Medical interactive processing method and system based on natural language processing

Abstract

The present invention provides a medical interaction processing method and system based on natural language processing. The method includes: when the medical end performs a dispensing operation, the dispensing collection is performed on the drug operation slot to obtain the dispensing information and the slot number; when the medical end performs a picking operation, the picking upload information uploaded by the medical end is obtained, and the picking collection is performed on the drug operation slot to obtain the picking collection information; based on the information differentiation of the picking upload information and the picking collection information, the picking difference value is determined; the use upload information uploaded by the medical end is obtained, and the drug recovery slot corresponding to the drug operation slot is determined in the drug residual material table to perform residual material collection to obtain residual material collection information; based on the residual material collection information and the use upload information, the use difference value is determined; the picking difference value and the use difference value are fused and calculated to obtain the difference evaluation value, and the medical end is evaluated for operation. The present invention at least improves the evaluation accuracy of the operation evaluation of the medical end.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a medical interaction processing method and system based on natural language processing. Background Art

[0002] Medication records during surgery play a vital role in medical records. They record information such as the type and dosage of drugs used during surgery, which is of great significance for the doctor's subsequent treatment plan formulation and the patient's condition follow-up.

[0003] The existing technology is unable to verify the actual medication use of doctors during surgery and to evaluate the doctors' corresponding operations. There is a lack of a method and system for processing medication records based on data processing technology. Summary of the invention

[0004] Based on the above problems, the present invention is proposed to provide a medical interaction processing method and system based on natural language processing to overcome the above problems or at least partially solve the above problems.

[0005] According to one aspect of the present invention, a medical interaction processing method based on natural language processing is provided, comprising the following steps:

[0006] In response to the medical end performing a dispensing operation on any drug operation slot located in the drug operation table, performing dispensing collection on the drug operation slot to obtain dispensing information and a slot number corresponding to the drug operation slot;

[0007] In response to the medical end performing a taking operation on the drug operation slot, obtaining the taking upload information uploaded by the medical end based on the drug operation slot, and performing taking collection on the drug operation slot to obtain the taking collection information corresponding to the drug operation slot;

[0008] Determining a take difference value corresponding to the drug operation slot based on information differentiation between the take upload information and the take collection information;

[0009] In response to obtaining the usage upload information uploaded by the medical terminal based on the drug operation slot, determining a drug recovery slot corresponding to the drug operation slot in the drug residual material table based on the slot number, and performing residual material collection on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot;

[0010] Determine a usage difference value corresponding to the drug operation slot based on the remaining material collection information and the usage upload information;

[0011] The taking difference value and the using difference value are fused and calculated, and the operation of the medical end is evaluated based on the obtained difference evaluation value.

[0012] Optionally, in the method according to the present invention, the drug operation table includes drug operation slots distributed in an array;

[0013] In response to the medical end performing a dispensing operation on any drug operation slot located in the drug operation table, performing dispensing collection on the drug operation slot to obtain dispensing information and a slot number corresponding to the drug operation slot, including:

[0014] Performing output detection on each gravity sensor disposed at the bottom of each drug operation slot in real time, and determining that the output state of any gravity sensor at any time is a valid collection state based on the output detection, the drug operation slot corresponding to the gravity sensor is determined as the dispensing operation slot;

[0015] Determine the time as the dispensing start time when the medical end starts to perform the dispensing operation on the dispensing operation slot, and establish a timed monitoring task corresponding to the gravity sensor based on the dispensing start time, so as to determine the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot based on the timed monitoring task;

[0016] Performing medication collection on the medication dispensing operation slot at any time after the medication dispensing end time to obtain a medication collection image corresponding to the medication dispensing operation slot;

[0017] The slot number corresponding to the medication dispensing operation slot and the actual type and actual dosage of the medication included in the medication dispensing information are determined based on the medication dispensing acquisition image.

[0018] Optionally, in the method according to the present invention, a timed monitoring task corresponding to the gravity sensor is established based on the medication start time, so as to determine the medication end time at which the medical end ends the medication dispensing operation for the medication dispensing operation slot based on the timed monitoring task, including:

[0019] Retrieving a reference monitoring period, and dividing the reference monitoring period into periods based on a half value to obtain a first monitoring period and a second monitoring period;

[0020] In response to the gravity sensor outputting a gravity acquisition value without gravity change in a reference monitoring period after the medication dispensing start time, determining a medication dispensing end time at which the medical terminal ends the medication dispensing operation on the medication dispensing operation slot based on the medication dispensing start time and the reference monitoring period;

[0021] In response to a gravity change in the gravity acquisition value output by the gravity sensor in a first monitoring period after the medication dispensing start time, the time corresponding to the gravity change is re-determined as the medication dispensing start time, and the medication dispensing end time at which the medical end ends the medication dispensing operation on the medication dispensing operation slot is determined again based on the timed monitoring task;

[0022] In response to a gravity change in the gravity collection value output by the gravity sensor between the first monitoring period and the second monitoring period, the time corresponding to the gravity change is re-determined as the dispensing start time, and a preset time increase coefficient is retrieved and multiplied by the reference monitoring period to obtain an updated reference monitoring period;

[0023] The scheduled monitoring task is updated based on the updated reference monitoring period, and the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot is determined based on the updated scheduled monitoring task.

[0024] Optionally, in the method according to the present invention, each of the drug operation slots includes spaced-apart accommodating sub-slots having a first size for accommodating medical drugs and numbered sub-slots having a second size for marking the drug operation slots, wherein different numbered sub-slots are filled with different slot numbers, and the first size is larger than the second size;

[0025] Determining the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information based on the dispensing acquisition image includes:

[0026] Obtaining a reference slot pixel value corresponding to the drug operation slot, and binarizing the medication collection image based on the reference slot pixel value to obtain each slot pixel point corresponding to the reference slot pixel value and each noise pixel point corresponding to the noise pixel value;

[0027] Connecting the pixels of each slot based on the pixels between adjacent positions, and determining a first image contour and a second image contour located in the medication collection image based on the pixel connections;

[0028] Determine a first number corresponding to each reference slot pixel point constituting the first image contour, and a second number corresponding to each reference slot pixel point constituting the second image contour, and determine the image contour with the largest corresponding number as the accommodating contour corresponding to the accommodating sub-slot, and determine the image contour with the smallest corresponding number as the numbered contour corresponding to the numbered sub-slot;

[0029] The dispensing acquisition image is segmented to obtain a containing image portion including the containing contour and a numbered image portion including the numbered contour, and based on the containing image portion and the numbered image portion, the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information are determined.

