CN117062280B - Automatic following system of neurosurgery self-service operating lamp - Google Patents
Automatic following system of neurosurgery self-service operating lamp Download PDFInfo
- Publication number
- CN117062280B CN117062280B CN202311039098.7A CN202311039098A CN117062280B CN 117062280 B CN117062280 B CN 117062280B CN 202311039098 A CN202311039098 A CN 202311039098A CN 117062280 B CN117062280 B CN 117062280B
- Authority
- CN
- China
- Prior art keywords
- doctor
- network
- operating lamp
- actor
- action
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 claims abstract description 12
- 230000000694 effects Effects 0.000 claims abstract description 6
- 230000009471 action Effects 0.000 claims description 53
- 238000000034 method Methods 0.000 claims description 31
- 230000006870 function Effects 0.000 claims description 26
- 230000008569 process Effects 0.000 claims description 26
- 238000005286 illumination Methods 0.000 claims description 23
- 238000001514 detection method Methods 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 14
- 238000004891 communication Methods 0.000 claims description 13
- 238000013135 deep learning Methods 0.000 claims description 10
- 238000001356 surgical procedure Methods 0.000 claims description 10
- 238000013461 design Methods 0.000 claims description 7
- 230000003287 optical effect Effects 0.000 claims description 7
- 238000011084 recovery Methods 0.000 claims description 7
- 238000002789 length control Methods 0.000 claims description 6
- 238000012423 maintenance Methods 0.000 claims description 6
- 230000003449 preventive effect Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000004092 self-diagnosis Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000007774 longterm Effects 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 3
- 238000004387 environmental modeling Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000015654 memory Effects 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims 1
- 230000033001 locomotion Effects 0.000 description 19
- 238000012360 testing method Methods 0.000 description 8
- 238000011156 evaluation Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008713 feedback mechanism Effects 0.000 description 2
- 230000009347 mechanical transmission Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 241000288105 Grus Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
- H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
- H05B47/10—Controlling the light source
- H05B47/105—Controlling the light source in response to determined parameters
- H05B47/115—Controlling the light source in response to determined parameters by determining the presence or movement of objects or living beings
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F21—LIGHTING
- F21V—FUNCTIONAL FEATURES OR DETAILS OF LIGHTING DEVICES OR SYSTEMS THEREOF; STRUCTURAL COMBINATIONS OF LIGHTING DEVICES WITH OTHER ARTICLES, NOT OTHERWISE PROVIDED FOR
- F21V14/00—Controlling the distribution of the light emitted by adjustment of elements
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F21—LIGHTING
- F21V—FUNCTIONAL FEATURES OR DETAILS OF LIGHTING DEVICES OR SYSTEMS THEREOF; STRUCTURAL COMBINATIONS OF LIGHTING DEVICES WITH OTHER ARTICLES, NOT OTHERWISE PROVIDED FOR
- F21V21/00—Supporting, suspending, or attaching arrangements for lighting devices; Hand grips
- F21V21/14—Adjustable mountings
- F21V21/15—Adjustable mountings specially adapted for power operation, e.g. by remote control
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
- H05B45/00—Circuit arrangements for operating light-emitting diodes [LED]
- H05B45/20—Controlling the colour of the light
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
- H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
- H05B47/10—Controlling the light source
- H05B47/105—Controlling the light source in response to determined parameters
- H05B47/115—Controlling the light source in response to determined parameters by determining the presence or movement of objects or living beings
- H05B47/125—Controlling the light source in response to determined parameters by determining the presence or movement of objects or living beings by using cameras
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05B—ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL
- H05B47/00—Circuit arrangements for operating light sources in general, i.e. where the type of light source is not relevant
- H05B47/10—Controlling the light source
- H05B47/175—Controlling the light source by remote control
- H05B47/19—Controlling the light source by remote control via wireless transmission
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F21—LIGHTING
- F21W—INDEXING SCHEME ASSOCIATED WITH SUBCLASSES F21K, F21L, F21S and F21V, RELATING TO USES OR APPLICATIONS OF LIGHTING DEVICES OR SYSTEMS
- F21W2131/00—Use or application of lighting devices or systems not provided for in codes F21W2102/00-F21W2121/00
- F21W2131/20—Lighting for medical use
- F21W2131/205—Lighting for medical use for operating theatres
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B20/00—Energy efficient lighting technologies, e.g. halogen lamps or gas discharge lamps
- Y02B20/40—Control techniques providing energy savings, e.g. smart controller or presence detection
Abstract
The invention provides an automatic following system of a self-help operating lamp for neurosurgery, belonging to the technical field of operating appliances; including operating lamp main part, positioning system and follow control system, wherein, positioning system includes: the position sensor is arranged on the operating lamp main body, and the positioning tag is arranged on the head or the hand of the doctor and is used for monitoring the position of the doctor and sending the position information of the doctor relative to the operating lamp to the following control system in real time; the following control system is used for automatically adjusting the position of the sight line of the doctor by driving the operating lamp main body according to the received doctor position information. The invention can meet the high requirement of neurosurgery and provide better operation conditions, thereby improving the working environment of doctors and improving the treatment effect of patients.
Description
Technical Field
The invention relates to the technical field of surgical tools, in particular to an automatic following system of a self-help operating lamp for neurosurgery.
Background
Neurosurgery has extremely high requirements on the operation precision, wherein the light irradiation of the operating lamp plays a crucial role in obtaining a clear visual field for doctors. Under conventional settings, the adjustment of the operating lights often relies on third party operators, which often causes delays in the movement of the lights, while the operator may have errors in understanding the effect of the illumination specifically desired by the physician, thereby affecting the efficiency and effectiveness of the operation. In order to solve the problem, it becomes necessary to provide an automatic following system of a neurosurgery self-service operating lamp, which can follow the operation of a doctor in real time and accurately adjust the illumination position and illumination intensity.
The automatic following system of the self-service operating lamp can efficiently solve the problems that the traditional operating lamp is inaccurate in irradiation, time delay is adjusted and real-time adjustment cannot be carried out following operation of a doctor, and further solves the problem that the operating lamp cannot zoom and irradiate according to a specific operating process.
Disclosure of Invention
The invention aims to solve the technical problems that the existing operating lamp is inaccurate in irradiation, time-delay in adjustment and cannot be adjusted in real time along with the operation of doctors.
In order to solve the technical problems, the invention provides the following technical scheme:
the utility model provides an automatic following system of neurosurgery self-service operating lamp, includes operating lamp main part, positioning system and follows control system, wherein, positioning system includes: the position sensor is arranged on the operating lamp main body, and the positioning tag is arranged on the head or the hand of the doctor and is used for monitoring the position of the doctor and sending the position information of the doctor relative to the operating lamp to the following control system in real time; the following control system is used for automatically adjusting the position of the sight line of the doctor by driving the operating lamp main body according to the received doctor position information.
