Hospital patient intelligent transfer method and system based on image recognition and intelligent terminal
The image recognition-based smart wheelchair intelligently identifies patient information and plans routes, solving the problem of difficulty for patients with mobility impairments to find departments in large hospitals, and improving examination efficiency and doctors' work efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI YANGPU CENT HOSPITAL
- Filing Date
- 2025-12-09
- Publication Date
- 2026-05-01
AI Technical Summary
In large general hospitals, inpatients with limited mobility often have difficulty finding the department they need to examine quickly, leading to low efficiency for examining doctors.
The system employs an image recognition-based smart wheelchair. By acquiring images of the patient's wrist, identifying wristband features and location, and scanning barcodes to obtain patient and location information, it determines the movement path based on a hospital model. The system then uses a transport device to automatically plan the path and intelligently control the wheelchair to move to the department where the examination is to be conducted. It adapts to different groups of people and elevator conditions, optimizing the examination route.
It improves the efficiency of patients finding the examination department, reduces examination time, enhances the work efficiency of examining doctors, and ensures smooth patient transfer in crowds and elevators.
Smart Images
Figure CN121964078A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of smart healthcare, and in particular to a method, system, and smart terminal for intelligent patient transfer in hospitals based on image recognition. Background Technology
[0002] With the development of information technology and my country's medical and health undertakings, the process of medical informatization in my country has also improved. A sound medical system needs to be assisted by an efficient smart information platform. Establishing a smart medical platform system in a timely manner is a challenge that hospitals face in the competition.
[0003] General hospitals are often large, and patients often cannot quickly and accurately locate their target department after entering the hospital, especially for inpatients with mobility issues who often require numerous physical examinations. After the doctor orders the tests, patients' unfamiliarity with the hospital and mobility limitations prevent them from finding the target department in a short time, resulting in lower work efficiency for the doctors in that department. Summary of the Invention
[0004] To enable patients to quickly locate their target department and improve the work efficiency of examining doctors, this invention provides a hospital patient intelligent transfer method, system, and intelligent terminal based on image recognition.
[0005] In a first aspect, the present invention provides a hospital patient intelligent transfer method based on image recognition, employing the following technical solution: A method for intelligent patient transfer in hospitals based on image recognition, comprising: Obtain images of the patient's wrist; Based on the wrist image information, the wristband features and wristband position are matched from a preset wristband feature library; Scan the preset identification barcode at the wristband position to obtain patient information and patient location information; Retrieve the department of the patient's pending examination based on the patient's information; Based on the department to be inspected, retrieve the department location information from the preset hospital model; The movement path is determined from a pre-set hospital model based on the patient's location information and the department's location information; The patient is moved to the department for the examination by a pre-set transport device at a preset speed according to the movement path control.
[0006] Optionally, the transfer device has a manual mode and an automatic mode, and the mode switching method of the transfer device includes: Acquire image information of the crowd along the movement path; Based on crowd image information and a pre-set human feature database, determine personnel characteristics and personnel distribution information; Determine the characteristics and width of the gaps between people based on the population distribution information; When the slit width is not less than the preset transfer device width, the transfer device is switched to automatic mode and a prompt is issued; When the slit width is less than the width of the transfer device, the transfer device is controlled to issue a prompt, and after a preset waiting time, the slit width is checked again to see if it is less than the width of the transfer device. The control of the transfer device is switched to manual mode and a prompt is issued only if the slit width value is less than the width of the transfer device.
[0007] Optionally, the control methods for the transfer device in a moving flow of people include: Acquire two images of pedestrian flow along the moving path within a preset shooting interval; Based on the pedestrian flow image information and personnel distribution information, determine the features of two people in adjacent positions, and determine the slit width value between the two personnel features. Define the personnel feature in front as the person moving forward, and the personnel feature in back as the person following behind. When the slit width is greater than the preset cutoff distance, the walking speed of the following person is determined based on the information from the two pedestrian flow images. The collision time is determined based on the slit width and walking speed. The theoretical extension speed of the retractable baffle preset on the transfer device is determined based on the preset extension length and collision time, and the actual extension speed of the baffle is determined based on the theoretical extension speed and the preset speed increase. The baffle extends according to the actual extension speed and inserts into the narrow gap between the following person and the person in front, and displays a prompt on the display screen preset on the baffle. The baffle is kept stationary, and the transfer device is controlled to move at a speed into the slit feature to retract the baffle.
[0008] Optional transfer methods when the patient's location information and the department's location information are not on the same floor include: Department floor information is determined based on department location information and hospital model; Based on patient location information and hospital model, determine the patient's floor information; When the patient's floor information is inconsistent with the department's floor information, the elevator characteristics and elevator location information closest to the movement path are determined based on the movement path and the hospital model, and the real-time floor dwell information of the elevator is obtained. The elevator arrival time is determined by real-time floor dwell information, patient floor information, and preset elevator speed. The travel distance is determined based on the patient's location information and the elevator's location information, and the arrival time of the person is determined based on the travel distance and the speed of movement. When the arrival time of the personnel is less than the arrival time of the elevator, the elevator is controlled to stop. When the arrival time of the personnel is not less than the arrival time of the elevator, obtain the image information of the elevator interior; The number of people is determined based on images inside the elevator and a human feature database. The elevator will stop and issue a warning only if the number of passengers is less than the preset standard number of passengers.