[0030] Optionally, in the method according to the present invention, determining the actual type of medication, the actual dosage of medication and the slot number corresponding to the medication dispensing operation slot based on the containing image portion and the numbering image portion includes:

[0031] Performing image recognition on the numbered image portion based on a pre-trained digital recognition model, and determining the slot number corresponding to the dispensing operation slot based on the recognition result;

[0032] Performing image recognition on the accommodating image portion based on a pre-trained contour recognition model, and determining each accommodating contour located in the accommodating image portion based on the recognition result;

[0033] Determine the size of each containment contour respectively, obtain the containment dimensions corresponding to each containment contour respectively, and determine the containment dimension with the largest corresponding dimension as the initial drug contour;

[0034] Performing distortion correction processing on the initial drug contour, and determining a corresponding drug size based on the obtained current drug contour;

[0035] Retrieving a preset drug comparison table, wherein the preset drug comparison table includes preset contours and preset sizes corresponding to different actual types of dispensed drugs;

[0036] Determine the confidence of the current drug contour and each preset contour corresponding to different actual types of medication based on the shape dimension, and obtain each shape confidence corresponding to different actual types of medication;

[0037] Determine the confidence of the drug size and each preset size corresponding to different actual types of medicines based on the size dimension, and obtain the confidence of each size corresponding to different actual types of medicines;

[0038] The shape confidence and size confidence corresponding to the same actual type of medicine are summed up to obtain various type confidences corresponding to different actual types of medicines, and the type confidence with the largest corresponding value among various type confidences is determined as the target confidence;

[0039] Determining the actual type of medicine dispensed corresponding to the target confidence as the actual type of medicine dispensed corresponding to the medicine dispensing operation slot, and obtaining the weight of the medicine;

[0040] Based on the gravity sensor, the slot gravity value corresponding to the dispensing operation slot at any time after the dispensing end time is obtained, and the slot gravity value is calculated by ratio with the medicine weight to obtain the actual dosage of the dispensing corresponding to the dispensing operation slot.

[0041] Optionally, in the method according to the present invention, in response to the medical end performing a taking operation on the drug operation slot, obtaining the taking upload information uploaded by the medical end based on the drug operation slot, and performing taking collection on the drug operation slot to obtain the taking collection information corresponding to the drug operation slot, including:

[0042] In response to a gravity collection value output by any gravity sensor at any time showing a decreasing trend, the time is determined as a starting time for the medical terminal to perform a taking operation on the drug operation slot, and the drug operation slot corresponding to the gravity sensor is determined as a taking operation slot;

[0043] Based on the picking up start time, the voice broadcast unit is triggered to output a preset voice, and the voice receiving module is triggered to receive voice in the current voice environment;

[0044] In response to the voice receiving module receiving the taking voice information output by the medical terminal at any time, the time is determined as the taking end time of the medical terminal performing the taking operation on the taking operation slot, and the taking upload type and the taking upload dosage corresponding to the taking operation slot are obtained based on the taking voice information;

[0045] Performing a pick-up and acquisition operation on the medication dispensing operation slot at any time after the medication dispensing end time, and obtaining a pick-up and acquisition image corresponding to the medication dispensing operation slot;

[0046] Determine the slot number corresponding to the taking operation slot and the actual taking type and the actual taking dose included in the taking information based on the taking acquisition image;

[0047] The dispensing information corresponding to the taking operation slot is determined based on the slot number, and the taking and collecting information corresponding to the taking operation slot is determined based on the information comparison between the dispensing information and the taking information.

[0048] Optionally, in the method according to the present invention, determining the pick-up difference value corresponding to the drug operation slot based on the information differentiation between the pick-up uploaded information and the pick-up collected information includes:

[0049] In response to the actual type of taking being different from the uploaded type of taking, determining a preset first comparison value as a taking difference value corresponding to the medicine operation slot;

[0050] In response to the actual type of taking being the same as the uploaded type of taking, based on the dosage ratio between the actual dosage taken and the uploaded dosage taken, and based on the dosage ratio, a numerical adjustment is made to the preset second comparison value to obtain a taking difference value corresponding to the drug operation slot.

[0051] Optionally, in the method according to the present invention, the taking difference value and the using difference value are fused and calculated, and the operation of the medical end is evaluated based on the obtained difference evaluation value.

[0052] Retrieving a take difference weight value and a use difference weight value, wherein the use difference weight value is different from the take difference weight value;

[0053] Respectively multiplying the taken difference value and the taken difference weight value, the used difference value and the used difference weight value to obtain a taken difference evaluation value and a used difference evaluation value;

[0054] The taking difference evaluation value and the using difference evaluation value are summed up and the operation of the medical end is evaluated based on the obtained difference evaluation value.

[0055] Optionally, in the method according to the present invention, the method further comprises:

[0056] In response to the interaction between the medical terminal and the foot pedal disposed under the medication operation table, the voice recognition unit is triggered to receive the newly added voice information output by the medical terminal, and a medication operation layer corresponding to the medication operation table is created, wherein the medication operation layer includes each virtual slot corresponding to each medication dispensing operation slot;

[0057] Filling each dispensing acquisition image corresponding to each dispensing operation slot into each corresponding virtual slot, marking the virtual slot filled with the dispensing acquisition image as a filled state, and marking the remaining virtual slots as an empty state;

[0058] Acquire a newly uploaded type and a newly uploaded dosage based on the newly added voice information, and divide any corresponding virtual slot in an empty state into slots to obtain a type sub-slot and a dosage sub-slot;

[0059] The newly uploaded types and the newly uploaded doses are respectively filled into the type sub-slot and the dose sub-slot corresponding to the same virtual slot, and the virtual slot is converted from the empty state to the filled state.

[0060] According to another aspect of the present invention, a medical interaction processing system based on natural language processing is provided, comprising:

[0061] The medicine dispensing collection module is configured to, in response to the medical end performing a medicine dispensing operation on any medicine operation slot located in the medicine operation table, perform medicine dispensing collection on the medicine operation slot to obtain the medicine dispensing information and slot number corresponding to the medicine operation slot;

[0062] A taking and collecting module is configured to, in response to the medical terminal performing a taking operation on the drug operation slot, obtain the taking and collecting information uploaded by the medical terminal based on the drug operation slot, and perform taking and collecting on the drug operation slot to obtain the taking and collecting information corresponding to the drug operation slot;

[0063] A first difference module is configured to determine a take difference value corresponding to the medicine operation slot based on information differentiation between the take upload information and the take collection information;

[0064] The residual material collection module is configured to, in response to obtaining the usage upload information uploaded by the medical terminal based on the drug operation slot, determine the drug recovery slot corresponding to the drug operation slot in the drug residual material table based on the slot number, and perform residual material collection on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot;

[0065] A second difference module is configured to determine a usage difference value corresponding to the drug operation slot based on the remaining material collection information and the usage upload information;

[0066] The operation evaluation module is configured to perform a fusion calculation on the taking difference value and the using difference value, and perform an operation evaluation on the medical end based on the obtained difference evaluation value.

