation.

Description

Technical Field

[0001] The present invention relates to the technical field of air purification, and particularly to an intelligent smoke removal device with AI smoke recognition function. Background Art

[0002] In the process of social development, artificial intelligence technology has gradually penetrated into all fields of our lives. In the medical field, the application of artificial intelligence technology is also becoming increasingly widespread, bringing many conveniences to doctors and patients.

[0003] In recent years, the problem of smoke in the operating room has attracted wide attention, because smoke not only affects the doctor's surgical operation, but also may pose a potential threat to the patient's health. The smoke generated during the operation mainly comes from surgical instruments, electrosurgical knives, lasers and other equipment. These smokes contain a large amount of harmful substances, such as carbon particles, heavy metals, etc., which will have an adverse impact on human health; the smoke will also affect the doctor's line of sight and reduce the success rate of the operation.

[0004] To solve this problem, there is an urgent need for an intelligent smoke removal device with AI smoke recognition function to change this situation. Summary of the Invention

[0005] The purpose of the present invention is to solve the disadvantages existing in the prior art, and to propose an intelligent smoke removal device with AI smoke recognition function. Its advantages lie in using an AI processor for deep learning to achieve accurate recognition of smoke, avoiding misjudgment and missed judgment. Doctors can intuitively understand the smoke concentration in the operating room and the operating status of the smoke removal system, which is convenient for adjustment and optimization, and improves the success rate of the operation.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] An intelligent smoke removal device with AI smoke recognition function, including a smoke removal action control unit, a smoke intelligent detection unit, a negative pressure smoke removal module and a foot pedal module. The smoke removal action control unit includes a touch screen, a smoke action execution module, a filter element management module and a power plug, and the power plug is used to supply power to the smoke removal device;

[0008] The intelligent smoke detection unit includes a data processing subunit and an AI smoke recognition subunit. The AI smoke recognition subunit is used to receive the data detected by the smoke detector and the air quality sensor, and perform identification and classification through machine learning algorithms to determine whether there is smoke and particulate matter. When the AI smoke recognition subunit determines the existence of smoke and particulate matter, it will transmit the unit detection result to the smoke removal action control unit through the USB interface, and the smoke removal action control unit will start the negative pressure smoke removal module to remove smoke, realizing the linkage and execution of the device; in this process, the smoke removal action control unit can also adjust the rotation speed of the fan according to the detected air quality to optimize the smoke removal effect.

[0009] The present invention is further configured such that the smoke removal device further includes a video acquisition module, which is used to acquire the video signal of the surgical process. This module can call the video summary storage algorithm to locally retain the summary video of the video when the user needs it.

[0010] The present invention is further configured such that the smoke removal action control unit is an MCU board, and the intelligent smoke detection unit is an NPU / GPU main board.

[0011] The present invention is further configured such that the touch screen can perform touch operations. When the video signal is input, the NPU / GPU main board runs the AI algorithm to perform image acquisition and smoke algorithm processing, export the surgical video signal, and the MCU board is responsible for the execution of the smoke removal action and the management of the filter element, processes the source of the smoke removal signal, sets parameters, and operates the system through the negative pressure smoke removal module to extract the smoke gas in the cavity.

[0012] The present invention is further configured such that the smoke removal device selects MobileNet_V3 as the main body of the smoke detection model, and constructs the intelligent smoke detection unit based on the local training, edge conversion and edge deployment methods.

[0013] The present invention is further configured such that the intelligent smoke detection unit accesses the endoscopic image in real time through interfaces such as HDMI. When smoke is identified in the image, it outputs a predefined signal to the USB port. When the smoke removal action control unit receives the predefined signal, it performs the smoke removal action. When the smoke removal action control unit does not receive the predefined signal, the smoke removal action stops.

[0014] The present invention is further configured such that the smoke removal device selects MobileNet_V3 as the main body of the smoke detection model, and constructs the intelligent smoke detection unit based on the local training, edge conversion and edge deployment methods. The calculation and storage of this unit are mainly composed of NVDIA JETSON AGX ORIN (GPU edge processor kit) and 1TB.M2 (solid state drive).

[0015] The present invention is further configured such that the intelligent smoke detection unit accesses the endoscopic image in real time through interfaces such as HDMI. When it identifies that there is smoke in the image, it outputs a predefined signal to the USB port. When the smoke removal action control unit receives the predefined signal, it performs the smoke removal action. When the smoke removal action control unit does not receive the predefined signal, the smoke removal action stops.