[0009] Optionally, patients can take the elevator in the following ways: The total station area is determined based on the number of people and the preset benchmark standing area; The allowable area for the transfer device is determined based on the preset elevator area and total station area. When the accommodating area exceeds the preset transfer device area, a prompt will be issued and the transfer device will be controlled to enter the elevator; When the capacity area is not greater than the area of the transfer device, a prompt is issued and the capacity area is checked again to see if it is greater than the area of the transfer device. Take the next elevator based on the elevator's arrival time, only if the area to be accommodated is not greater than the area of the transfer device. When the elevator reaches the department floor information, determine whether there are human characteristics between the transfer device and the elevator door based on the elevator's internal image information and the human feature database. When there are people with distinctive features between the transfer device and the elevator door, a avoidance warning is issued, and the people's features are reconfirmed. Currently, the elevator stops at a preset long dwell time only when there are people present, and the transfer device is switched to manual mode.
[0010] Optional, also includes: Acquire external image information of the elevator; The number of waiting people is determined based on external elevator images and a human feature database. The number of people allowed to enter is determined based on the area of the accommodation and the area of the transfer equipment. The elevator will issue a warning when the number of people waiting exceeds the number of people allowed to enter.
[0011] Optional transfer methods when there are multiple departments requiring examination include: Obtain the personnel queuing information for each department with each item to be inspected; The estimated queuing time is determined based on the queuing information and the preset queuing time per person. Based on the department to be inspected, retrieve the department location information of each department from the preset hospital model; The estimated arrival time is determined based on the patient's location information, the department's location information, and the speed at which the transport device moves. The examination and transfer time between departments is determined based on the location information of each department and the moving speed of the transfer device; Generate optimized transfer route plans based on estimated queuing time, expected arrival time, and inspection and transfer time. The system controls the transport device to register patients with the department where the examination is to be performed, and then controls the transport device to transport patients to the department where the examination is to be performed according to the optimized transport route plan.
[0012] Optional optimized transit route options include: The on-site waiting time for each department to be inspected is determined based on the estimated queuing time and the expected arrival time. When the on-site waiting time is not greater than the preset overtime waiting time, the department to be examined is determined according to the on-site waiting time and defined as the preferred examination department. The transfer device is then controlled to transfer the patient to the preferred examination department. When the on-site waiting time exceeds the preset overtime waiting time, the on-site waiting time of each department to be inspected will be sorted, and the department with the shortest on-site waiting time will be defined as the preferred inspection department. The remaining waiting time for each department is determined based on the transfer time between the preferred examination department and the department for each examination item, and the estimated waiting time. The remaining waiting times for each department to be examined are sorted to obtain the examination order. Based on the examination order, the transport device is controlled to transport the patient to the department to be examined.
[0013] Secondly, this application provides an intelligent patient transport system for hospitals based on image recognition, employing the following technical solution: A hospital patient intelligent transfer system based on image recognition includes: The acquisition module is used to acquire patient wrist image information, patient information, patient location information, crowd image information, crowd flow image information, real-time elevator floor dwell information, elevator interior image information, elevator exterior image information, and personnel queuing information for each department to be examined. A memory for storing the program of any of the above-described image recognition-based intelligent patient transfer methods in hospitals; The processor and the program in the memory can be loaded and executed by the processor to implement any of the above-described image recognition-based intelligent patient transfer methods for hospitals.
[0014] Thirdly, this application provides a smart terminal, which adopts the following technical solution: A smart terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed any of the above-described image recognition-based intelligent patient transfer methods for hospitals.
[0015] In summary, this application includes at least one of the following beneficial technical effects: 1. The system identifies the patient's examination items and the department to be examined by recognizing information on the patient's wristband through the wheelchair. Based on the department location information and the patient's location information, the system forms the optimal movement path. Then, the system controls the wheelchair to carry the patient to the examination along the optimal movement path. Through the above method, the efficiency of the patient's examination is high, and the efficiency of the examining doctor's examination is also high. 2. The wheelchair can automatically identify crowds blocking its path, find a wide gap to pass through in stationary crowds, and create space for the wheelchair to pass through in moving crowds by using barriers. This makes it less likely for patients to be blocked by crowds in hospitals and delay their examination time, thereby improving the efficiency of patients' examinations. 3. The wheelchair can interact with the elevator, controlling the elevator's stop based on the number of people inside and outside the elevator, and creating an opportunity for the patient to enter the elevator and go to the department where the examination is to be performed as soon as possible, thereby improving the efficiency of the patient's examination. Attached Figure Description
[0016] Figure 1 This is a flowchart of a method for intelligent patient transfer in hospitals based on image recognition, according to an embodiment of the invention. Figure 2 This is a flowchart of the mode switching method of the transfer device according to an embodiment of the invention; Figure 3 This is a flowchart illustrating the method of controlling the transfer device in a moving flow of people according to an embodiment of the invention. Figure 4 This is a flowchart of a transfer method when the patient's location information and the department's location information are not on the same floor, according to an embodiment of the invention. Figure 5 This is a flowchart illustrating a method for a patient to ride an elevator, according to an embodiment of the invention. Detailed Implementation
[0017] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.
[0018] This application discloses an intelligent patient transport method for hospitals based on image recognition. In this solution, an intelligent transport device is used for transportation. This device can be a smart hospital bed or a smart wheelchair. In this embodiment, a smart wheelchair is used to transport the patient for various examinations. Based on the patient's required examinations and the relevant department, the smart wheelchair can automatically plan the route to the target department, automatically arrange the order of examinations, and automatically register the patient, enabling the patient to complete the examinations quickly.
[0019] Reference Figure 1A method for intelligent patient transfer in hospitals based on image recognition includes the following steps: Step S100: Obtain image information of the patient's wrist.
[0020] The smart wheelchair is equipped with a camera, and the wrist image information refers to the image of the patient's wrist captured by the camera. The patient wears a wristband that records patient information, and the wristband is identified by acquiring the wrist image information.