[0067] According to the solution of the present invention, the server can not only perform corresponding dispensing and taking collection according to the dispensing operation and taking operation performed by the medical end on any drug operation slot, but also obtain the dispensing information, slot number and taking collection information corresponding to the drug operation slot according to the collection results, and compare the taking upload information with the taking collection information uploaded by the medical end to determine the corresponding information differentiation, and determine the taking difference value corresponding to the drug operation slot according to the information differentiation. After obtaining the use upload information uploaded by the medical end, the server can also determine the drug recovery slot corresponding to the drug operation slot in the drug residual material table according to the slot number, and perform residual material collection on the drug recovery slot, so as to obtain the residual material collection information corresponding to the drug recovery slot. Then, the server will determine the use difference value corresponding to the drug operation slot according to the residual material collection information and the use upload information, and then fuse the taking difference value and the use difference value to obtain the difference evaluation value, and perform corresponding operation evaluation on the medical end according to the difference evaluation value. The present invention can clearly record the information of medicines taken and used by the medical staff, and objectively evaluate the operation of the medical staff, thereby improving the evaluation accuracy to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A flowchart of a medical interaction processing method based on natural language processing according to an embodiment of the present invention is shown;

[0069] Figure 2 A schematic diagram of a drug operation slot according to an embodiment of the present invention is shown;

[0070] Figure 3 A schematic diagram of a virtual slot according to an embodiment of the present invention is shown;

[0071] Figure 4 A structural block diagram of a medical interaction processing system based on natural language processing according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0072] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0073] Medication records during surgery play a vital role in medical records. They record information such as the type and dosage of drugs used during surgery, which is of great significance for the doctor's subsequent treatment plan formulation and the patient's condition follow-up.

[0074] The existing technology is unable to verify the actual medication use of doctors during surgery and to evaluate the doctors' corresponding operations. There is a lack of a method and system for processing medication records based on data processing technology.

[0075] In order to solve the above problems in the prior art, the inventor proposes the solution of the present invention. An embodiment of the present invention provides a medical interaction processing method based on natural language processing, which can be executed in a computing device.

[0076] Figure 1 A flowchart of a medical interaction processing method based on natural language processing according to an embodiment of the present invention is shown, and the method is suitable for being executed in a computing device.

[0077] like Figure 1 As shown, the medical interaction processing method based on natural language processing proposed in this embodiment begins with step S102. In step S102, the following contents are included:

[0078] In response to the medical staff performing a dispensing operation on any drug operation slot located in the drug operation table, the dispensing collection is performed on the drug operation slot to obtain the dispensing information and the slot number corresponding to the drug operation slot.

[0079] For example, in this embodiment, the medicine operation table can be understood as a table for medical staff to perform medicine dispensing operations. There are multiple slots in the medicine operation table, namely, various medicine operation slots.

[0080] When the medical staff places the medicine into any medicine operation slot, that is, performs the medicine dispensing operation, the server will perform medicine dispensing collection on the medicine operation slot, thereby obtaining the medicine dispensing information and slot number corresponding to the medicine operation slot.

[0081] Furthermore, the above-mentioned “in response to the medical end performing a dispensing operation on any drug operation slot located in the drug operation table, performing dispensing collection on the drug operation slot to obtain the dispensing information and slot number corresponding to the drug operation slot” also includes the following steps:

[0082] Performing output detection on each gravity sensor disposed at the bottom of each drug operation slot in real time, and determining that the output state of any gravity sensor at any time is a valid collection state based on the output detection, the drug operation slot corresponding to the gravity sensor is determined as the dispensing operation slot;

[0083] Determine the time as the dispensing start time when the medical end starts to perform the dispensing operation on the dispensing operation slot, and establish a timed monitoring task corresponding to the gravity sensor based on the dispensing start time, so as to determine the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot based on the timed monitoring task;

[0084] Performing medication collection on the medication dispensing operation slot at any time after the medication dispensing end time to obtain a medication collection image corresponding to the medication dispensing operation slot;

[0085] The slot number corresponding to the medication dispensing operation slot and the actual medication type and the actual medication dosage included in the medication dispensing information are determined based on the medication dispensing acquisition image.

[0086] For example, in this embodiment, the drug operation table includes a plurality of drug operation slots distributed in an array, and a gravity sensor is provided at the bottom of each drug operation slot. The server will perform output detection on each gravity sensor in real time, so that the server can timely determine whether the medical staff performs corresponding drug dispensing operation on the drug operation slot.

[0087] When the gravity value outputted by the gravity sensor at any time is greater than the preset gravity value, it indicates that the medical end has performed a dispensing operation on the drug operation slot corresponding to the gravity sensor. Therefore, the server will determine the output state in which the gravity value outputted by any gravity sensor at any time in the output detection is greater than the preset gravity value as a valid collection state, and determine the drug operation slot corresponding to the gravity sensor as a dispensing operation slot.

[0088] Then, the server will determine this moment as the starting moment for the medical staff to start dispensing the medicine for this dispensing operation slot, and establish a timed monitoring task for the gravity sensor based on the starting moment of dispensing, so as to determine the ending moment for the medical staff to end the dispensing operation for this dispensing operation slot based on the timed monitoring task.

[0089] Next, the server will capture the image of the dispensing operation slot at any time after the end of the dispensing, that is, perform dispensing capture, so as to obtain the dispensing capture image corresponding to the dispensing operation slot, and then determine the slot number of the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information based on the dispensing capture image, so as to facilitate further verification by the subsequent server.

[0090] Furthermore, the above-mentioned “establishing a timed monitoring task corresponding to the gravity sensor based on the dispensing start time, so as to determine the dispensing end time when the medical end ends the dispensing operation for the dispensing operation slot based on the timed monitoring task” further includes the following steps:

[0091] Retrieving a reference monitoring period, and dividing the reference monitoring period into periods based on a half value to obtain a first monitoring period and a second monitoring period;

[0092] In response to the gravity sensor outputting a gravity acquisition value without gravity change in a reference monitoring period after the medication dispensing start time, determining a medication dispensing end time at which the medical terminal ends the medication dispensing operation on the medication dispensing operation slot based on the medication dispensing start time and the reference monitoring period;

[0093] In response to a gravity change in the gravity acquisition value output by the gravity sensor in a first monitoring period after the medication dispensing start time, the time corresponding to the gravity change is re-determined as the medication dispensing start time, and the medication dispensing end time at which the medical end ends the medication dispensing operation on the medication dispensing operation slot is determined again based on the timed monitoring task;

[0094] In response to a gravity change in the gravity collection value output by the gravity sensor between the first monitoring period and the second monitoring period, the time corresponding to the gravity change is re-determined as the dispensing start time, and a preset time increase coefficient is retrieved and multiplied by the reference monitoring period to obtain an updated reference monitoring period;

[0095] The scheduled monitoring task is updated based on the updated reference monitoring period, and the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot is determined based on the updated scheduled monitoring task.

[0096] For example, in this embodiment, the management end may pre-set a reference monitoring period in the server, for example, the reference monitoring period may be set to 5 seconds. The server will retrieve the reference monitoring period and divide the reference monitoring period into periods based on the half value, thereby obtaining a first monitoring period and a second monitoring period. For example, if the reference monitoring period is 5 seconds, the first monitoring period and the second monitoring period obtained are both 2.5 seconds.