Preferably, the position sensor is used for receiving the signal sent by the positioning tag, and the position sensor comprises an optical sensor or a radio frequency sensor.
Preferably, the positioning system further comprises at least two 3D cameras, and the positioning system is realized by constructing a three-dimensional space coordinate system to more accurately acquire the position information of the head or the hand of the doctor, specifically by the following steps:
positioning and calibrating a camera: the two cameras are arranged on two sides of the operating lamp main body in an operating room for fixation, and the relative positions and the orientations of the two cameras are accurately determined through calibration so as to establish a local coordinate system;
stereoscopic vision matching: after capturing a scene image by a camera, finding out characteristic points related to a positioning label in the image, and performing stereo matching;
three-dimensional reconstruction: when the matched characteristic points are found in the two camera views, reconstructing a three-dimensional model in space by using the matched points;
positioning tag tracking: in the constructed three-dimensional coordinate system, the system continuously monitors and updates the position of the positioning tag, and performs feature extraction, matching and reconstruction processes on each frame of image to acquire the position information of the head or hand of the doctor in real time.
Preferably, the following control system adopts a deep learning algorithm of a long-short-term memory network (LSTM), predicts a movement track of a doctor by collecting and learning operation habits of the doctor in a history operation, and adjusts a position of an operating lamp according to a prediction result, and in the automatic following system, a doctor's movement is regarded as time-series data, and the LSTM is used for predicting a next possible movement of the doctor;
LSTM includes a forget gate (forget gate), an input gate (input gate), and an output gate (output gate), which together determine the flow of information;
inputs are x_t (input of current time) and h_ (t-1) (hidden state of last moment), and outputs are h_t (hidden state of current time), the following expression formula of LSTM is:
forgetting the door: f_t=σ (w_f· [ h_ (t-1), x_t ] +b_f)
An input door: i_t=σ (w_i· [ h_ (t-1), x_t ] +b_i)
Candidate cell status:
updating the cell state:
output door: o_t=σ (w_o [ h_ (t-1), x_t ] +b_o)
Updating the hidden state: h_t=o_t tan H (c_t)
Wherein sigma represents a sigmoid function, [ H_ (t-1), X_t ] represents that H_ (t-1) and X_t are spliced, W and b are respectively a weight matrix and bias, and X represents element multiplication;
x_t is the doctor's current action, C_t and H_t represent the internal state and predictions of the system's future actions, respectively, by training the LSTM model, learning to predict the future actions of the doctor based on the doctor's past and current actions, and adjusting the position of the operating lights based on the predictions.
Preferably, the following control system further comprises an artificial intelligence assistant, which speech recognizes the doctor's instructions to further fine control the position and illumination intensity of the operating lights.
Preferably, the operating lamp main body adopts a high-efficiency LED lamp and is provided with a color temperature adjusting module for adjusting the color temperature in real time according to the requirements of doctors so as to provide different visual effects of the requirements.
Preferably, the system also comprises a communication module, wherein the following control system is connected with an information system of a hospital through a wireless network through the communication module, and uploads and synchronizes information of the operation process in real time, wherein the information comprises the position and illumination intensity of an operation lamp, the position of the head and the hand of a doctor and any information acquired through a position sensor and a positioning tag.
Preferably, the following control system further comprises a fault detection module for detecting the operation state of the system, finding and processing problems in time, the fault detection module comprises the operation state of hardware components and the operation condition of software processes,
hardware monitoring: the fault detection module continuously monitors the operating lamp main body, the position sensor, the positioning tag and the working state of the communication module hardware component;
and (3) software monitoring: the fault detection module is responsible for monitoring the running condition of the software process, including the running of a deep learning algorithm, the collection and processing of data and the communication with a hospital information system;
preventive maintenance: the fault detection module also executes preventive maintenance tasks, and the module predicts possible problems and intervenes in advance by analyzing data of long-term operation of the system;
self-diagnosis and recovery: when a fault occurs, the system performs self-diagnosis and attempts automatic recovery.
Preferably, the operating lamp main body further includes a light focal length control module for adjusting the focusing degree of illumination, and the light focal length control module is based on an Actor-Critic model and specifically includes
Environmental modeling and state definition: in the model, the environment is defined as an operating room, and the states comprise the position of a doctor, the operation time point and the current illumination intensity and focusing degree parameters of an operation lamp;
action definition: the action is defined as the adjustment of the focusing degree of the operating lamp, and in a given state, the Actor network generates a continuous action value which represents the adjustment of the focusing degree of the operating lamp;
and (3) bonus function design: the reward function is based on surgical efficiency and physician feedback;
network structure and training: in the Actor-Critic model, an Actor network and a Critic network are realized based on a deep neural network, model training is carried out according to historical operation data of doctors in the training process, the Actor network updates a value function by maximizing rewards given by the Critic network and updating strategies, and the Critic network updates the value function by comparing expected rewards with actual rewards;
model application: in the operation process, the system acquires the position of a doctor and an operation time point in real time, then inputs the position and the operation time point into an Actor network, and the Actor network outputs an action value which determines the adjustment of the focusing degree of the operation lamp so as to adjust the focusing degree of the operation lamp according to the action value;
the model specifically operates as follows:
an Actor network function a (s, θ_a), where s represents the current state, θ_a is a parameter of the Actor network, which outputs an action a that is optimal for changing the environmental state, state s including the doctor's location, the surgical time point, and the current parameter of the surgical lamp;
a Critic network function C (s, a, θ_c), where s and a represent the current state and the action selected by the Actor, respectively, θ_c represents a parameter of the Critic network, which function outputs a value v representing the expected return for performing action a in state s;
and (3) updating a network: the network parameters θ_a and θ_c of the Actor and Critic are updated according to the following formula:
parameter updating of the Actor:
parameter update of Critic:
where alpha and beta are learning rates, R is the actual return,and->Is the gradient of the Actor and Critic networks;
model application: given the current state s, action a is selected using Actor network a (s, θ_a), then the effect of this action is evaluated using Critic network C (s, a, θ_c), and the parameters of the network are updated using the above equation, optimizing the policy.
Compared with the prior art, the invention has at least the following beneficial effects:
in the scheme, the operation efficiency is improved: the system can monitor the operation position of doctors in real time, automatically adjust the irradiation position and the illumination intensity of the operating lamp, reduce the interruption in the operation process and improve the operation efficiency.
Improving the operation precision: through accurate illumination, doctors can more clearly observe the surgical site, thereby improving the precision of the operation.
The system operation stability is improved: the fault detection module is arranged in the system, so that the problems can be found and processed in time, and the stable operation of the system is ensured.
And (3) improving positioning accuracy: and a three-dimensional space coordinate system is constructed through the 3D camera, so that the position information of the head or the hand of the doctor is acquired more accurately, and the positioning accuracy is further improved.