[0021] Step S101: Match wristband features and wristband position from a preset wristband feature library based on wrist image information.
[0022] The wristband feature library refers to a database containing various medical wristband images obtained through machine learning, which will not be elaborated on here. By matching wrist image information with the wristband feature library, the wristband features on the patient's wrist can be obtained, and the location of the wristband features on the patient's wrist can be determined.
[0023] Step S102: Scan the preset identification barcode at the wristband position to obtain patient information and patient location information.
[0024] The wristband is an electronic device that records patient information and displays it on the wristband's screen as a QR code. When the patient is seated in the smart wheelchair, the wheelchair's camera can identify the QR code to obtain the patient's information, including the patient's basic information and the tests the doctor has ordered for the patient's next steps.
[0025] The positioning chip in the wristband can locate the patient's position in the hospital and store the location information in the identification barcode on the wristband. When the camera identifies the identification barcode, it can also obtain the patient's current location information.
[0026] Step S103: Retrieve the department of the patient's examination items based on the patient information.
[0027] The department for the examination items is the department corresponding to the examination order issued by the doctor. After obtaining the patient's information, the smart wheelchair can retrieve the department for the examination items that the patient needs to undergo from the system.
[0028] Step S104: Retrieve the department location information from the preset hospital model according to the department of the item to be inspected.
[0029] The pre-set hospital model is a database of hospital models created by technicians based on the actual structure of the hospital. The hospital model includes the various departments of the hospital and their locations, all the corridor paths of the hospital, the distribution of elevators, etc.
[0030] Department location information refers to the location of the department to be examined within the hospital, which is retrieved from the hospital model.
[0031] Step S105: Determine the movement path from the preset hospital model based on the patient's location information and the department's location information.
[0032] Based on the patient's location information, their position within the hospital model can be determined. Then, the movement path between the patient's location and the department's location can be determined from the hospital model. The movement path refers to the shortest distance from the patient's location to the department's location.
[0033] Step S106: Based on the movement path, control the preset transport device to move the patient to the department of the examination item at a preset speed.
[0034] The preset moving speed is the speed set by the technicians when the smart wheelchair carries the patient. At this speed, the wheelchair moves more stably and is less likely to cause harm to the patient, which will not be elaborated here.
[0035] The system determines the movement path from the patient's location information to the department's location information, and the smart wheelchair moves the patient to the department where the examination is to be performed based on the movement path prompts in the system.
[0036] The smart wheelchair also has a display screen that shows a QR code containing the wheelchair's current location. As the wheelchair moves along its path, cameras along the path can recognize the QR code on the wheelchair, locate the patient and the wheelchair in real time, and report the location back to the system to correct the wheelchair's movement path.
[0037] Reference Figure 2 The transport device has manual and automatic modes. Manual mode allows the patient to control the wheelchair independently, while automatic mode is controlled by the system, requiring no patient intervention. The mode switching method for the transport device includes the following steps: Step S200: Obtain image information of the crowd along the movement path.
[0038] Crowd image information refers to images of people blocking the path of a wheelchair as it moves with a patient, captured by a camera mounted on the wheelchair. Here, "crowd" refers to people standing still. As the wheelchair moves with the patient, it takes pictures of the path ahead, thus obtaining crowd image information.
[0039] Step S201: Determine the characteristics and distribution information of the people based on the crowd image information and the preset human feature database.
[0040] A human feature database refers to a database containing human images obtained through machine learning, which will not be elaborated upon here. By matching crowd image information with the human feature database, it is possible to determine all the features of individuals in the crowd image information and the positional distribution of each individual feature within the crowd image information. Personnel distribution information refers to the distribution of individual features along a movement path.
[0041] Step S202: Determine the slit characteristics and slit width values between people based on the personnel distribution information.
[0042] When people are not in close proximity, gaps exist between them. All gap features can be identified and analyzed using crowd image information and personnel distribution information. Based on a reference point placed on the wheelchair and its position relative to the camera, combined with the reference point's position in the crowd image information and the width of the gap features within the crowd image information, the actual gap width can be determined.
[0043] Step S2021: When the slit width value is not less than the preset transfer device width, control the transfer device to switch to automatic mode and issue a prompt.
[0044] The preset width of the transfer device is the standard width of a smart wheelchair, which will not be elaborated here.
[0045] When the slit width is not less than the width of the transfer device, it means the wheelchair can pass directly through the crowd at the slit. Therefore, switch the wheelchair to automatic mode, and the wheelchair can automatically carry the patient through the crowd. When the wheelchair passes through the slit, it will issue a warning to alert the crowd to give way.
[0046] Step S2022: When the slit width value is less than the width of the transfer device, control the transfer device to issue a prompt, and after a preset waiting time, determine again whether the slit width value is less than the width of the transfer device.
[0047] When the slit width is less than the width of the transfer device, the wheelchair cannot pass through the slit directly. At this time, the wheelchair will issue a warning to the people in front to make way for the patient to pass through.
[0048] The preset waiting time is the time set by the technicians to allow personnel to react and make way, which will not be elaborated here.
[0049] After the waiting period, the system will re-detect the crowds and gaps in the movement path to determine whether the wheelchair can pass through.
[0050] Step S20221: If and only if the slit width value is less than the width of the transfer device, control the transfer device to switch to manual mode and issue a prompt.
[0051] If no one moves aside and the gap width is insufficient for the wheelchair to pass, the wheelchair will switch to manual mode. The wheelchair will then be unable to pass through the gap automatically, requiring the patient to manually control it through the crowd. During the passage, the wheelchair will also issue warnings to avoid collisions with the patient.