[0097] When the gravity sensor outputs no gravity change in the gravity collection value during the reference monitoring period after the dispensing start time, it means that the medical end does not continue to perform the dispensing operation on the dispensing operation slot. Therefore, the server will determine the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot according to the dispensing start time and the reference monitoring period. For example, if the dispensing start time is 12:00:00 and the reference monitoring period is 5 seconds, the server will determine 12:00:05 as the dispensing end time.

[0098] When a gravity change occurs in the gravity collection value output by the gravity sensor during the first monitoring period after the start time of dispensing, it indicates that the medical staff is still dispensing medicine at the dispensing operation slot and the dispensing speed is relatively fast. Therefore, the server will redefine the moment when the gravity change occurs as the start time of dispensing medicine, and again determine the end time of dispensing medicine when the medical staff ends the dispensing operation at the dispensing operation slot based on the scheduled monitoring task.

[0099] When the gravity sensor outputs a gravity change in the gravity collection value between the first monitoring period and the second monitoring period, it indicates that the medical staff is still dispensing medicine at the dispensing operation slot, and the dispensing speed is slow. The medical staff may perform dispensing operations after the benchmark monitoring period.

[0100] In order to determine the end time of dispensing more accurately, the server will redefine the time when gravity changes as the start time of dispensing, and call the preset time increase coefficient and multiply it with the benchmark monitoring period to obtain an updated benchmark monitoring period, so as to make the updated benchmark monitoring period longer and avoid missing the monitoring of dispensing operations on the medical side.

[0101] Next, the server will update the scheduled monitoring task based on the updated reference monitoring period, and thereby determine the dispensing end time when the medical staff ends the dispensing operation for the dispensing operation slot based on the updated scheduled monitoring task.

[0102] Furthermore, the above-mentioned “determining the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information based on the dispensing acquisition image” further includes the following steps:

[0103] Obtaining a reference slot pixel value corresponding to the drug operation slot, and binarizing the medication collection image based on the reference slot pixel value to obtain each slot pixel point corresponding to the reference slot pixel value and each noise pixel point corresponding to the noise pixel value;

[0104] Connecting the pixels of each slot based on the pixels between adjacent positions, and determining a first image contour and a second image contour located in the medication collection image based on the pixel connections;

[0105] Determine a first number corresponding to each reference slot pixel point constituting the first image contour, and a second number corresponding to each reference slot pixel point constituting the second image contour, and determine the image contour with the largest corresponding number as the accommodating contour corresponding to the accommodating sub-slot, and determine the image contour with the smallest corresponding number as the numbered contour corresponding to the numbered sub-slot;

[0106] The dispensing acquisition image is segmented to obtain a containing image portion including the containing contour and a numbered image portion including the numbered contour, and based on the containing image portion and the numbered image portion, the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information are determined.

[0107] For example, in this embodiment, each drug operation slot has a spaced-apart accommodating sub-slot with a first size for accommodating medical drugs and a numbering sub-slot with a second size for marking the drug operation slot, such as Figure 2 As shown. Different slot numbers are filled in different numbered sub-slots, which can facilitate the subsequent server to determine the corresponding relationship between each drug operation slot and each drug recovery slot according to the slot number. Since the accommodating sub-slot may need to accommodate more medical drugs, the first size of the accommodating sub-slot is larger than the second size of the numbered sub-slot.

[0108] Under normal circumstances, the material and color of the containing sub-slot and the numbered sub-slot in each drug operation slot are the same, so the server can obtain the pixel value corresponding to the drug operation slot, that is, the reference slot pixel value. Then, the server will binarize the medication collection image according to the reference slot pixel value, thereby obtaining each slot pixel point corresponding to the reference slot pixel value and each noise pixel point corresponding to the noise pixel value other than each slot pixel point.

[0109] Then, the server connects the pixels of each slot at adjacent positions, thereby obtaining two image contours in the medication collection image, and determines the two image contours as the first image contour and the second image contour, respectively.

[0110] Next, the server will respectively determine the first number of pixels of each reference slot that constitutes the first image contour and the second number of pixels of each reference slot that constitutes the second image contour. Since the first size of the accommodating subslot is larger than the second size of the numbered subslot, the number of pixels of the reference slot corresponding to the accommodating subslot will also be larger than the number of pixels of the reference slot corresponding to the numbered subslot. Therefore, the server will determine the image contour with the largest corresponding number as the accommodating contour corresponding to the accommodating subslot and the image contour with the smallest corresponding number as the numbered contour corresponding to the numbered subslot.

[0111] At this time, the server will perform image segmentation on the dispensing acquisition image according to the containing contour and the numbering contour, thereby obtaining a containing image portion including the containing contour and a numbering image portion including the numbering contour, and determine the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information based on the containing image portion and the numbering image portion.

[0112] This embodiment can determine the corresponding accommodating image portion and the numbering image portion based on the size difference between the accommodating sub-slot and the numbering sub-slot, so as to facilitate the server to determine the slot number and the dispensing information.

[0113] Furthermore, the above-mentioned “determining the actual type of medication, the actual dosage of medication and the slot number corresponding to the medication dispensing operation slot based on the containing image portion and the numbering image portion” further includes the following steps:

[0114] Performing image recognition on the numbered image portion based on a pre-trained digital recognition model, and determining the slot number corresponding to the dispensing operation slot based on the recognition result;

[0115] Performing image recognition on the accommodating image portion based on a pre-trained contour recognition model, and determining each accommodating contour located in the accommodating image portion based on the recognition result;

[0116] Determine the size of each containment contour respectively, obtain the containment dimensions corresponding to each containment contour respectively, and determine the containment dimension with the largest corresponding dimension as the initial drug contour;

[0117] Performing distortion correction processing on the initial drug contour, and determining a corresponding drug size based on the obtained current drug contour;

[0118] Retrieving a preset drug comparison table, wherein the preset drug comparison table includes preset contours and preset sizes corresponding to different actual types of dispensed drugs;

[0119] Determine the confidence of the current drug contour and each preset contour corresponding to different actual types of medication based on the shape dimension, and obtain each shape confidence corresponding to different actual types of medication;

[0120] Determine the confidence of the drug size and each preset size corresponding to different actual types of medicines based on the size dimension, and obtain the confidence of each size corresponding to different actual types of medicines;

[0121] The shape confidence and size confidence corresponding to the same actual type of medicine are summed up to obtain various type confidences corresponding to different actual types of medicines, and the type confidence with the largest corresponding value among various type confidences is determined as the target confidence;

[0122] Determining the actual type of medicine dispensed corresponding to the target confidence as the actual type of medicine dispensed corresponding to the medicine dispensing operation slot, and obtaining the weight of the medicine;

[0123] Based on the gravity sensor, the slot gravity value corresponding to the dispensing operation slot at any time after the dispensing end time is obtained, and the slot gravity value is calculated by ratio with the medicine weight to obtain the actual dosage of the dispensing corresponding to the dispensing operation slot.