Accurately responds and provides the light ray focusing power which meets the operation requirement of doctors: the system is based on a reinforcement learning Actor-Critic model, can intelligently adjust the focusing degree of the operating lamp, realizes real-time light focusing degree adjustment according to specific requirements of operation, can recognize the requirements when performing large-scale operation, and correspondingly adjusts the light diffusion irradiation wider range of the operating lamp. And when doctors perform fine and small-range operation, such as operations of cutting off and the like, the system can timely adjust the light rays of the operation lamp, focus the light rays on an operation point, realize automatic intelligent zooming, and the system for intelligently adjusting the light ray focal power can accurately adapt to the change of the operation, provide illumination conditions most suitable for the operation needs, and greatly improve the operation efficiency. Meanwhile, the system can also lighten the burden of doctors, so that the doctors can concentrate on the operation, and the accuracy and the safety of the operation are improved.
Drawings
The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments of the present disclosure and, together with the description, further serve to explain the principles of the disclosure and to enable a person skilled in the pertinent art to make and use the disclosure.
FIG. 1 is a schematic diagram of a system logic according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a three-dimensional construction flow according to an embodiment of the present invention.
Detailed Description
The following detailed description of an automatic following system for a neurosurgical self-service operating lamp is provided with reference to the accompanying drawings and specific embodiments. While the invention has been described herein in terms of the preferred and preferred embodiments, the following embodiments are intended to be more illustrative, and may be implemented in many alternative ways as will occur to those of skill in the art; and the accompanying drawings are only for the purpose of describing the embodiments more specifically and are not intended to limit the invention specifically.
It should be noted that references in the specification to "one embodiment," "an example embodiment," "some embodiments," etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the relevant art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.
Generally, the terminology may be understood, at least in part, from the use of context. For example, the term "one or more" as used herein may be used to describe any feature, structure, or characteristic in a singular sense, or may be used to describe a combination of features, structures, or characteristics in a plural sense, depending at least in part on the context. In addition, the term "based on" may be understood as not necessarily intended to convey an exclusive set of factors, but may instead, depending at least in part on the context, allow for other factors that are not necessarily explicitly described.
It will be understood that the meanings of "on … …", "over … …" and "over … …" in this disclosure should be interpreted in the broadest sense so that "on … …" means not only "directly on" but also includes meaning "directly on" something with intervening features or layers therebetween, and "over … …" or "over … …" means not only "on" or "over" something, but also may include its meaning "on" or "over" something without intervening features or layers therebetween.
Furthermore, spatially relative terms such as "under …," "under …," "lower," "above …," "upper," and the like may be used herein for ease of description to describe one element or feature's relationship to another element or feature as illustrated in the figures. Spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. The device may be otherwise oriented and the spatially relative descriptors used herein may similarly be interpreted accordingly.
As shown in fig. 1 and 2, an embodiment of the present invention provides an automatic following system of a self-service operating lamp for neurosurgery, including an operating lamp body, a positioning system, and a following control system, wherein the positioning system includes: the position sensor is arranged on the operating lamp main body, and the positioning tag is arranged on the head or the hand of the doctor and is used for monitoring the position of the doctor and sending the position information of the doctor relative to the operating lamp to the following control system in real time; the following control system is used for automatically adjusting the position of the sight line of the doctor by driving the operating lamp main body according to the received doctor position information; the system is particularly designed for neurosurgery, realizes automatic and accurate following of the operating lamp, and improves the accuracy and efficiency of the operation.
The position sensor is used for receiving the signal sent by the positioning tag, and comprises an optical sensor or a radio frequency sensor, for example, if the optical sensor is used, the position of the positioning tag can be determined by measuring the intensity and the angle of light reflected from the positioning tag, and the position sensor has the characteristics of high precision and quick response regardless of the type of the position sensor so as to meet the requirement of neurosurgery.
The positioning tag is used to send out a signal that can be received and interpreted by the position sensor to monitor the doctor's position in real time. The location tag may select different modes of operation depending on the type of location sensor used. For example, if an optical sensor is used, the positioning tag may be made of a reflective material to improve the detection efficiency of the optical sensor; if a radio frequency sensor is used, the location tag will emit a radio frequency signal at a particular frequency. Regardless of the type of locating tag, it should be lightweight and not interfere with the physician's operation, while having sufficient signal strength to ensure accurate transmission of the location.
Through the cooperation of position sensor and location label, neurosurgery self-service operating lamp automatic following system can monitor doctor's position in real time accurately, makes the operating lamp can follow doctor's operation automatically and accurately, improves the precision and the efficiency of operation.
The positioning system further comprises at least two 3D cameras, and position information of the head or the hand of a doctor is more accurately obtained by constructing a three-dimensional space coordinate system, and the positioning system is realized by the following steps:
positioning and calibrating a camera: the two cameras are arranged on two sides of the operating lamp main body in an operating room for fixation, and the relative positions and the orientations of the two cameras are accurately determined through calibration so as to establish a local coordinate system;
stereoscopic vision matching: after capturing a scene image by a camera, finding out characteristic points related to a positioning tag in the image, and carrying out stereo matching, wherein the purpose of the stereo matching is to find out the same characteristic points in two camera views, which is necessary for subsequent three-dimensional reconstruction;
three-dimensional reconstruction: when the matched characteristic points are found in the two camera views, a three-dimensional model is rebuilt in space by using the matched points, the process generally comprises the steps of calculating depth information of the points, converting two-dimensional image information into three-dimensional space coordinates, and completing the construction of a three-dimensional space coordinate system;
positioning tag tracking: in the constructed three-dimensional coordinate system, the system continuously monitors and updates the position of the positioning label, and performs feature extraction, matching and reconstruction processes on each frame of image so as to acquire the position information of the head or the hand of a doctor in real time;
the above is a general procedure for constructing a three-dimensional space coordinate system using a 3D camera system. It should be noted that this process requires reliance on efficient computing platforms and advanced algorithm support to ensure real-time and accuracy of the system. Meanwhile, depending on the specific environment and requirements of the operating room, some customized designs and optimizations may be required.
The specific driving mode of the following control system is that the system receives the doctor position information and then realizes the accurate control of the operating lamp main body. This requires the use of motor drives in order to achieve flexible movements and accurate positioning. In particular, this may involve the following parts:
and (3) motor driving: in the automatic following system of the operating lamp, a stepping motor or a servo motor may be used, which are different in rotation accuracy, control manner and performance. Stepper motors can provide a defined angle of rotation and are therefore more common where precise rotation or positioning is required. The servo motor is more suitable for application occasions requiring quick response and large moment.
The control circuit: the driving of the motor needs to be accomplished by a specific control circuit. The control circuit receives the signal following the control system and then converts the signal into a motor driving signal to control the rotation of the motor. The design of the control circuit needs to take into account the accuracy, stability and response speed of signal conversion.