[0052] Reference Figure 3 The wheelchair's path may encounter numerous pedestrians, and the system can adjust its control based on the crowd when the wheelchair passes through them. The control method for the transfer device in moving crowds includes the following steps: Step S300: Acquire two images of pedestrian flow along the moving path within a preset shooting interval.
[0053] The pedestrian flow image information refers to images of people on the path of the wheelchair, captured by a camera mounted on the wheelchair. Here, the pedestrian flow image information focuses on people who are walking.
[0054] When the wheelchair moves in front of a group of people, the camera on the wheelchair will take pictures.
[0055] The preset shooting interval refers to the time interval set by the technician for the camera to continuously capture multiple images, which will not be elaborated here.
[0056] Step S301: Based on the pedestrian flow image information and the personnel distribution information, determine the features of two people in adjacent positions, and determine the slit width value between the two personnel features. Define the personnel feature in front as the person moving forward, and the personnel feature in the back as the person following behind.
[0057] The system can identify the two people walking in front and behind with the largest gaps based on pedestrian flow image information and personnel distribution information, and determine the gap width value. The method for identifying personnel features is as follows: first, all gap features are identified through image recognition; then, the gap features are compared in size to select the largest gap feature; finally, the person walking in front and the person following behind are determined based on the largest gap feature. The method for determining the gap width value is the same as in step S202 and will not be repeated here. The gap width value here refers to the gap width when the camera on the wheelchair is taking the picture.
[0058] Step S302: When the slit width value is greater than the preset cutoff distance, determine the walking speed of the following person based on the information from the two pedestrian flow images.
[0059] In this embodiment, a retractable baffle is provided on the wheelchair. When the wheelchair moves in front of a person walking, the baffle can extend from the wheelchair and insert into the narrow gap of the pedestrian flow, thereby creating a gap for the wheelchair to move.
[0060] The cutoff distance is a reference value set by technicians based on the normal walking speed of people, which is the distance at which a following person is unlikely to collide with the baffle when the baffle is inserted into the slit. It will not be elaborated here.
[0061] When the slit width is not greater than the cutoff distance, the wheelchair will not insert the baffle into the slit feature to avoid the person being unable to react in time and causing an impact. In this case, the wheelchair will wait for a period of time and reselect the slit feature.
[0062] When the slit width is greater than the cutoff distance, the wheelchair will insert the baffle into the slit feature, at the end furthest from the following person. The baffle's parameters are determined based on the following person's walking speed. The following person's walking speed is determined by: using the pedestrian flow images acquired from two separate shots to determine the distance the following person moves within the shooting interval, and then calculating the walking speed from the distance moved and the shooting interval.
[0063] Step S303: Determine the collision time based on the slit width value and walking speed.
[0064] The collision time refers to the time, calculated theoretically, for a person walking at a walking speed to collide with the baffle when it is inserted into the slit feature. The collision time is the quotient of the slit width and the walking speed.
[0065] Step S304: Determine the theoretical extension speed of the retractable baffle preset on the transfer device based on the preset extension length and collision time, and determine the actual extension speed of the baffle based on the theoretical extension speed and the preset speed increase.
[0066] The theoretical extension speed is the speed at which the barrier will inevitably collide with the following person if they do not yield. The preset extension length is a fixed length that the barrier can extend to, as set by the technicians, and will not be elaborated here. The theoretical extension speed is the quotient of the extension length and the collision time. In order to allow the following person time to react and prevent them from colliding with the barrier, the actual extension speed of the barrier must be greater than the theoretical extension speed.
[0067] The preset speed increase is the speed that the technician adds based on the theoretical extension speed of the baffle to improve the actual extension speed of the baffle, which will not be elaborated here.
[0068] The actual extension speed of the baffle is the sum of the theoretical extension speed and the speed increase.
[0069] Step S305: Control the baffle to extend and insert into the narrow gap between the following person and the person in front according to the actual extension speed, and display the prompt content on the display screen preset on the baffle.
[0070] When a wheelchair is in front of a moving crowd, the wheelchair uses a camera to determine the walking speed of the people and the gap between them, and determines the actual extension speed of the barrier. Then, the wheelchair controls the barrier to insert into the gap at the actual extension speed to block the following people, allowing them to avoid the gap. At this time, the wheelchair has enough space to move forward.
[0071] The baffle is equipped with a display screen. When the baffle is inserted into the slit feature, the display screen will show a prompt and provide an audio prompt to remind people to avoid it.
[0072] Step S306: Control the baffle to remain stationary, and control the transfer device to enter the slit feature at a moving speed to retract the baffle.
[0073] Once the baffle is inserted into the slit feature, suction cups will extend from the bottom of the baffle and adhere to the ground, allowing the baffle to remain stationary. Then, the wheelchair will enter the slit feature at a preset speed and simultaneously retract the baffle.
[0074] Reference Figure 4 When the patient's location information and the department's location information are not on the same floor, the transfer method includes the following steps: Step S400: Determine the department floor information based on the department location information and the hospital model.
[0075] The department / floor information refers to the floor where the department to be inspected is located. The hospital model also divides the hospital into floors based on its actual structure, with each department corresponding to a specific floor. The floor where the department to be inspected is located can be determined based on the department's location information.
[0076] Step S401: Determine the patient's floor information based on the patient's location information and the hospital model.
[0077] Patient floor information refers to the floor a patient is currently on, which can be determined based on the patient's current location information in the hospital model.
[0078] Step S402: When the patient's floor information is inconsistent with the department's floor information, determine the elevator features and elevator location information that are closest to the movement path based on the movement path and the hospital model, and obtain the elevator's real-time floor dwell information.