[0124] For example, in this embodiment, the server will perform image recognition on the numbered image portion through a pre-trained digital recognition model, and determine the slot number of the dispensing operation slot according to the recognition result. Then, the server will perform image recognition on the containing image portion through a pre-trained contour recognition model. Since a plurality of medical drugs may be contained in the containing sub-slot, the server can determine the contours or partial contours of the plurality of medical drugs located in the containing image portion, that is, each containing contour, according to the recognition result.

[0125] Since there may be a situation where drugs are stacked when a plurality of drugs are contained in a containing sub-slot, the server may not be able to completely determine the outline of each drug. However, in this case, the server can determine the outline of at least one drug. For example, if the drug is located at the top of the stacked drugs, the containing size of the drug will also be the largest. Therefore, the server will determine the size of each containing outline respectively, thereby obtaining the containing sizes corresponding to each containing outline, and determine the containing size with the largest corresponding size as the initial drug outline.

[0126] When placing each medicine, the medical staff may tilt the medicine, which may cause a certain distortion in the initial medicine outline determined by the server, which will affect the subsequent server's determination of the actual type of medicine corresponding to the initial medicine outline and cause corresponding errors. Therefore, the server will perform distortion correction processing on the initial medicine outline according to the existing technology, and determine the corresponding medicine size based on the obtained current medicine outline.

[0127] Then, the server will retrieve the preset drug comparison table, which includes preset contours and preset sizes corresponding to different actual types of medicines. Next, the server will determine the confidence of the current drug contour and the preset contours corresponding to different actual types of medicines based on the shape dimension, thereby obtaining the confidence of each shape corresponding to different actual types of medicines. Then, the drug size and the preset sizes corresponding to different actual types of medicines are determined based on the size dimension, thereby obtaining the confidence of each size corresponding to different actual types of medicines.

[0128] Next, the server will sum up the shape confidence and size confidence corresponding to the same actual type of medication, so as to obtain various type confidences corresponding to different actual types of medication. The higher the type confidence value, the more consistent the actual type of medication with the current drug outline. Therefore, the server will determine the type confidence with the largest corresponding value among the various type confidences as the target confidence, and then determine the actual type of medication corresponding to the target confidence as the actual type of medication corresponding to the medication operation slot, and obtain the weight of the actual type of medication.

[0129] Then, the server will obtain the slot gravity value of the dispensing operation slot at any time after the dispensing end time through the gravity sensor, and calculate the ratio of the slot gravity value to the medicine weight, so as to obtain the actual dosage of the medicine corresponding to the dispensing operation slot. For example, if the slot gravity value is 30mg and the medicine weight is 10mg, the actual dosage of the medicine is 3.

[0130] This embodiment determines the actual type of the medicine to be dispensed through shape dimensions and size dimensions, which has a certain degree of accuracy.

[0131] In step S104, the following contents are included:

[0132] In response to the medical end performing a picking operation on the drug operation slot, the picking upload information uploaded by the medical end based on the drug operation slot is obtained, and the picking collection is performed on the drug operation slot to obtain the picking collection information corresponding to the drug operation slot.

[0133] For example, in this embodiment, when the medical staff takes the medicine in the medicine operation slot, that is, performs the taking operation, the server will obtain the taking upload information output by the medical staff to the server through voice based on the medicine operation slot. For example, the obtained taking upload information is: "I took 30 mg of epinephrine."

[0134] Next, the server will perform the picking and collecting operation on the drug operation slot, thereby obtaining the picking and collecting information corresponding to the drug operation slot.

[0135] Further, the above-mentioned “in response to the medical end performing a taking operation on the drug operation slot, obtaining the taking upload information uploaded by the medical end based on the drug operation slot, and performing taking collection on the drug operation slot to obtain the taking collection information corresponding to the drug operation slot” also includes the following steps:

[0136] In response to a gravity collection value output by any gravity sensor at any time showing a decreasing trend, the time is determined as a starting time for the medical terminal to perform a taking operation on the drug operation slot, and the drug operation slot corresponding to the gravity sensor is determined as a taking operation slot;

[0137] Based on the picking up start time, the voice broadcast unit is triggered to output a preset voice, and the voice receiving module is triggered to receive voice in the current voice environment;

[0138] In response to the voice receiving module receiving the taking voice information output by the medical terminal at any time, the time is determined as the taking end time of the medical terminal performing the taking operation on the taking operation slot, and the taking upload type and the taking upload dosage corresponding to the taking operation slot are obtained based on the taking voice information;

[0139] Performing a pick-up and acquisition operation on the medication dispensing operation slot at any time after the medication dispensing end time, and obtaining a pick-up and acquisition image corresponding to the medication dispensing operation slot;

[0140] Determine the slot number corresponding to the taking operation slot and the actual taking type and the actual taking dose included in the taking information based on the taking acquisition image;

[0141] The dispensing information corresponding to the taking operation slot is determined based on the slot number, and the taking and collecting information corresponding to the taking operation slot is determined based on the information comparison between the dispensing information and the taking information.

[0142] For example, in this embodiment, when the gravity collection value output by any gravity sensor at any time shows a decreasing trend, it indicates that the medical staff is performing a taking operation on the medicine. Therefore, the server will determine this moment as the starting moment for the medical staff to take the medicine operation slot, and determine the medicine operation slot corresponding to the gravity sensor as the taking operation slot.

[0143] When the server determines the start time of taking, it will trigger the voice broadcast unit to output the preset voice, and trigger the voice receiving module to receive the voice in the current voice environment. For example, the preset voice can be: "What kind of medicine did you take and the specific dosage?"

[0144] When the voice receiving module receives the picking voice information output by the medical end at any time, it will determine the time as the end time of the picking operation of the picking operation slot by the medical end, and obtain the picking upload type and picking upload dose corresponding to the picking operation slot according to the picking voice information.

[0145] Next, the server will perform a collection of the medication dispensing operation slot at any time after the medication dispensing end time, thereby obtaining a collection image corresponding to the medication dispensing operation slot. The slot number corresponding to the medication dispensing operation slot and the actual type and dosage of medication taken included in the medication dispensing information are then determined based on the collection image.

[0146] Finally, the server will determine the dispensing information corresponding to the taking operation slot according to the slot number, and compare the dispensing information with the taking information, so as to determine the taking collection information corresponding to the taking operation slot.

[0147] In step S106, the following contents are included:

[0148] Based on the information differentiation between the taking upload information and the taking collection information, a taking difference value corresponding to the medicine operation slot is determined.

[0149] For example, in this embodiment, the server determines the information difference between the uploaded information and the collected information, and determines the difference value of taking corresponding to the drug operation slot according to the information difference.