The mechanical transmission structure comprises: the rotation of the motor needs to be translated into movement of the operating lamp by a mechanical transmission. Such structures may include gears, belts, screws, etc., with different structures corresponding to different transmission efficiencies and accuracies.
Feedback mechanism: to achieve more accurate control, the follower control system may also include a feedback mechanism. For example, the actual position of the operating lamp can be monitored in real time by a position sensor arranged on the operating lamp and fed back to a control system to realize closed-loop control.
The above is one possible specific driving form of the following control system. It should be noted that the design of this drive form requires consideration of many factors including, but not limited to, the weight of the surgical lamp, the range of movement, the speed of response, and the accuracy requirements. The actual driving form may be optimized according to the specific design and use of the operating lamp. Deep learning has demonstrated powerful performance in many areas, including computer vision, natural language processing, recommendation systems, and the like. In our automatic following system of neurosurgery lamps, deep learning can be used to improve the accuracy of the system's prediction of the actions of the doctor, so that the lamps can follow the actions of the doctor more accurately.
One possible algorithm is Recurrent Neural Networks (RNNs) and variants thereof, such as long short-term memory networks (LSTM) and gated loop units (GRUs). These networks are all capable of processing sequence data, such as time series or text, etc. In our system, the actions of the physician can be regarded as a time series, so RNN and its variants are a good choice.
Specifically, the network may be enabled to receive historical data of doctor actions over a period of time and predict the next possible actions of the doctor. The predicted motion may be translated into a position to which the operatory lamp should be moved, which is then effected by the motor drive system. To train this network, data of doctor actions and operating lamp movement positions can be collected during the operation and then used for supervised learning.
The following control system adopts a deep learning algorithm of a long-short-term memory network (LSTM), predicts the action track of a doctor by collecting and learning the operation habit of the doctor in a history operation, and adjusts the position of an operating lamp according to a prediction result;
LSTM includes a forget gate (forget gate), an input gate (input gate), and an output gate (output gate), which together determine the flow of information;
inputs are x_t (input of current time) and h_ (t-1) (hidden state of last moment), and outputs are h_t (hidden state of current time), the following expression formula of LSTM is:
forgetting the door: f_t=σ (w_f· [ h_ (t-1), x_t ] +b_f)
An input door: i_t=σ (w_i· [ h_ (t-1), x_t ] +b_i)
Candidate cell status:
updating the cell state:
output door: o_t=σ (w_o [ h_ (t-1), x_t ] +b_o)
Updating the hidden state: h_t=o_t tan H (c_t)
Wherein sigma represents a sigmoid function, [ H_ (t-1), X_t ] represents that H_ (t-1) and X_t are spliced, W and b are respectively a weight matrix and bias, and X represents element multiplication;
x_t is the doctor's current motion (e.g., head or hand position information captured by a position sensor), C_t and H_t represent the internal state and predictions of the system's future motion, respectively, of the doctor, and by training the LSTM model, the physician learns to predict his future motion from his past and current motion, and adjusts the position of the surgical lamp based on the predictions.
The follow-up control system also comprises an artificial intelligence assistant and a voice recognition doctor to further finely control the position and illumination intensity of the operation lamp.
The operating lamp main body adopts a high-efficiency LED lamp and is provided with a color temperature adjusting module for adjusting the color temperature in real time according to the requirements of doctors so as to provide different visual effects of the requirements.
The system also comprises a communication module, wherein the follow-up control system is connected with an information system (such as an HIS (hospital information management system)) of a hospital through a wireless network, and uploads and synchronizes information of the operation process in real time, wherein the information comprises the position and illumination intensity of an operation lamp, the position of the head and the hand of a doctor and any information acquired through a position sensor and a positioning tag; through this connection, the information of the surgical procedure can be entered in real time into the database of the hospital, facilitating the statistics and analysis of the later data. For example, a data analyst may analyze movement of the operating lights and changes in illumination intensity during the surgical procedure to assess performance and efficiency of the system. They can also optimize the deep learning algorithm by analyzing the movement habits of the doctor so that the system better adapts to the doctor's surgical habits.
In addition, the real-time uploaded surgical information can also be used for remote monitoring. For example, a hospital administrator may view real-time information of a currently ongoing operation at any time to ensure that the operation is performed smoothly. Data support can also be provided for remote medical consultation in emergency.
In addition, information of the surgical procedure can also be stored as part of the electronic medical record for future review. For example, if it is desired to review the surgical procedure, or to evaluate and learn the effectiveness of the procedure, such detailed surgical procedure data would be a very valuable resource.
In general, the real-time wireless connection of the follow-up control system and the hospital information system greatly improves the informatization degree of the operation process, provides important data support for medical service, and is beneficial to improving the medical quality and efficiency.
The following control system also comprises a fault detection module for detecting the running state of the system and timely finding and processing problems, the fault detection module comprises the working state of hardware components and the running condition of software processes,
hardware monitoring: the fault detection module continuously monitors the operating state of the operating lamp body, the position sensor, the positioning tag and the communication module hardware component, for example, detects whether the movement of the operating lamp is normal, whether the light source is stable, whether the sensor accurately collects data, whether the communication module normally transmits and receives data, and the like. If any anomalies are detected, the system will immediately issue a warning and describe the nature of the problem and the possible solutions as detailed as possible.
And (3) software monitoring: the fault detection module is responsible for monitoring the operation of the software process, including the operation of the deep learning algorithm, the collection and processing of data and the communication with the hospital information system, and if the system finds any software errors, such as the algorithm fails to operate properly, the data is lost or the communication fails, the system will immediately record these problems and attempt an automatic recovery.
Preventive maintenance: the fault detection module also performs preventive maintenance tasks by analyzing data of the long-term operation of the system, the module predicts possible problems and intervenes in advance, for example, if the data shows that a certain part of the operating lamp gradually declines in performance over a period of time, the system can predict in advance that the part is likely to fail and remind relevant personnel to check or replace.
Self-diagnosis and recovery: upon failure, the system performs a self-diagnosis and attempts an automatic recovery, e.g., if a hardware failure is detected, the system may attempt to restart the hardware component; if a software failure is detected, the system may attempt to reload or update the problematic software module.
Through the four main functions, the fault detection module can effectively monitor and maintain the running state of the system, and ensure the stability and reliability of the system so as to meet the high requirements of neurosurgery on the equipment performance.