[0079] If the patient's floor information matches the department's floor information, the wheelchair can carry the patient directly through the hospital corridor to the department where the examination is to be performed.
[0080] If the patient's floor information does not match the department's floor information, it means that the patient is not on the same floor as the department where the examination is to be performed. In this case, the patient needs to take the elevator to the floor where the department where the examination is to be performed is located.
[0081] The wheelchair system can retrieve all elevator features from the system based on the hospital model and select the elevator feature closest to the patient's travel path as the elevator the patient needs to take, thereby reducing the time the patient spends traveling to the department where the examination is to be performed. Based on the selected elevator feature and the hospital model, the elevator's location information can be determined.
[0082] Real-time elevator floor information refers to the current floor the elevator is on. Elevators are part of the hospital's system, and wheelchair users can directly access this information from the system.
[0083] Step S403: Determine the elevator arrival time based on real-time elevator floor dwell information, patient floor information, and preset elevator speed.
[0084] The preset elevator speed is the standard lifting speed set by the technicians, which will not be elaborated here.
[0085] Elevator arrival time refers to the time required for the elevator to travel from its current floor to the patient's floor. Based on real-time elevator floor information and the patient's floor information, the number of floors the elevator needs to ascend or descend can be determined. Then, based on the number of floors and the elevator speed, the arrival time can be calculated. Step S404: Determine the travel distance based on the patient's location information and the elevator's location information, and determine the arrival time of the person based on the travel distance and the moving speed.
[0086] The travel distance refers to the distance between the patient's current location and the elevator's location. The path distance between the two can be determined from the hospital model based on the patient's location information and the elevator's location information.
[0087] Personnel arrival time refers to the time required for a patient to travel from their current location to the location of the selected elevator. The arrival time can be determined based on the travel distance and the speed at which the wheelchair moves.
[0088] Step S4041: When the arrival time of the personnel is less than the arrival time of the elevator, control the elevator to stop.
[0089] If the arrival time of the person is less than the arrival time of the elevator, it means that when the patient arrives at the selected elevator location, the elevator has not yet reached the patient's floor. At this time, the wheelchair can interact with the elevator, and the wheelchair can control the elevator to stop at the patient's floor through the system, and output the floor information of the department to be examined to the elevator so that the elevator can stop at the floor of the department to be examined.
[0090] Step S4042: When the arrival time of the personnel is not less than the arrival time of the elevator, obtain the image information inside the elevator.
[0091] If the arrival time of the person is not less than the elevator arrival time, it means that when the patient arrives at the elevator location, the elevator may have already left the patient's current floor. In order to ensure that the patient does not miss the elevator, it is necessary to determine the number of people in the elevator and whether the patient has space to enter the elevator.
[0092] Elevator interior image information refers to images inside the elevator obtained through cameras installed inside the elevator, which contain the characteristics of the people inside the elevator.
[0093] Step S40421: Determine the number of people based on the elevator interior image information and the human feature database.
[0094] The number of people refers to the number of people inside the elevator. This number can be determined through image recognition based on images of the elevator interior and a human feature database. The method for recognizing the number of heads using image recognition is existing technology and will not be elaborated upon here.
[0095] Step S40422: If and only if the number of people is less than the preset standard number of passengers, control the elevator to stop and issue a prompt.
[0096] The preset standard number of passengers is a reference value set by technicians. When the number of people in the elevator is less than the standard number of passengers, the elevator can also accommodate wheelchairs, which will not be elaborated here.
[0097] If the number of people in the elevator is not less than the standard number of passengers, even if the elevator stops at the patient's floor, the patient cannot enter the elevator in a wheelchair, so the wheelchair will not interfere with the normal operation of the elevator.
[0098] If the number of people in the elevator is less than the standard passenger capacity, the patient and wheelchair may still be able to use the elevator. In this case, even if the patient has not yet reached the elevator, the wheelchair can be controlled by the system to stop the elevator. Furthermore, when the elevator stops at the patient's floor, it will issue a notification informing passengers that a person with special needs is in the elevator and requesting them to wait patiently.
[0099] Reference Figure 5 The procedure for patients to take the elevator includes the following steps: This embodiment addresses the situation where the arrival time of the person is less than the arrival time of the elevator. At this point, the patient is already waiting at the elevator door, and the number of people inside the elevator is uncertain.
[0100] Step S500: Determine the total station area based on the number of people and the preset benchmark standing area.
[0101] The baseline standing area refers to the ground area required for a person to stand, as determined based on the general population, and will not be elaborated upon here.
[0102] The total standing area refers to the floor area inside the elevator occupied by all people inside. The total standing area is the product of the number of people and the baseline standing area.
[0103] Step S501: Determine the allowable area for the transfer device based on the preset elevator area and total station area.
[0104] The elevator area is a standard value and will not be elaborated here.
[0105] The capacity area refers to the area inside the elevator after subtracting the area occupied by all standing people. This area can be used for wheelchair placement. The capacity area is the difference between the elevator area and the total station area.
[0106] Step S5011: When the accommodating area is greater than the preset transfer device area, issue a prompt and control the transfer device to enter the elevator.
[0107] When the elevator's capacity is greater than the area of the transfer device, there is still space inside the elevator for a wheelchair. When the elevator door opens, the elevator will issue a warning to inform the people inside to make way for the patient and wheelchair.
[0108] Step S5012: When the accommodating area is not greater than the area of the transfer device, issue a prompt and determine again whether the accommodating area is greater than the area of the transfer device.