[0150] Furthermore, the above-mentioned “determining the taking difference value corresponding to the drug operation slot based on the information differentiation between the taking uploaded information and the taking collected information” further includes the following steps:

[0151] In response to the actual type of taking being different from the uploaded type of taking, determining a preset first comparison value as a taking difference value corresponding to the medicine operation slot;

[0152] In response to the actual type of taking being the same as the uploaded type of taking, based on the dosage ratio between the actual dosage taken and the uploaded dosage taken, and based on the dosage ratio, a numerical adjustment is made to the preset second comparison value to obtain a taking difference value corresponding to the drug operation slot.

[0153] For example, in this embodiment, when the actual type taken is different from the uploaded type taken, it means that the medical staff may have misspoken about the type of medicine. At this time, the server will determine the preset first comparison value as the taking difference value corresponding to the medicine operation slot.

[0154] When the actual type of drug taken is the same as the uploaded type of drug taken, it means that the drug type in the uploaded information taken by the medical staff is correct but the drug dosage does not match the actual one. Therefore, the server will perform a dosage ratio between the actual dosage taken and the uploaded dosage taken, and then adjust the preset second comparison value numerically based on the obtained dosage ratio, so as to obtain the dosage difference value corresponding to the drug operation slot.

[0155] In step S108, the following contents are included:

[0156] In response to obtaining the usage upload information uploaded by the medical end based on the drug operation slot, a drug recovery slot corresponding to the drug operation slot is determined in the drug residual table based on the slot number, and residual material collection is performed on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot.

[0157] For example, in this embodiment, the residual medicine table can be understood as a table surface for medical staff to place residual medicines, and there are multiple medicine recovery slots corresponding to the medicine operation slots in the residual medicine table.

[0158] The server will obtain the usage upload information output by the medical staff to the server through voice based on a drug operation slot. For example, the usage upload information obtained is: "I used 20 mg of epinephrine."

[0159] Then, the server will determine the drug recovery slot corresponding to the drug operation slot in the drug residual material table through the slot number of the drug operation slot, and then perform residual material collection on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot.

[0160] In step S110, the following contents are included:

[0161] A usage difference value corresponding to the drug operation slot is determined based on the residual material collection information and the usage upload information.

[0162] For example, in this embodiment, the server will compare the obtained residual material collection information and the usage upload information to determine the usage difference value corresponding to the drug operation slot.

[0163] In step S112, the following contents are included:

[0164] The taking difference value and the using difference value are fused and calculated, and the operation of the medical end is evaluated based on the obtained difference evaluation value.

[0165] For example, in this embodiment, the server will perform a fusion calculation on the obtained difference value and the used difference value to obtain a difference evaluation value, and then perform a corresponding operation evaluation on the medical end according to the difference evaluation value.

[0166] Furthermore, the above-mentioned “combining and calculating the taking difference value and the using difference value, and evaluating the operation of the medical end based on the obtained difference evaluation value” further includes the following steps:

[0167] Retrieving a take difference weight value and a use difference weight value, wherein the use difference weight value is less than the take difference weight value;

[0168] Respectively multiplying the taken difference value and the taken difference weight value, the used difference value and the used difference weight value to obtain a taken difference evaluation value and a used difference evaluation value;

[0169] The taking difference evaluation value and the using difference evaluation value are summed up and the operation of the medical end is evaluated based on the obtained difference evaluation value.

[0170] For example, in this embodiment, the server will call out the taking difference weight value and the use difference weight value. Since the danger caused by the taking difference is greater, the use difference weight value may be smaller than the taking difference weight value; or, based on other considerations, the corresponding use difference weight value may be set to be greater than the taking difference weight value.

[0171] Then, the server will multiply the difference value and the difference weight value, the difference value and the difference weight value, respectively, to obtain the difference evaluation value and the difference evaluation value, and then sum the difference evaluation value and the difference evaluation value to obtain the difference evaluation value, and evaluate the operation of the medical end according to the difference evaluation value.

[0172] Furthermore, the method further comprises:

[0173] In response to the interaction between the medical terminal and the foot pedal disposed under the medication operation table, the voice recognition unit is triggered to receive the newly added voice information output by the medical terminal, and a medication operation layer corresponding to the medication operation table is created, wherein the medication operation layer includes each virtual slot corresponding to each medication dispensing operation slot;

[0174] Filling each dispensing acquisition image corresponding to each dispensing operation slot into each corresponding virtual slot, marking the virtual slot filled with the dispensing acquisition image as a filled state, and marking the remaining virtual slots as an empty state;

[0175] Acquire a newly uploaded type and a newly uploaded dosage based on the newly added voice information, and divide any corresponding virtual slot in an empty state into slots to obtain a type sub-slot and a dosage sub-slot;

[0176] The newly uploaded types and the newly uploaded doses are respectively filled into the type sub-slot and the dose sub-slot corresponding to the same virtual slot, and the virtual slot is converted from the empty state to the filled state.

[0177] For example, in this embodiment, in certain special circumstances, the medical staff may need to use medical drugs that have not been dispensed in advance due to some emergency situation. At this time, in order to make corresponding medication records for this situation, the medical staff can step on the foot pedal under the drug operating table to indicate that new medical drugs are needed.

[0178] When the medical terminal interacts with the foot pedal disposed under the medication operation table, the server triggers the voice recognition unit to receive the new voice information output by the medical terminal. For example, the new voice information may be: "I just took 30 mg of lidocaine."

[0179] At this time, the server will also create a drug operation layer corresponding to the drug operation table, and the drug operation layer includes virtual slots corresponding to each drug dispensing operation slot. The server will fill each drug dispensing acquisition image corresponding to each drug dispensing operation slot into the corresponding virtual slot, and mark the virtual slot filled with the drug dispensing acquisition image as filled, and mark the remaining virtual slots as empty. Figure 3 shown.

[0180] Then, the server will obtain the newly uploaded type and the newly uploaded dosage according to the newly added voice information. For example, if the newly added voice information is: "I just took 30 mg of lidocaine", the newly uploaded type will be lidocaine and the newly uploaded dosage will be 30 mg.

[0181] Next, the server will divide the slots of any virtual slot corresponding to the vacant state to obtain type sub-slots and dosage sub-slots, and then fill the newly uploaded types and newly uploaded dosages into the type sub-slots and dosage sub-slots corresponding to the same virtual slot, and convert the slot state of the virtual slot from an vacant state to a filled state.

[0182] According to the solution of the present invention, the server can not only perform corresponding dispensing and taking collection according to the dispensing operation and taking operation performed by the medical end on any drug operation slot, but also obtain the dispensing information, slot number and taking collection information corresponding to the drug operation slot according to the collection results, and compare the taking upload information with the taking collection information uploaded by the medical end to determine the corresponding information differentiation, and determine the taking difference value corresponding to the drug operation slot according to the information differentiation. After obtaining the use upload information uploaded by the medical end, the server can also determine the drug recovery slot corresponding to the drug operation slot in the drug residual material table according to the slot number, and perform residual material collection on the drug recovery slot, so as to obtain the residual material collection information corresponding to the drug recovery slot. Then, the server will determine the use difference value corresponding to the drug operation slot according to the residual material collection information and the use upload information, and then fuse the taking difference value and the use difference value to obtain the difference evaluation value, and perform corresponding operation evaluation on the medical end according to the difference evaluation value. The present invention can clearly record the information of medicines taken and used by the medical staff, and objectively evaluate the operation of the medical staff, thereby improving the evaluation accuracy to a certain extent.