The operating lamp main body also comprises a lamplight focal length control module for adjusting the focusing degree of illumination, wherein the lamplight focal length control module is based on an Actor-Critic model and specifically comprises the following components
Environmental modeling and state definition: in the model, the environment is defined as an operating room, and the states comprise the position of a doctor, the operation time point and the current illumination intensity and focusing degree parameters of an operation lamp;
action definition: the action is defined as the adjustment of the focusing degree of the operating lamp, and in a given state, the Actor network generates a continuous action value which represents the adjustment of the focusing degree of the operating lamp;
and (3) bonus function design: the reward function is based on the surgical efficiency and the doctor's feedback, e.g., if the doctor's surgical efficiency is improved, or if the doctor's feedback is positive, the system will get a positive reward. Conversely, if the efficiency of the procedure is reduced, or if the physician's feedback is negative, the system will be rewarded negatively;
network structure and training: in the Actor-Critic model, an Actor network and a Critic network are realized based on a deep neural network, model training is carried out according to historical operation data of doctors in the training process, the Actor network updates a value function by maximizing rewards given by the Critic network and updating strategies, and the Critic network updates the value function by comparing expected rewards with actual rewards;
model application: in the operation process, the system acquires the position of a doctor and an operation time point in real time, then inputs the position and the operation time point into an Actor network, and the Actor network outputs an action value which determines the adjustment of the focusing degree of the operation lamp so as to adjust the focusing degree of the operation lamp according to the action value;
in addition, the system can upload and synchronize the information of the operation process in real time through the wireless network connection with the hospital information system, thereby further improving the operation efficiency and the safety. Meanwhile, by continuously monitoring and learning the operation behaviors of doctors, the system can also optimize and improve the performance of the system.
In practice, we need to ensure that the data collection of the surgical procedure is complete and accurate, and in the case of compliance, to ensure privacy protection for the patient and doctor. We also need to perform extensive training and testing to ensure the effectiveness and stability of the model;
the model specifically operates as follows:
an Actor network function a (s, θ_a), where s represents the current state, θ_a is a parameter of the Actor network, which outputs an action a that is optimal for changing the environmental state, state s including the doctor's location, the surgical time point, and the current parameter of the surgical lamp;
a Critic network function C (s, a, θ_c), where s and a represent the current state and the action selected by the Actor, respectively, θ_c represents a parameter of the Critic network, which function outputs a value v representing the expected return for performing action a in state s;
and (3) updating a network: the network parameters θ_a and θ_c of the Actor and Critic are updated according to the following formula:
parameter updating of the Actor:
parameter update of Critic:
where alpha and beta are learning rates, R is the actual return,and->Is the gradient of the Actor and Critic networks;
model application: using an Actor network A (s, theta_a) to select an action a given a current state s, then using a Critic network C (s, a, theta_c) to evaluate the effect of the action, and then using the above formula to update parameters of the network to optimize the strategy;
this model is based on the idea of reinforcement learning, by continually trying and learning in the environment, gradually finding the optimal strategy for a particular task. In this problem, the task is to adjust the degree of focus of the operating lamp to provide optimal lighting conditions, by using this model we can automatically adjust the degree of focus of the operating lamp according to the doctor's position and the operating time point.
The system of the invention is tested, and the test process is as follows:
system integration and debugging: firstly, integrating all hardware devices (such as an operating lamp main body, a position sensor, a positioning tag, a 3D camera, a communication module and the like) and a software system (comprising a positioning system, a following control system, a fault detection module and the like), and carrying out necessary debugging to ensure the normal operation of the system.
Model training: the physician's motion data during the procedure is collected and used to train the LSTM model with the aim of enabling the model to predict future motion of the physician based on his historical and current motion.
And (3) system testing: in a simulation environment of a non-real operation, the function of the system is tested, for example, to let a doctor simulate the operation, and whether the operation lamp can accurately follow the action of the doctor is tested.
Actual application evaluation: in an actual surgical procedure, the system is applied and data of the surgical procedure is recorded, including the position and illumination intensity of the surgical lights, the position of the doctor's head and hands.
Analysis:
model evaluation: the prediction accuracy of the model is evaluated by comparing the doctor's actions predicted by the LSTM model with the doctor's actual actions.
Evaluation of system efficacy: by analyzing the data of the actual surgical procedure, the performance of the system is evaluated, for example, whether the operating lamp can accurately and timely follow the action of a doctor, whether the illumination intensity and the focal length can meet the requirements of the operation, whether the system stably operates, and whether faults occur or not are checked.
Doctor feedback: the physician using the system is queried for feedback regarding the performance of the system to assess the utility of the system.
The following table is based on data of system tests and actual application evaluations:
in the above table, we can see the performance of the neurosurgical self-service operating lamp automatic following system in the test phase and the practical application phase, this comparison helps us to see if the performance of the system meets the expectations and further improves it, the analysis of this table is as follows:
light following precision: the method is improved by 5 percent compared with the test stage in the actual application stage. This may be due to the fact that in a real environment the system learns and adapts to the behavioural pattern of the doctor, thereby improving the accuracy.
Color temperature adjustment accuracy and focus adjustment accuracy: the actual application stage is improved because the system is self-adjusted and optimized according to the actual operation environment.
LSTM prediction accuracy: the method is improved by 5 percent in the actual application stage compared with the test stage, and the method is capable of effectively predicting the action track of a doctor in a complex real environment based on a deep learning algorithm of a long short term memory network (LSTM).
Failure occurrence rate: the drop of 2 percentage points in the practical application stage is a positive trend, which shows that the stability of the system is improved in long-term use.
Doctor satisfaction: 92% was achieved, which is a high satisfaction, indicating that most doctors are satisfied with the performance of the system.
AI helper identification accuracy: the method improves the practical application stage by 4 percent points, which indicates that the AI assistant can effectively identify and understand the instruction of the doctor.
Overall, the performance of the neurosurgical self-service operating lamp automatic following system in practical application is better than that of the test stage, which shows that the system can be effectively learned and optimized from practical application, and better performance is achieved.
The technical scheme provided by the invention improves the operation efficiency: the system can monitor the operation position of doctors in real time, automatically adjust the irradiation position and the illumination intensity of the operating lamp, reduce the interruption in the operation process and improve the operation efficiency.
Improving the operation precision: through accurate illumination, doctors can more clearly observe the surgical site, thereby improving the precision of the operation.
The system operation stability is improved: the fault detection module is arranged in the system, so that the problems can be found and processed in time, and the stable operation of the system is ensured.
And (3) improving positioning accuracy: and a three-dimensional space coordinate system is constructed through the 3D camera, so that the position information of the head or the hand of the doctor is acquired more accurately, and the positioning accuracy is further improved.
Convenient later data statistics and analysis: the following control system can be in wireless connection with an information system of a hospital, and information of a surgical process is transmitted in real time, so that statistics and analysis of later data are facilitated.
The invention is intended to cover any alternatives, modifications, equivalents, and variations that fall within the spirit and scope of the invention. In the following description of preferred embodiments of the invention, specific details are set forth in order to provide a thorough understanding of the invention, and the invention will be fully understood to those skilled in the art without such details. In other instances, well-known methods, procedures, flows, components, circuits, and the like have not been described in detail so as not to unnecessarily obscure aspects of the present invention.