[0109] When the elevator's capacity is no larger than the transfer device's area, the elevator will issue a prompt after opening, informing passengers whether someone can leave the elevator to allow the wheelchair and patient to enter. After the prompt, the camera will again detect the number of people in the elevator and determine the capacity.
[0110] Step S50121: Take the next elevator only if and only if the area to be accommodated is not greater than the area of the transfer device.
[0111] If the available space is larger than the transfer device's area, the wheelchair will be guided into the elevator, and a warning will be issued to avoid it. If the available space is still smaller than the transfer device's area, the wheelchair will wait at the elevator door for the next elevator to stop.
[0112] Step S50122: When the elevator reaches the department floor information, determine whether there are human characteristics between the transfer device and the elevator door based on the elevator internal image information and the human feature database.
[0113] When a patient arrives at the floor where the department to be examined is located, there may be people who do not get off the elevator between the wheelchair and the elevator door. These people will block the path for the wheelchair to exit the elevator, so it is necessary to check whether there are any people between the wheelchair and the elevator door.
[0114] Step S50123: When there are personnel characteristics between the transfer device and the elevator door, issue an avoidance warning and reconfirm the personnel characteristics.
[0115] If there are no people obstructing the way, the wheelchair can exit the elevator and proceed to the department to be inspected, following the designated movement path.
[0116] If human characteristics are present in both the elevator and the wheelchair, both will issue a warning to prompt the person to move aside. After issuing the warning, the system will check again to see if human characteristics are still present.
[0117] Step S50124: Currently, the elevator stops for a preset long dwell time only when there are personnel characteristics, and the transfer device is switched to manual mode.
[0118] The preset long dwell time is the time set by technicians for the elevator to wait for the patient to exit. The long dwell time is longer than the normal dwell time of the elevator, which will not be elaborated here.
[0119] If people have already moved aside, the wheelchair can automatically exit the elevator.
[0120] If there are still human characteristics between the two, the system controls the elevator to stay based on the long dwell time, and at this time the wheelchair will automatically switch to manual mode, and the patient can control the wheelchair to move slowly out of the elevator. The long dwell time is sufficient for the patient to control the wheelchair to exit the elevator.
[0121] When there are many people outside the elevator who need to enter, the patient should take the elevator using the following steps: Step S600: Obtain external image information of the elevator.
[0122] External elevator image information refers to images captured by cameras installed on the outside of the elevator, showing people waiting outside the elevator.
[0123] Step S601: Determine the number of waiting people based on the external image information of the elevator and the human feature database.
[0124] The number of waiting people refers to the number of people waiting outside the elevator. The method for determining this number is the same as in step S40421, and will not be repeated here.
[0125] Step S602: Determine the number of personnel who can enter based on the capacity area and the area of the transfer device.
[0126] If there are already people in the elevator, the remaining space can accommodate wheelchairs and some other people; the number of these additional people determines the total number of people allowed in. The difference between the existing space and the transfer device area determines the remaining standing space, and the difference between this remaining standing space and the baseline standing space determines the total number of people allowed in.
[0127] Step S603: When the number of waiting people exceeds the number of people who can enter, the elevator issues a warning.
[0128] If the number of people waiting is no greater than the number of people who can enter, then when the elevator arrives, all people waiting outside the elevator, as well as patients in wheelchairs, can enter the elevator.
[0129] If the number of people waiting exceeds the number of people allowed to enter, some of those waiting outside the elevator will be unable to enter when a patient needs to. To prevent those outside from occupying all the elevator's interior space and to allow patients and wheelchairs to enter, the elevator will issue a notification to those outside, informing them to let the patient and wheelchair enter first. The method to ensure patients don't miss the elevator is as follows: Step S6031: Determine whether a wheelchair exists inside the elevator based on the elevator interior image information and the preset wheelchair feature database.
[0130] The wheelchair feature database is a collection of wheelchair images obtained through machine learning. The system matches the elevator interior images with the wheelchair feature database to determine whether a wheelchair is inside the elevator, and thus decides whether a patient waiting outside the elevator should enter.
[0131] Step S6032: If and only if there is no wheelchair in the elevator, issue a prompt and control the elevator to stay for a long time.
[0132] If a patient in a wheelchair is unable to enter the elevator, both the elevator and the wheelchair will issue a yielding warning, prompting people inside the elevator to make way and controlling the elevator to remain stationary for an extended period until the patient enters.
[0133] When there are multiple departments requiring examination, the transfer process includes the following steps: Step S700: Obtain the personnel queuing information for each department of the item to be inspected.
[0134] The personnel queuing information refers to the queuing status of individuals waiting for their examinations within the respective departments. The queuing information for each department can be retrieved directly from the system. Here, "personnel queuing information" refers to the number of people in the queue.
[0135] Step S701: Determine the estimated queuing time based on the queuing information and the preset queuing time per person.
[0136] The queuing time per person is a reference value for the estimated testing time for each test item, and will not be elaborated here.
[0137] Estimated waiting time refers to the time a patient needs to wait from registration at the department where the test is to begin. Estimated waiting time is the product of the number of people queuing and the individual waiting time.
[0138] Step S702: Retrieve the department location information of each department to be inspected from the preset hospital model according to the department to be inspected.
[0139] The same as step S104, so it will not be repeated here.
[0140] Step S703: Determine the estimated arrival time based on the patient's location information, the department's location information, and the moving speed of the transport device.
[0141] Estimated arrival time refers to the time required for the patient to travel from their current location to the location of each department performing the required examinations. The distance between the patient's location and the department's location can be determined, and the estimated arrival time can be calculated based on this distance and the speed of the transport device.