[0183] Another embodiment of the present invention provides a medical interaction processing system based on natural language processing. Figure 4 As the corresponding system block diagram, the system includes:

[0184] The medicine dispensing collection module is configured to, in response to the medical end performing a medicine dispensing operation on any medicine operation slot located in the medicine operation table, perform medicine dispensing collection on the medicine operation slot to obtain the medicine dispensing information and slot number corresponding to the medicine operation slot;

[0185] A taking and collecting module is configured to, in response to the medical terminal performing a taking operation on the drug operation slot, obtain the taking and collecting information uploaded by the medical terminal based on the drug operation slot, and perform taking and collecting on the drug operation slot to obtain the taking and collecting information corresponding to the drug operation slot;

[0186] A first difference module is configured to determine a take difference value corresponding to the medicine operation slot based on information differentiation between the take upload information and the take collection information;

[0187] The residual material collection module is configured to, in response to obtaining the usage upload information uploaded by the medical terminal based on the drug operation slot, determine the drug recovery slot corresponding to the drug operation slot in the drug residual material table based on the slot number, and perform residual material collection on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot;

[0188] A second difference module is configured to determine a usage difference value corresponding to the drug operation slot based on the remaining material collection information and the usage upload information;

[0189] The operation evaluation module is configured to perform a fusion calculation on the taking difference value and the using difference value, and perform an operation evaluation on the medical end based on the obtained difference evaluation value.

[0190] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the preferred embodiment of the present invention.

[0191] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0192] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0193] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0194] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of submodules or subunits or subcomponents.

[0195] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present invention and to form different embodiments.

[0196] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0197] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.

[0198] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is primarily selected for readability and instructional purposes, rather than for explaining or limiting the subject matter of the present invention.

Claims

1. A medical interaction processing method based on natural language processing, characterized in that: The following steps are involved: In response to the medical end performing a dispensing operation on any drug operation slot located in the drug operation table, performing dispensing collection on the drug operation slot to obtain dispensing information and a slot number corresponding to the drug operation slot; In response to the medical end performing a taking operation on the drug operation slot, obtaining the taking upload information uploaded by the medical end based on the drug operation slot, and performing taking collection on the drug operation slot to obtain the taking collection information corresponding to the drug operation slot; Determining a take difference value corresponding to the drug operation slot based on information differentiation between the take upload information and the take collection information; In response to obtaining the usage upload information uploaded by the medical terminal based on the drug operation slot, determining a drug recovery slot corresponding to the drug operation slot in the drug residual material table based on the slot number, and performing residual material collection on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot; Determine a usage difference value corresponding to the drug operation slot based on the remaining material collection information and the usage upload information; The taking difference value and the using difference value are fused and calculated, and the operation of the medical end is evaluated based on the obtained difference evaluation value.

2. The medical interaction processing method based on natural language processing according to claim 1, characterized in that: The drug operation table includes drug operation slots arranged in an array; In response to the medical end performing a dispensing operation on any drug operation slot located in the drug operation table, performing dispensing collection on the drug operation slot to obtain dispensing information and a slot number corresponding to the drug operation slot, including: Performing output detection on each gravity sensor disposed at the bottom of each drug operation slot in real time, and determining that the output state of any gravity sensor at any time is a valid collection state based on the output detection, the drug operation slot corresponding to the gravity sensor is determined as the dispensing operation slot; Determine the time as the dispensing start time when the medical end starts to perform the dispensing operation on the dispensing operation slot, and establish a timed monitoring task corresponding to the gravity sensor based on the dispensing start time, so as to determine the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot based on the timed monitoring task; Performing medication collection on the medication dispensing operation slot at any time after the medication dispensing end time to obtain a medication collection image corresponding to the medication dispensing operation slot; The slot number corresponding to the medication dispensing operation slot and the actual type and actual dosage of the medication included in the medication dispensing information are determined based on the medication dispensing acquisition image.

3. The medical interaction processing method based on natural language processing according to claim 2, characterized in that: Establishing a timed monitoring task corresponding to the gravity sensor based on the medication dispensing start time, so as to determine the medication dispensing end time at which the medical end ends the medication dispensing operation on the medication dispensing operation slot based on the timed monitoring task, including: Retrieving a reference monitoring period, and dividing the reference monitoring period into periods based on a half value to obtain a first monitoring period and a second monitoring period; In response to the gravity sensor outputting a gravity acquisition value without gravity change in a reference monitoring period after the medication dispensing start time, determining a medication dispensing end time at which the medical terminal ends the medication dispensing operation on the medication dispensing operation slot based on the medication dispensing start time and the reference monitoring period; In response to a gravity change in the gravity acquisition value output by the gravity sensor in a first monitoring period after the medication dispensing start time, the time corresponding to the gravity change is re-determined as the medication dispensing start time, and the medication dispensing end time at which the medical end ends the medication dispensing operation on the medication dispensing operation slot is determined again based on the timed monitoring task; In response to a gravity change in the gravity collection value output by the gravity sensor between the first monitoring period and the second monitoring period, the time corresponding to the gravity change is re-determined as the dispensing start time, and a preset time increase coefficient is retrieved and multiplied by the reference monitoring period to obtain an updated reference monitoring period; The scheduled monitoring task is updated based on the updated reference monitoring period, and the dispensing end time when the medical end ends the dispensing operation on the dispensing operation slot is determined based on the updated scheduled monitoring task.

4. The medical interaction processing method based on natural language processing according to claim 2, characterized in that: Each of the drug operation slots includes spaced sub-slots with a first size for accommodating medical drugs and numbered sub-slots with a second size for marking the drug operation slots, wherein different numbered sub-slots are filled with different slot numbers, and the first size is larger than the second size; Determining the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information based on the dispensing acquisition image includes: Obtaining a reference slot pixel value corresponding to the drug operation slot, and binarizing the medication collection image based on the reference slot pixel value to obtain each slot pixel point corresponding to the reference slot pixel value and each noise pixel point corresponding to the noise pixel value; Connecting the pixels of each slot based on the pixels between adjacent positions, and determining a first image contour and a second image contour located in the medication collection image based on the pixel connections; Determine a first number corresponding to each reference slot pixel point constituting the first image contour, and a second number corresponding to each reference slot pixel point constituting the second image contour, and determine the image contour with the largest corresponding number as the accommodating contour corresponding to the accommodating sub-slot, and determine the image contour with the smallest corresponding number as the numbered contour corresponding to the numbered sub-slot; The dispensing acquisition image is segmented to obtain a containing image portion including the containing contour and a numbered image portion including the numbered contour, and based on the containing image portion and the numbered image portion, the slot number corresponding to the dispensing operation slot and the actual type and actual dosage of the dispensing included in the dispensing information are determined.