Those of ordinary skill in the art will appreciate that all or a portion of the steps in implementing the methods of the embodiments described above may be implemented by a program that instructs associated hardware, and the program may be stored on a computer readable storage medium, such as: ROM/RAM, magnetic disks, optical disks, etc.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (7)
1. The automatic following system of the self-service operating lamp for the neurosurgery is characterized by comprising an operating lamp main body, a positioning system and a following control system;
the positioning system comprises: the position sensor is arranged on the operating lamp main body, and the positioning tag is arranged on the head or the hand of the doctor and is used for monitoring the position of the doctor and sending the position information of the doctor relative to the operating lamp to the following control system in real time;
the following control system is used for automatically adjusting the position of the sight line of the doctor by driving the operating lamp main body according to the received doctor position information;
the operating lamp main body adopts a high-efficiency LED lamp and is provided with a color temperature adjusting module for adjusting the color temperature in real time according to the requirements of doctors so as to provide different visual effects of the requirements;
the operating lamp main body also comprises a lamplight focal length control module for adjusting the focusing degree of illumination, wherein the lamplight focal length control module is based on an Actor-Critic model and specifically comprises the following components
Environmental modeling and state definition: in the model, the environment is defined as an operating room, and the states comprise the position of a doctor, the operation time point and the current illumination intensity and focusing degree parameters of an operation lamp;
action definition: the action is defined as the adjustment of the focusing degree of the operating lamp, and in a given state, the Actor network generates a continuous action value which represents the adjustment of the focusing degree of the operating lamp;
and (3) bonus function design: the reward function is based on surgical efficiency and physician feedback;
network structure and training: in the Actor-Critic model, an Actor network and a Critic network are realized based on a deep neural network, model training is carried out according to historical operation data of doctors in the training process, the Actor network updates a value function by maximizing rewards given by the Critic network and updating strategies, and the Critic network updates the value function by comparing expected rewards with actual rewards;
model application: in the operation process, the system acquires the position of a doctor and an operation time point in real time, then inputs the position and the operation time point into an Actor network, and the Actor network outputs an action value which determines the adjustment of the focusing degree of the operation lamp so as to adjust the focusing degree of the operation lamp according to the action value;
the model specifically operates as follows:
an Actor network function a (s, θ_a), where s represents the current state, θ_a is a parameter of the Actor network, which outputs an action a that is optimal for changing the environmental state, state s including the doctor's location, the surgical time point, and the current parameter of the surgical lamp;
a Critic network function C (s, a, θ_c), where s and a represent the current state and the action selected by the Actor, respectively, θ_c represents a parameter of the Critic network, which function outputs a value v representing the expected return for performing action a in state s;
and (3) updating a network: the network parameters θ_a and θ_c of the Actor and Critic are updated according to the following formula:
parameter updating of the Actor:
parameter update of Critic:
where alpha and beta are learning rates, R is the actual return,and->Is the gradient of the Actor and Critic networks;
model application: given the current state s, action a is selected using Actor network a (s, θ_a), then the effect of this action is evaluated using Critic network C (s, a, θ_c), and the parameters of the network are updated using the above equation, optimizing the policy.
2. The neurosurgical self-service surgical light automatic following system of claim 1, wherein the position sensor is configured to receive the position tag signal, and wherein the position sensor comprises an optical sensor or a radio frequency sensor.
3. The neurosurgical self-service surgical light automatic following system according to claim 2, wherein the positioning system further comprises at least two 3D cameras, and the position information of the head or hand of the doctor is more accurately obtained by constructing a three-dimensional space coordinate system, specifically by the following steps:
positioning and calibrating a camera: the two cameras are arranged in an operating room and fixed on two sides of the operating lamp main body, and the relative positions and the orientations of the two cameras are accurately determined through calibration so as to establish a local coordinate system;
stereoscopic vision matching: after capturing a scene image by a camera, finding out characteristic points related to a positioning label in the image, and performing stereo matching;
three-dimensional reconstruction: when the matched characteristic points are found in the two camera views, reconstructing a three-dimensional model in space by using the matched characteristic points;
positioning tag tracking: in the constructed three-dimensional coordinate system, the system continuously monitors and updates the position of the positioning tag, and performs feature extraction, matching and reconstruction processes on each frame of image to acquire the position information of the head or hand of the doctor in real time.
4. The automatic following system of a neurosurgical self-service operation lamp according to claim 1, wherein the following control system adopts a deep learning algorithm of a long-short-term memory network, predicts the action track of a doctor by collecting and learning the operation habit of the doctor in a history operation, and adjusts the position of the operation lamp according to the prediction result, in the automatic following system, the action of the doctor is regarded as time-series data, and LSTM is used for predicting the next possible action of the doctor;
the LSTM comprises a forgetting gate, an input gate and an output gate, and the flow of information is determined together;
the inputs are X_t and H_ (t-1), the output is H_t, wherein X_t is the input of the current time, H_ (t-1) is the hidden state of the last moment, H_t is the hidden state of the current time, and the following expression formula of LSTM is as follows:
forgetting the door: f_t=σ (w_f· [ h_ (t-1), x_t ] +b_f)
An input door: i_t=σ (w_i· [ h_ (t-1), x_t ] +b_i)
Candidate cell status:
updating the cell state:
output door: o_t=σ (w_o [ h_ (t-1), x_t ] +b_o)
Updating the hidden state: h_t=o_t tan H (c_t)
Wherein sigma represents a sigmoid function, [ H_ (t-1), X_t ] represents that H_ (t-1) and X_t are spliced, W and b are respectively a weight matrix and bias, and X represents element multiplication;
x_t is the doctor's current action, C_t and H_t represent the internal state and predictions of the system's future actions, respectively, by training the LSTM model, learning to predict the future actions of the doctor based on the doctor's past and current actions, and adjusting the position of the operating lights based on the predictions.
5. The automated surgical light following system of claim 1, wherein the following control system further comprises an artificial intelligence assistant, voice recognition doctor instructions to further fine control the position and illumination intensity of the surgical light.
6. The automated surgical light following system of claim 1, further comprising a communication module, wherein the following control system is connected to a hospital information system via a wireless network, and wherein the information of the surgical procedure, including the position and illumination intensity of the surgical light, the position of the doctor's head and hands, and any information obtained via position sensors and positioning tags, is uploaded and synchronized in real time.