[0142] Step S704: Determine the examination and transfer time between departments based on the location information of each department and the moving speed of the transfer device.
[0143] The examination transfer time refers to the time required for a patient to travel by wheelchair from one department to another for an examination. The calculation method for the examination transfer time is the same as in step S703, and will not be repeated here.
[0144] Step S705: Generate an optimized transfer route plan based on the estimated queuing time, estimated arrival time, and transfer time check.
[0145] The optimized transport route plan refers to the shortest and most efficient route that allows the system to transport patients in wheelchairs to the various departments where examinations are to be performed. The wheelchair moves according to this route plan, enabling patients to complete their examinations in a shorter time. The optimized transport route plan is generated based on the estimated queuing time, expected arrival time, and examination transport time. The generation method will not be detailed here but will be described in detail in subsequent embodiments.
[0146] Step S706: Control the transport device to register the patient with the department where the examination is to be performed, and control the transport device to transport the patient to the department where the examination is to be performed according to the optimized transport route plan.
[0147] After the system identifies all the departments requiring examinations, it will pre-register the patient for each department, allowing for quicker examinations upon arrival. Once all departments are identified, the wheelchair will move according to an optimized transport route, leading the patient to each department for their respective examinations.
[0148] The optimized transit route plan includes the following steps: Step S800: Determine the on-site waiting time for each department to be inspected based on the estimated queuing time and expected arrival time.
[0149] On-site waiting time refers to the time a patient needs to wait from arriving at the department where the examination is to begin. On-site waiting time is the difference between the estimated waiting time and the expected arrival time.
[0150] Step S8001: When the on-site waiting time is not greater than the preset overtime waiting time, determine the department of the examination item to be matched according to the on-site waiting time and define it as the preferred examination department, and control the transfer device to transfer the patient to the preferred examination department.
[0151] The overtime waiting time refers to the time given to patients to prepare before their turn for examination. The standard value set by the hospital system will not be elaborated here. If the on-site waiting time is no greater than the overtime waiting time, it means that when the patient arrives at the department for the examination at the speed of a wheelchair, the examination can be performed immediately.
[0152] If the on-site waiting time in a certain examination department is no greater than the overtime waiting time, then the department with the examination to be performed needs to be prioritized to avoid missing the appointment. Therefore, the department with the examination to be performed is designated as the first-choice examination department, and the system will transfer the patient to the first-choice examination department first.
[0153] Step S8002: When the on-site waiting time is greater than the preset overtime waiting time, sort the on-site waiting times of each department to be inspected, and define the department with the shortest on-site waiting time as the preferred inspection department.
[0154] If the on-site waiting time in all departments is greater than the overtime waiting time, it means that the patient will need to continue waiting regardless of which department the patient arrives at first. In this case, the system will sort the departments according to the overtime waiting time and transfer the patient to the department with the shortest overtime waiting time.
[0155] Step S80021: Determine the remaining waiting time for each department based on the transfer time between the preferred examination department and the department for each examination item, and the estimated queuing time.
[0156] The remaining waiting time refers to the time a patient needs to wait from the department where they completed one examination to the department where they will have their remaining examinations. The remaining waiting time is the difference between the examination transfer time and the estimated waiting time.
[0157] Step S80022: Sort the remaining waiting times of each department for each remaining examination item to obtain the examination order, and control the transfer device to transfer the patient to the department for the examination item according to the examination order.
[0158] At this point, the remaining waiting time in each department will be used as the basis for queuing for examinations, and each subsequent examination will be carried out according to the queue order.
[0159] Based on the same inventive concept, embodiments of the present invention provide a hospital patient intelligent transfer system based on image recognition, comprising: The acquisition module is used to acquire patient wrist image information, patient information, patient location information, crowd image information, crowd flow image information, real-time elevator floor dwell information, elevator interior image information, elevator exterior image information, and personnel queuing information for each department to be examined. A memory for storing programs for any image recognition-based intelligent patient transport method in hospitals; The processor and memory can load and execute programs to implement any image recognition-based intelligent patient transfer method in hospitals.
[0160] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0161] Based on the same inventive concept, embodiments of the present invention provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to perform a hospital patient intelligent transfer method based on image recognition.
[0162] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
Claims
1. A method for intelligent patient transfer in hospitals based on image recognition, characterized in that, include: Obtain images of the patient's wrist; Based on the wrist image information, the wristband features and wristband position are matched from a preset wristband feature library; Scan the preset identification barcode at the wristband position to obtain patient information and patient location information; Retrieve the department of the patient's pending examination based on the patient's information; Based on the department to be inspected, retrieve the department location information from the preset hospital model; The movement path is determined from a pre-set hospital model based on the patient's location information and the department's location information; The patient is moved to the department for the examination by a pre-set transport device at a preset speed according to the movement path control.
2. The intelligent hospital patient transfer method based on image recognition according to claim 1, characterized in that, The transfer device has manual and automatic modes, and the mode switching methods of the transfer device include: Acquire image information of the crowd along the movement path; Based on crowd image information and a pre-set human feature database, determine personnel characteristics and personnel distribution information; Determine the characteristics and width of the gaps between people based on the population distribution information; When the slit width is not less than the preset transfer device width, the transfer device is switched to automatic mode and a prompt is issued; When the slit width is less than the width of the transfer device, the transfer device is controlled to issue a prompt, and after a preset waiting time, the slit width is checked again to see if it is less than the width of the transfer device. The control of the transfer device is switched to manual mode and a prompt is issued only if the slit width value is less than the width of the transfer device.