5. The medical interaction processing method based on natural language processing according to claim 4, characterized in that: Determining the actual type of medication, the actual dosage of medication, and the slot number corresponding to the medication dispensing operation slot based on the containing image portion and the numbering image portion includes: Performing image recognition on the numbered image portion based on a pre-trained digital recognition model, and determining the slot number corresponding to the dispensing operation slot based on the recognition result; Performing image recognition on the accommodating image portion based on a pre-trained contour recognition model, and determining each accommodating contour located in the accommodating image portion based on the recognition result; Determine the size of each containment contour respectively, obtain the containment dimensions corresponding to each containment contour respectively, and determine the containment dimension with the largest corresponding dimension as the initial drug contour; Performing distortion correction processing on the initial drug contour, and determining a corresponding drug size based on the obtained current drug contour; Retrieving a preset drug comparison table, wherein the preset drug comparison table includes preset contours and preset sizes corresponding to different actual types of dispensed drugs; Determine the confidence of the current drug contour and each preset contour corresponding to different actual types of medication based on the shape dimension, and obtain each shape confidence corresponding to different actual types of medication; Determine the confidence of the drug size and each preset size corresponding to different actual types of medicines based on the size dimension, and obtain the confidence of each size corresponding to different actual types of medicines; The shape confidence and size confidence corresponding to the same actual type of medicine are summed up to obtain various type confidences corresponding to different actual types of medicines, and the type confidence with the largest corresponding value among various type confidences is determined as the target confidence; Determining the actual type of medicine dispensed corresponding to the target confidence as the actual type of medicine dispensed corresponding to the medicine dispensing operation slot, and obtaining the weight of the medicine; Based on the gravity sensor, the slot gravity value corresponding to the dispensing operation slot at any time after the dispensing end time is obtained, and the slot gravity value is calculated by ratio with the medicine weight to obtain the actual dosage of the dispensing corresponding to the dispensing operation slot.

6. The medical interaction processing method based on natural language processing according to claim 5, characterized in that: In response to the medical end performing a taking operation on the drug operation slot, obtaining the taking upload information uploaded by the medical end based on the drug operation slot, and performing taking collection on the drug operation slot to obtain the taking collection information corresponding to the drug operation slot, including: In response to a gravity collection value output by any gravity sensor at any time showing a decreasing trend, the time is determined as a starting time for the medical terminal to perform a taking operation on the drug operation slot, and the drug operation slot corresponding to the gravity sensor is determined as a taking operation slot; Based on the picking up start time, the voice broadcast unit is triggered to output a preset voice, and the voice receiving module is triggered to receive voice in the current voice environment; In response to the voice receiving module receiving the taking voice information output by the medical terminal at any time, the time is determined as the taking end time of the medical terminal performing the taking operation on the taking operation slot, and the taking upload type and the taking upload dosage corresponding to the taking operation slot are obtained based on the taking voice information; Performing a pick-up and acquisition operation on the medication dispensing operation slot at any time after the medication dispensing end time, and obtaining a pick-up and acquisition image corresponding to the medication dispensing operation slot; Determine the slot number corresponding to the taking operation slot and the actual taking type and the actual taking dose included in the taking information based on the taking acquisition image; The dispensing information corresponding to the taking operation slot is determined based on the slot number, and the taking and collecting information corresponding to the taking operation slot is determined based on the information comparison between the dispensing information and the taking information.

7. The medical interaction processing method based on natural language processing according to claim 6, characterized in that: Determining a taking difference value corresponding to the drug operation slot based on the information differentiation between the taking uploaded information and the taking collected information includes: In response to the actual type of taking being different from the uploaded type of taking, determining a preset first comparison value as a taking difference value corresponding to the medicine operation slot; In response to the actual type of taking being the same as the uploaded type of taking, based on the dosage ratio between the actual dosage taken and the uploaded dosage taken, and based on the dosage ratio, a numerical adjustment is made to the preset second comparison value to obtain a taking difference value corresponding to the drug operation slot.

8. The medical interaction processing method based on natural language processing according to claim 1, characterized in that: The taking difference value and the using difference value are fused and calculated, and the operation of the medical end is evaluated based on the obtained difference evaluation value, including: Retrieving a take difference weight value and a use difference weight value, wherein the use difference weight value is different from the take difference weight value; Respectively multiplying the taken difference value and the taken difference weight value, the used difference value and the used difference weight value to obtain a taken difference evaluation value and a used difference evaluation value; The taking difference evaluation value and the using difference evaluation value are summed up and the operation of the medical end is evaluated based on the obtained difference evaluation value.

9. The medical interaction processing method based on natural language processing according to claim 2, characterized in that: The method further comprises: In response to the interaction between the medical terminal and the foot pedal disposed under the medication operation table, the voice recognition unit is triggered to receive the newly added voice information output by the medical terminal, and a medication operation layer corresponding to the medication operation table is created, wherein the medication operation layer includes each virtual slot corresponding to each medication dispensing operation slot; Filling each dispensing acquisition image corresponding to each dispensing operation slot into each corresponding virtual slot, marking the virtual slot filled with the dispensing acquisition image as a filled state, and marking the remaining virtual slots as an empty state; Acquire a newly uploaded type and a newly uploaded dosage based on the newly added voice information, and divide any corresponding virtual slot in an empty state into slots to obtain a type sub-slot and a dosage sub-slot; The newly uploaded types and the newly uploaded doses are respectively filled into the type sub-slot and the dose sub-slot corresponding to the same virtual slot, and the virtual slot is converted from the empty state to the filled state.

10. A medical interactive processing system based on natural language processing, characterized in that: The following steps are involved: The medicine dispensing collection module is configured to, in response to the medical end performing a medicine dispensing operation on any medicine operation slot located in the medicine operation table, perform medicine dispensing collection on the medicine operation slot to obtain the medicine dispensing information and slot number corresponding to the medicine operation slot; A taking and collecting module is configured to, in response to the medical terminal performing a taking operation on the drug operation slot, obtain the taking and collecting information uploaded by the medical terminal based on the drug operation slot, and perform taking and collecting on the drug operation slot to obtain the taking and collecting information corresponding to the drug operation slot; A first difference module is configured to determine a take difference value corresponding to the medicine operation slot based on information differentiation between the take upload information and the take collection information; The residual material collection module is configured to, in response to obtaining the usage upload information uploaded by the medical terminal based on the drug operation slot, determine the drug recovery slot corresponding to the drug operation slot in the drug residual material table based on the slot number, and perform residual material collection on the drug recovery slot to obtain residual material collection information corresponding to the drug recovery slot; A second difference module is configured to determine a usage difference value corresponding to the drug operation slot based on the remaining material collection information and the usage upload information; The operation evaluation module is configured to perform a fusion calculation on the taking difference value and the using difference value, and perform an operation evaluation on the medical end based on the obtained difference evaluation value.

Images