7. The automatic following system of a neurosurgical self-service operating lamp according to claim 6, wherein the following control system further comprises a fault detection module for detecting the operation state of the system, finding and handling problems in time, the fault detection module comprising the operation state of hardware components and the operation state of software processes,
hardware monitoring: the fault detection module continuously monitors the operating lamp main body, the position sensor, the positioning tag and the working state of the communication module hardware component;
and (3) software monitoring: the fault detection module is responsible for monitoring the running condition of the software process, including the running of a deep learning algorithm, the collection and processing of data and the communication with a hospital information system;
preventive maintenance: the fault detection module also executes preventive maintenance tasks, and the module predicts possible problems and intervenes in advance by analyzing data of long-term operation of the system;
self-diagnosis and recovery: when a fault occurs, the system performs self-diagnosis and attempts automatic recovery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311039098.7A CN117062280B (en) | 2023-08-17 | 2023-08-17 | Automatic following system of neurosurgery self-service operating lamp |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311039098.7A CN117062280B (en) | 2023-08-17 | 2023-08-17 | Automatic following system of neurosurgery self-service operating lamp |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117062280A CN117062280A (en) | 2023-11-14 |
| CN117062280B true CN117062280B (en) | 2024-03-08 |
Family
ID=88665874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311039098.7A Active CN117062280B (en) | 2023-08-17 | 2023-08-17 | Automatic following system of neurosurgery self-service operating lamp |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117062280B (en) |
Citations (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080086143A (en) * | 2007-03-22 | 2008-09-25 | 한양대학교 산학협력단 | Lighting robot system using the position pursuit function and surgery having the same and lighting robot position moving method using the position pursuit function |
| CN103337094A (en) * | 2013-06-14 | 2013-10-02 | 西安工业大学 | Method for realizing three-dimensional reconstruction of movement by using binocular camera |
| CN103971378A (en) * | 2014-05-29 | 2014-08-06 | 福州大学 | Three-dimensional reconstruction method of panoramic image in mixed vision system |
| CN107062074A (en) * | 2017-04-12 | 2017-08-18 | 成都迅德科技有限公司 | A kind of adjustable focal length operating lamp with microphone |
| WO2017175232A1 (en) * | 2016-04-07 | 2017-10-12 | M.S.T. Medical Surgery Technologies Ltd. | Vocally activated surgical control system |
| CN108024839A (en) * | 2015-08-13 | 2018-05-11 | 卡尔莱宾格医疗技术有限责任两合公司 | The handle apparatus and operating lamp of surgery operating lamp with acoustic control |
| CN108549237A (en) * | 2018-05-16 | 2018-09-18 | 华南理工大学 | Preview based on depth enhancing study controls humanoid robot gait's planing method |
| CN109089365A (en) * | 2018-10-30 | 2018-12-25 | 成都大学附属医院 | A kind of intelligence adjusts the control system and method for operating lamp |
| CN109636062A (en) * | 2018-12-25 | 2019-04-16 | 湖北工业大学 | A kind of students ' behavior analysis method and system based on big data analysis |
| CN110288661A (en) * | 2019-06-19 | 2019-09-27 | 深圳市睿智龙电子科技有限公司 | Location regulation method, device, computer equipment and the storage medium of operating lamp |
| CN110989576A (en) * | 2019-11-14 | 2020-04-10 | 北京理工大学 | Target following and dynamic obstacle avoidance control method for differential slip steering vehicle |
| CN111631719A (en) * | 2020-05-21 | 2020-09-08 | 北京城市系统工程研究中心 | Method for predicting falling risk of old people |
| CN112389436A (en) * | 2020-11-25 | 2021-02-23 | 中汽院智能网联科技有限公司 | Safety automatic driving track-changing planning method based on improved LSTM neural network |
| CN112435503A (en) * | 2020-10-30 | 2021-03-02 | 江苏大学 | Intelligent automobile active collision avoidance method for identifying intention of high-risk pedestrians |
| CN112509125A (en) * | 2020-12-14 | 2021-03-16 | 广州迦恩科技有限公司 | Three-dimensional reconstruction method based on artificial markers and stereoscopic vision |
| CN112883738A (en) * | 2021-03-23 | 2021-06-01 | 西南交通大学 | Medical entity relation extraction method based on neural network and self-attention mechanism |
| CN113111581A (en) * | 2021-04-09 | 2021-07-13 | 重庆邮电大学 | LSTM trajectory prediction method combining space-time factors and based on graph neural network |
| CN113254197A (en) * | 2021-04-30 | 2021-08-13 | 西安电子科技大学 | Network resource scheduling method and system based on deep reinforcement learning |
| CN113269963A (en) * | 2021-05-20 | 2021-08-17 | 东南大学 | Internet vehicle signal lamp control intersection economic passing method based on reinforcement learning |
| CN113573284A (en) * | 2021-06-21 | 2021-10-29 | 吉林大学 | Random access backoff method for large-scale machine type communication based on machine learning |
| CN113763723A (en) * | 2021-09-06 | 2021-12-07 | 武汉理工大学 | Traffic signal lamp control system and method based on reinforcement learning and dynamic timing |
| CN115033022A (en) * | 2022-06-28 | 2022-09-09 | 华南理工大学 | DDPG unmanned aerial vehicle landing method based on expert experience and oriented to mobile platform |
| CN115311405A (en) * | 2021-05-06 | 2022-11-08 | 珠海市迪谱医疗科技有限公司 | Three-dimensional reconstruction method of binocular endoscope |
| CN115665941A (en) * | 2022-11-04 | 2023-01-31 | 浙江大学医学院附属第一医院 | Sight following surgical illuminating lamp and sight following method |
| CN116259175A (en) * | 2022-12-30 | 2023-06-13 | 北京邮电大学 | Vehicle speed recommendation method and device for diversified dynamic signal lamp modes |
| WO2023123906A1 (en) * | 2021-12-31 | 2023-07-06 | 深圳云天励飞技术股份有限公司 | Traffic light control method and related device |
| CN116513273A (en) * | 2023-01-03 | 2023-08-01 | 南京工业大学 | Train operation scheduling optimization method based on deep reinforcement learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060142740A1 (en) * | 2004-12-29 | 2006-06-29 | Sherman Jason T | Method and apparatus for performing a voice-assisted orthopaedic surgical procedure |
-
2023
- 2023-08-17 CN CN202311039098.7A patent/CN117062280B/en active Active
Patent Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080086143A (en) * | 2007-03-22 | 2008-09-25 | 한양대학교 산학협력단 | Lighting robot system using the position pursuit function and surgery having the same and lighting robot position moving method using the position pursuit function |
| CN103337094A (en) * | 2013-06-14 | 2013-10-02 | 西安工业大学 | Method for realizing three-dimensional reconstruction of movement by using binocular camera |