3. The intelligent hospital patient transfer method based on image recognition according to claim 2, characterized in that, The control methods for transfer devices in moving crowds include: Acquire two images of pedestrian flow along the moving path within a preset shooting interval; Based on the pedestrian flow image information and personnel distribution information, determine the features of two people in adjacent positions, and determine the slit width value between the two personnel features. Define the personnel feature in front as the person moving forward, and the personnel feature in back as the person following behind. When the slit width is greater than the preset cutoff distance, the walking speed of the following person is determined based on the information from the two pedestrian flow images. The collision time is determined based on the slit width and walking speed. The theoretical extension speed of the retractable baffle preset on the transfer device is determined based on the preset extension length and collision time, and the actual extension speed of the baffle is determined based on the theoretical extension speed and the preset speed increase. The baffle extends according to the actual extension speed and inserts into the narrow gap between the following person and the person in front, and displays a prompt on the display screen preset on the baffle. The baffle is kept stationary, and the transfer device is controlled to move at a speed into the slit feature to retract the baffle.
4. The intelligent hospital patient transfer method based on image recognition according to claim 2, characterized in that, When the patient's location information and the department's location information are not on the same floor, the transfer methods include: Department floor information is determined based on department location information and hospital model; Based on patient location information and hospital model, determine the patient's floor information; When the patient's floor information is inconsistent with the department's floor information, the elevator characteristics and elevator location information closest to the movement path are determined based on the movement path and the hospital model, and the real-time floor dwell information of the elevator is obtained. The elevator arrival time is determined by real-time floor dwell information, patient floor information, and preset elevator speed. The travel distance is determined based on the patient's location information and the elevator's location information, and the arrival time of the person is determined based on the travel distance and the speed of movement. When the arrival time of the personnel is less than the arrival time of the elevator, the elevator is controlled to stop. When the arrival time of the personnel is not less than the arrival time of the elevator, obtain the image information of the elevator interior; The number of people is determined based on images inside the elevator and a human feature database. The elevator will stop and issue a warning only if the number of passengers is less than the preset standard number of passengers.
5. The intelligent hospital patient transfer method based on image recognition according to claim 4, characterized in that, Methods for patients to take the elevator include: The total station area is determined based on the number of people and the preset benchmark standing area; The allowable area for the transfer device is determined based on the preset elevator area and total station area. When the accommodating area exceeds the preset transfer device area, a prompt will be issued and the transfer device will be controlled to enter the elevator; When the capacity area is not greater than the area of the transfer device, a prompt is issued and the capacity area is checked again to see if it is greater than the area of the transfer device. Take the next elevator based on the elevator's arrival time, only if the area to be accommodated is not greater than the area of the transfer device. When the elevator reaches the department floor information, determine whether there are human characteristics between the transfer device and the elevator door based on the elevator's internal image information and the human feature database. When there are people with distinctive features between the transfer device and the elevator door, a avoidance warning is issued, and the people's features are reconfirmed. Currently, the elevator stops at a preset long dwell time only when there are people present, and the transfer device is switched to manual mode.
6. The intelligent hospital patient transfer method based on image recognition according to claim 5, characterized in that, Also includes: Acquire external image information of the elevator; The number of waiting people is determined based on external elevator images and a human feature database. The number of people allowed to enter is determined based on the area of the accommodation and the area of the transfer equipment. The elevator will issue a warning when the number of people waiting exceeds the number of people allowed to enter.
7. The intelligent hospital patient transfer method based on image recognition according to claim 1, characterized in that, When there are multiple departments requiring examinations, the transfer methods include: Obtain the personnel queuing information for each department with each item to be inspected; The estimated queuing time is determined based on the queuing information and the preset queuing time per person. Based on the department to be inspected, retrieve the department location information of each department from the preset hospital model; The estimated arrival time is determined based on the patient's location information, the department's location information, and the speed at which the transport device moves. The examination and transfer time between departments is determined based on the location information of each department and the moving speed of the transfer device; Generate optimized transfer route plans based on estimated queuing time, expected arrival time, and inspection and transfer time. The system controls the transport device to register patients with the department where the examination is to be performed, and then controls the transport device to transport patients to the department where the examination is to be performed according to the optimized transport route plan.
8. The intelligent hospital patient transfer method based on image recognition according to claim 7, characterized in that, The optimized transit route plan includes: The on-site waiting time for each department to be inspected is determined based on the estimated queuing time and the expected arrival time. When the on-site waiting time is not greater than the preset overtime waiting time, the department to be examined is determined according to the on-site waiting time and defined as the preferred examination department. The transfer device is then controlled to transfer the patient to the preferred examination department. When the on-site waiting time exceeds the preset overtime waiting time, the on-site waiting time of each department to be inspected will be sorted, and the department with the shortest on-site waiting time will be defined as the preferred inspection department. The remaining waiting time for each department is determined based on the transfer time between the preferred examination department and the department for each examination item, and the estimated waiting time. The remaining waiting times for each department to be examined are sorted to obtain the examination order. Based on the examination order, the transport device is controlled to transport the patient to the department to be examined.
9. A hospital patient intelligent transfer system based on image recognition, characterized in that, include: The acquisition module is used to acquire patient wrist image information, patient information, patient location information, crowd image information, crowd flow image information, real-time elevator floor dwell information, elevator interior image information, elevator exterior image information, and personnel queuing information for each department to be examined. A memory for storing the program of the image recognition-based intelligent patient transfer method for hospitals as described in any one of claims 1 to 8; The processor and the program in the memory can be loaded and executed by the processor to implement the image recognition-based intelligent patient transfer method for hospitals as described in any one of claims 1 to 8.
10. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 8, which is a hospital patient intelligent transfer method based on image recognition.