[0016] The present invention is further configured such that while the intelligent smoke detection unit detects whether there is smoke, the edge computing unit runs the surgical video summary algorithm in real time. Without affecting the video image quality, this algorithm removes similar frames, performs key frame scoring on the remaining frames, and uses the knapsack optimization algorithm to store the key information of the video at the original resolution.

[0017] The beneficial effects of the present invention are as follows: By using AI technology, the present invention realizes intelligent recognition and automatic processing, solves the problem of interference caused by smoke during medical surgery, uses an AI processor for deep learning to accurately identify smoke, and avoids misjudgment and missed judgment. At the same time, the video acquisition and medical surgery smoke recognition functions are added, enabling doctors to timely understand the surgical situation and take corresponding measures. Through the control interface, doctors can intuitively understand the smoke concentration in the operating room and the operating status of the smoke removal system, facilitating adjustment and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is the overall structural schematic diagram of the intelligent smoke removal device with AI smoke recognition function proposed by the present invention;

[0019] Figure 2 is the structural schematic diagram of the intelligent smoke detection unit function of the intelligent smoke removal device with AI smoke recognition function proposed by the present invention;

[0020] Figure 3 is the structural schematic diagram of the intelligent smoke detection video summary of the intelligent smoke removal device with AI smoke recognition function proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The technical solutions of this patent will be further described in detail below in conjunction with the specific embodiments.

[0022] The embodiments of this patent will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below by referring to the drawings are exemplary and are only used to explain this patent and should not be construed as a limitation of this patent.

[0023] In the description of this patent, it should be understood that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings. These are only for the convenience of describing this patent and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to this patent.

[0024] In the description of this patent, it should be noted that unless otherwise clearly specified and defined, the terms "installation", "connection", "linkage", "setting" should be understood in a broad sense. For example, it can be fixedly connected and set, or detachably connected and set, or integrally connected and set. For those of ordinary skill in the art, the specific meanings of the above terms in this patent can be understood according to specific circumstances.

[0025] Referring to Figure 1 , a smart smoke removal device with AI smoke recognition function, including a smoke removal action control unit, a smoke intelligent detection unit, a negative pressure smoke removal module and a foot pedal module. The smoke removal action control unit includes a touch screen, a smoke action execution module, a filter element management module and a power plug. The power plug is used to supply power to the smoke removal device;

[0026] The smoke intelligent detection unit includes a data processing sub-unit and an AI smoke recognition sub-unit. The AI smoke recognition sub-unit is used to receive the data detected by the smoke detector and the air quality sensor, and perform recognition and classification through machine learning algorithms to determine whether there is smoke and particulate matter. When the AI smoke recognition sub-unit determines that there is smoke and particulate matter, it will transmit the unit detection result to the smoke removal action control unit through the USB interface, and the smoke removal action control unit will start the negative pressure smoke removal module to remove smoke, for the linkage and execution of the device; in this process, the smoke removal action control unit can also adjust the rotation speed of the fan according to the detected air quality to optimize the smoke removal effect, reduce the smoke concentration in the cavity environment, and improve the clarity of the surgical field.

[0027] In this embodiment, the smoke removal device further includes a video acquisition module. The video acquisition module is used to acquire the video signal of the surgical process. This module can call the video summary storage algorithm, and when the user needs it, it can locally retain the summary video of the video; the smoke removal action control unit is an MCU board, and the smoke intelligent detection unit is an NPU / GPU main board; the touch screen can perform touch operations. When the video signal is input, the NPU / GPU main board runs the AI algorithm to perform image acquisition and smoke algorithm processing, and exports the surgical video signal. The MCU board is responsible for the execution of the smoke removal action and the management of the filter element, processes the source of the smoke removal signal, sets parameters, and operates the system through the negative pressure smoke removal module to extract the smoke gas in the cavity.

[0028] Referring to Figure 2 , the smoke removal device selects MobileNet_V3 as the main body of the smoke detection model, and constructs a smoke intelligent detection unit based on local training, edge conversion and edge deployment methods; the smoke intelligent detection unit accesses the endoscopic image in real time through interfaces such as HDMI. When smoke is identified in the image, a predefined signal is output to the USB port. When the smoke removal action control unit receives the predefined signal, it performs the smoke removal action. When the smoke removal action control unit does not receive the predefined signal, the smoke removal action stops.

[0029] Referring to Figure 3 , the smoke removal device selects MobileNet_V3 as the main body of the smoke detection model, and constructs a smoke intelligent detection unit based on local training, edge conversion and edge deployment methods. The calculation and storage of this unit are mainly composed of NVDIA JETSON AGX ORIN (GPU edge processor kit) and 1TB.M2 (solid state drive); the smoke intelligent detection unit accesses the endoscopic image in real time through interfaces such as HDMI. When smoke is identified in the image, a predefined signal is output to the USB port. When the smoke removal action control unit receives the predefined signal, it performs the smoke removal action. When the smoke removal action control unit does not receive the predefined signal, the smoke removal action stops.

[0030] Furthermore, while the smoke intelligent detection unit detects whether there is smoke, the edge computing unit runs the surgical video summary algorithm in real time. Without affecting the video image quality, this algorithm removes similar frames, performs key scoring on the remaining frames, and uses the knapsack optimization algorithm to achieve the storage of key video information at the original resolution.

[0031] By using AI technology to achieve intelligent recognition and automatic processing, the problem of interference caused by smoke during medical surgery is solved. Deep learning is carried out using an AI processor to achieve accurate recognition of smoke and avoid misjudgment and missed judgment. At the same time, the video acquisition and medical surgery smoke recognition functions are added, enabling doctors to timely understand the surgical situation and take corresponding measures. Through the control interface, doctors can intuitively understand the smoke concentration in the operating room and the operating status of the smoke removal system, which is convenient for adjustment and optimization.

[0032] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. An intelligent smoke removal device with AI smoke recognition function, characterized in that, It includes a smoke removal action control unit, a smoke intelligent detection unit, a negative pressure smoke removal module and a foot pedal module. The smoke removal action control unit includes a touch screen, a smoke action execution module, a filter element management module and a power plug, and the power plug is used to supply power to the smoke removal device; The smoke intelligent detection unit includes a data processing sub-unit and an AI smoke recognition sub-unit. The AI smoke recognition sub-unit is used to receive the data detected by the smoke detector and the air quality sensor, and perform recognition and classification through machine learning algorithms to determine whether there is smoke and particulate matter. When the AI smoke recognition sub-unit determines that there is smoke and particulate matter, it will transmit the unit detection result to the smoke removal action control unit through the USB interface, and the smoke removal action control unit will start the negative pressure smoke removal module to remove smoke, realizing the linkage and execution of the device; during this process, the smoke removal action control unit adjusts the rotation speed of the fan according to the detected air quality; The smoke removal device further includes a video acquisition module, which is used to acquire the video signal of the surgical process. This module can call the video summary storage algorithm to locally retain the summary video of the video when the user needs it; the smoke removal device selects MobileNet_V3 as the main body of the smoke detection model and constructs the smoke intelligent detection unit based on the local training, edge conversion and edge deployment methods; the smoke intelligent detection unit is connected to the endoscopic image in real time through the HDMI interface. When it recognizes that there is smoke in the image, it outputs a convention signal to the USB port, and the smoke removal action control unit performs the smoke removal action when it receives the convention signal, and stops the smoke removal action when the smoke removal action control unit does not receive the convention signal.

2. The intelligent smoke removal device with AI smoke recognition function according to claim 1, wherein, The smoke removal action control unit is an MCU board, and the smoke intelligent detection unit is an NPU / GPU main board.

3. The intelligent smoke removal device with AI smoke recognition function according to claim 2, characterized in that, The touch screen can perform touch operations. When the video signal is input, the NPU / GPU main board runs the AI algorithm to perform image acquisition and smoke algorithm processing, and exports the surgical video signal. The MCU board is responsible for the execution of the smoke removal action and the management of the filter element, processes the source of the smoke removal signal, sets parameters, and operates the system through the negative pressure smoke removal module to extract the smoke gas in the cavity.

4. The intelligent smoke removal device with AI smoke recognition function according to claim 3, characterized in that, While the smoke intelligent detection unit detects whether there is smoke, the edge computing unit runs the surgical video summary algorithm in real time. Without affecting the video picture quality, this algorithm removes similar frames, performs key frame scoring on the remaining frames, and uses the knapsack optimization algorithm to realize the storage of the key information of the video at the original resolution.

Images