| CN103971378A (en) * | 2014-05-29 | 2014-08-06 | 福州大学 | Three-dimensional reconstruction method of panoramic image in mixed vision system |
| CN108024839A (en) * | 2015-08-13 | 2018-05-11 | 卡尔莱宾格医疗技术有限责任两合公司 | The handle apparatus and operating lamp of surgery operating lamp with acoustic control |
| CN109863553A (en) * | 2016-04-07 | 2019-06-07 | M.S.T.医学外科技术有限公司 | The operation control system of voice activation |
| WO2017175232A1 (en) * | 2016-04-07 | 2017-10-12 | M.S.T. Medical Surgery Technologies Ltd. | Vocally activated surgical control system |
| CN107062074A (en) * | 2017-04-12 | 2017-08-18 | 成都迅德科技有限公司 | A kind of adjustable focal length operating lamp with microphone |
| CN108549237A (en) * | 2018-05-16 | 2018-09-18 | 华南理工大学 | Preview based on depth enhancing study controls humanoid robot gait's planing method |
| CN109089365A (en) * | 2018-10-30 | 2018-12-25 | 成都大学附属医院 | A kind of intelligence adjusts the control system and method for operating lamp |
| CN109636062A (en) * | 2018-12-25 | 2019-04-16 | 湖北工业大学 | A kind of students ' behavior analysis method and system based on big data analysis |
| CN110288661A (en) * | 2019-06-19 | 2019-09-27 | 深圳市睿智龙电子科技有限公司 | Location regulation method, device, computer equipment and the storage medium of operating lamp |
| CN110989576A (en) * | 2019-11-14 | 2020-04-10 | 北京理工大学 | Target following and dynamic obstacle avoidance control method for differential slip steering vehicle |
| CN111631719A (en) * | 2020-05-21 | 2020-09-08 | 北京城市系统工程研究中心 | Method for predicting falling risk of old people |
| CN112435503A (en) * | 2020-10-30 | 2021-03-02 | 江苏大学 | Intelligent automobile active collision avoidance method for identifying intention of high-risk pedestrians |
| CN112389436A (en) * | 2020-11-25 | 2021-02-23 | 中汽院智能网联科技有限公司 | Safety automatic driving track-changing planning method based on improved LSTM neural network |
| CN112509125A (en) * | 2020-12-14 | 2021-03-16 | 广州迦恩科技有限公司 | Three-dimensional reconstruction method based on artificial markers and stereoscopic vision |
| CN112883738A (en) * | 2021-03-23 | 2021-06-01 | 西南交通大学 | Medical entity relation extraction method based on neural network and self-attention mechanism |
| CN113111581A (en) * | 2021-04-09 | 2021-07-13 | 重庆邮电大学 | LSTM trajectory prediction method combining space-time factors and based on graph neural network |
| CN113254197A (en) * | 2021-04-30 | 2021-08-13 | 西安电子科技大学 | Network resource scheduling method and system based on deep reinforcement learning |
| CN115311405A (en) * | 2021-05-06 | 2022-11-08 | 珠海市迪谱医疗科技有限公司 | Three-dimensional reconstruction method of binocular endoscope |
| CN113269963A (en) * | 2021-05-20 | 2021-08-17 | 东南大学 | Internet vehicle signal lamp control intersection economic passing method based on reinforcement learning |
| CN113573284A (en) * | 2021-06-21 | 2021-10-29 | 吉林大学 | Random access backoff method for large-scale machine type communication based on machine learning |
| CN113763723A (en) * | 2021-09-06 | 2021-12-07 | 武汉理工大学 | Traffic signal lamp control system and method based on reinforcement learning and dynamic timing |
| WO2023123906A1 (en) * | 2021-12-31 | 2023-07-06 | 深圳云天励飞技术股份有限公司 | Traffic light control method and related device |
| CN115033022A (en) * | 2022-06-28 | 2022-09-09 | 华南理工大学 | DDPG unmanned aerial vehicle landing method based on expert experience and oriented to mobile platform |
| CN115665941A (en) * | 2022-11-04 | 2023-01-31 | 浙江大学医学院附属第一医院 | Sight following surgical illuminating lamp and sight following method |
| CN116259175A (en) * | 2022-12-30 | 2023-06-13 | 北京邮电大学 | Vehicle speed recommendation method and device for diversified dynamic signal lamp modes |
| CN116513273A (en) * | 2023-01-03 | 2023-08-01 | 南京工业大学 | Train operation scheduling optimization method based on deep reinforcement learning |
Non-Patent Citations (2)
| Title |
|---|
| Mingrui Yu ; Jaijun Chai ; Yisheng Lv ; Gang Xiong.An effective deep reinforcement learning approach for adaptive traffic signal control.2020 Chinese Automation Congress (CAC).2021,正文第6419-6425页. * |
| 一种基于机器学习的自然调光LED节能灯;李峦;陈英革;王小英;;电脑知识与技术(第05期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117062280A (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11442429B2 (en) | Systems and methods for advance anomaly detection in a discrete manufacturing process with a task performed by a human-robot team | |
| US20230117254A1 (en) | Context and state aware treatment room efficiency | |
| US20200226751A1 (en) | Surgical workflow and activity detection based on surgical videos | |
| CN110163485A (en) | A kind of computer room cruising inspection system | |
| CN108879440A (en) | Intelligent examination and repair system and method based on wearable augmented reality display terminal and cloud platform | |
| CN102348101A (en) | Examination room intelligence monitoring system and method thereof | |
| CN111046943A (en) | Method and system for automatically identifying state of isolation switch of transformer substation | |
| CN109120889A (en) | A kind of Good Laboratory Control management system and method | |
| DE102020133445A1 (en) | CAMERA ARCHESTRY TECHNOLOGY TO IMPROVE AUTOMATIC PERSONAL IDENTIFICATION | |
| Diethe et al. | Releasing ehealth analytics into the wild: lessons learnt from the sphere project | |
| CN117062280B (en) | Automatic following system of neurosurgery self-service operating lamp | |
| CN115620888A (en) | Digital twin-based surgical robot monitoring method, device and system | |
| CN116127790A (en) | Predictive maintenance management method and system for industrial robot | |
| Zou et al. | Evaluation of operation state for operators in NPP Main control room using human behavior recognition | |
| CN117111660A (en) | Unattended intelligent granary system and method | |
| CN116690605A (en) | Intelligent robot on duty for fire control center | |
| KR102415912B1 (en) | Artificial Intelligence life monitoring system with personal information protection function | |
| Wang et al. | Guided activity prediction for minimally invasive surgery safety improvement in the internet of medical things | |
| US20240112466A1 (en) | System and method for analyzing periodic task | |
| CN111652911B (en) | Target monitoring method, device and equipment | |
| Lu | REAL-TIME EYE BLINK DETECTION WITH GENERAL CAMERAS | |
| CN116091027A (en) | Electric power equipment maintenance method, device and system | |
| Padoy | Toward an Operating Room Control Tower? | |
| US20240135525A1 (en) | Method and system of providing assistance during an operation performed on an equipment | |
| CN110675950B (en) | Paralyzed patient intelligent nursing system based on Internet of things cloud platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |