Liquid medicine extraction process detection control system and method for intelligent medicine dispensing

Abstract

The invention discloses a liquid medicine extraction process detection control system and method for intelligent medicine dispensing, and belongs to the technical field of equipment detection.The method comprises the following steps that medicine dispensing information is received, and an intelligent medicine dispensing system is controlled to grab medicine bottles according to the medicine dispensing information; extracting liquid medicine; the extraction torque of each pipettor is detected, and the extraction torque value is stored according to the detection time; the liquid medicine extraction amount is calculated; calculating a current average extraction torque value; comparing the current average extracted torque value with a torque target preset value; if the current average extraction torque value is smaller than the torque target preset value, liquid medicine extraction is stopped, and an alarm is given out. The method has the advantages that whether the medicine bottle is damaged or not can be automatically detected, and even if the damage is very small, accurate detection can be achieved.

Description

Technical Field

[0001] The present invention relates to the field of equipment detection, and in particular to a detection control system and method for a liquid medicine extraction process of intelligent medicine dispensing. Background Art

[0002] At present, intelligent dispensing refers to the process of using artificial intelligence technology, computer vision, control systems and other advanced technologies to achieve the automation, precision configuration and management of medical drugs. After obtaining the demand for drugs, the intelligent dispensing system grabs the required medicine bottles through visual recognition, number recognition and other methods, and drives the pipette through the motor to suck a fixed amount of liquid medicine into a unified container for mixing and dispensing. The liquid medicine extraction equipment usually includes a manipulator gripper, a suction device, a suction head and a drive component. The working principle of the pipette is that the piston realizes liquid suction and discharge through the expansion and contraction movement of the spring. Under the push of the piston, part of the air is discharged, and the liquid is sucked in by atmospheric pressure; then the piston pushes the air to discharge the liquid. The speed and force of the pipette can be well controlled by operating the pipette with the elastic characteristics of the spring. Since the intelligent dispensing system is an automated system as a whole, there is no personnel to supervise most of the time during the liquid medicine extraction process, so a detection system is needed to monitor the liquid medicine extraction process and repair it in time after a problem occurs.

[0003] The above-mentioned existing technical solutions have the following defects: currently, the extraction process of the medicine liquid needs to be ensured to proceed smoothly. If the medicine bottle is damaged, the pressure in the bottle will change during the extraction process, which may make it difficult to control the extraction amount of the medicine liquid, or even make it impossible to extract the medicine liquid. Summary of the invention

[0004] In order to accurately and automatically detect whether a medicine bottle is damaged, the present application provides a smart medicine dispensing liquid extraction process detection control system and method.

[0005] On the one hand, the present application provides a method for detecting and controlling the liquid extraction process of intelligent medicine dispensing, which adopts the following technical solution: A method for detecting and controlling a liquid medicine extraction process for intelligent medicine dispensing comprises the following steps: Set the liquid extraction speed, extraction dose and torque target preset values, and calculate the piston stroke of the pipette according to the extraction dose; Receive medication dispensing information, and control the intelligent medication dispensing system to grab the medicine bottle according to the medication dispensing information; Control the intelligent dispensing system to add the pipette into the grabbed medicine bottle and extract the medicine liquid; Detect the extraction torque of each pipette and store the extraction torque value; Detecting the current piston stroke of the pipette, and calculating the amount of liquid medicine extracted according to the current piston stroke of the pipette; When the current piston stroke of the pipette reaches the required piston stroke of the pipette, all extraction torque values ​​within the detection time are read to calculate the current average extraction torque value; Compare the current average extracted torque value with the torque target preset value; If the current average extraction torque value is less than the torque target preset value, the extraction of liquid medicine is stopped and an alarm is issued; If the current average extraction torque value is greater than or equal to the torque target preset value, the detection process is repeated until the amount of liquid medicine extracted reaches the requirements of the dispensing information.

[0006] By adopting the above scheme, during the process of dispensing medicine in the intelligent dispensing system, after the capillary is inserted into the medicine bottle, the motor drives the vacuum pump or the exhaust fan to absorb the liquid medicine. The system can estimate whether the medicine bottle is damaged by comparing the extraction torque of the motor with the preset torque target value. The current average extraction torque value is calculated because the torque of the motor will change slightly during extraction, and the current average extraction torque value calculation is more stable and accurate.

[0007] Preferably, the method further comprises the following steps: Pre-stored complete medicine bottle pictures from multiple angles; Take pictures of the medicine bottle and the intelligent medicine dispensing system, and mark them as medicine bottle image and system image respectively; In the system image information, a region similar to any complete medicine bottle image is selected, and the selected region is used as the system medicine bottle image; Find the closest complete medicine picture corresponding to the medicine bottle image and the system medicine bottle image and associate them, and compare the medicine bottle image or the system medicine bottle image with the associated complete medicine picture to determine whether there are cracks or damages; If it is determined that the medicine bottle image or the system medicine bottle image is cracked or damaged, an alarm will be issued and dispensing will be stopped.

[0008] By adopting the above solution, the medicine bottles with serious damage can be identified through visual recognition technology, and the dispensing of medicine can be stopped in advance.

[0009] Preferably, the method further comprises the following steps: The total capacity of the medicine bottles is pre-stored; Determine the ratio of the shadow part in the medicine bottle image and the system medicine bottle image to the entire medicine bottle, and calculate the remaining medicine amount in the medicine bottle according to the ratio of the shadow part to the entire medicine bottle; comparing the remaining drug amount in the drug bottle image and the system drug bottle image before the extraction with the drug amount of the corresponding drug type in the dispensing information; If the remaining medicine amount is less than the medicine amount of the corresponding medicine type in the dispensing information, an alarm is issued and dispensing is stopped; Calculate the amount of medicine to be extracted according to the extraction time and extraction torque of the system medicine bottle image in the extraction process, subtract the amount of medicine to be extracted from the amount of medicine of the corresponding medicine type in the dispensing information to obtain the remaining amount of medicine, and compare the remaining amount of medicine in the system medicine bottle image in the extraction process with the remaining amount of medicine to be extracted; If the remaining amount of the system medicine bottle image during the extraction is less than the expected remaining amount of medicine, the medicine bottle is judged to be leaking, an alarm is issued and the dispensing is stopped; If the remaining amount of medicine in the system medicine bottle image during extraction is greater than the expected remaining amount of medicine, the medicine bottle is judged to be leaking and an alarm is issued.

[0010] By adopting the above solution, since there are obvious traces of liquid medicine in the medicine bottle, the visual recognition technology can also estimate the remaining amount of medicine in the medicine bottle, and can determine whether the liquid medicine in the medicine bottle is enough to prepare the medicine, and judge whether the medicine bottle is leaking or leaking by comparing the remaining amount of medicine in the process of extracting the liquid medicine with the amount of medicine that should be left. Leakage will make the remaining amount of medicine less than the amount of medicine that should be left, and leakage will make it difficult for the motor to extract the liquid medicine due to the air pressure problem in the medicine bottle, resulting in the remaining amount of medicine being greater than the amount of medicine that should be left, making the judgment result more accurate.

[0011] Preferably, the method further comprises the following steps: The medicine bottle is laid flat and then tilted, and image information of the medicine bottle is continuously collected at fixed time intervals; All the continuously acquired image information is sorted according to time, the shadow part of the medicine bottle in each image information is identified and the end of the shadow part closest to the bottle mouth is marked; Estimate the centipoise value according to the movement process of the marking point in the image information; Adjust the motor's extraction torque according to the centipoise value until the actual liquid extraction speed is equivalent to the set liquid extraction speed.

[0012] By adopting the above solution, the approximate viscosity of the medicine liquid can be estimated by the flow rate of the medicine liquid in the medicine bottle, so that the extraction torque of the motor can be adjusted more conveniently.

[0013] Preferably, the method further comprises the following steps: Preset maximum torque value; When the extraction torque exceeds the maximum torque value, stop adjusting the extraction torque and keep the extraction torque at the maximum torque value; Collect image information of the extracted medicine bottle; The final remaining height of the medicine bottle is calculated according to the dispensing information, and the remaining height of the medicine bottle is calculated through the shadow part in the image information of the extracted medicine bottle. When the calculated remaining height of the medicine bottle is equal to the final remaining height of the medicine bottle, the extraction is stopped.

[0014] By adopting the above solution, visual recognition technology can be used to monitor from image information whether the amount of liquid medicine extracted meets the requirements of the dispensing information, and the dispensing is more accurate.

[0015] On the other hand, the present application provides a smart drug dispensing liquid extraction process detection and control system that adopts the following technical solutions: A method for detecting and controlling the liquid medicine extraction process of intelligent medicine dispensing, comprising a data storage module, a medicine dispensing analysis module, a torque detection module, an extraction calculation module, a damage detection module, and a medicine dispensing stop module; The data storage module receives and stores data; The dispensing analysis module receives dispensing information input from the outside and transmits it to the data storage module. The dispensing information includes the type of medicine and the extraction dose. The piston stroke of the pipette is calculated according to the extraction dose. The intelligent dispensing system is controlled to grab the corresponding medicine bottle according to the dispensing information, and the pipette is inserted into the grabbed medicine bottle, and the motor is started to extract the medicine liquid, and the start signal and the piston stroke of the pipette are transmitted to the torque detection module and the extraction calculation module. After receiving the start signal, the torque detection module starts to detect the extraction torque of each pipette and the current piston stroke of the pipette. When the current piston stroke of the pipette reaches the piston stroke that the pipette should travel, all extraction torque values ​​within the detection time are read, the current average extraction torque value is calculated, the current average extraction torque value is transmitted to the damage detection module, and the current piston stroke of the pipette is transmitted to the extraction calculation module; The extraction calculation module is preset with a stroke medicine volume calculation formula. After receiving the start signal, the extraction calculation module (4) receives the current piston stroke of the pipette, calculates the amount of liquid medicine extracted according to the current piston stroke of the pipette and the stroke medicine volume calculation formula, and transmits the amount of liquid medicine extracted to the dispensing stop module; The damage detection module is preset with a torque target preset value, and the damage detection module compares the current average extracted torque value with the torque target preset value. If the current average extracted torque value is less than the torque target preset value, a stop signal is sent to the dispensing stop module and an alarm is issued; The dispensing stop module controls the intelligent dispensing system to stop dispensing after receiving the stop signal. The dispensing stop module calls the dispensing information of the data storage module, compares the received amount of liquid medicine extracted with the extracted dose in the dispensing information, and when the received amount of liquid medicine extracted is equal to the extracted dose in the dispensing information, controls the intelligent dispensing system to stop dispensing.

[0016] By adopting the above scheme, during the process of dispensing medicine in the intelligent dispensing system, after the capillary is inserted into the medicine bottle, the motor drives the vacuum pump or the exhaust fan to absorb the liquid medicine. The system can estimate whether the medicine bottle is damaged by comparing the extraction torque of the motor with the preset torque target value. The current average extraction torque value is calculated because the torque of the motor will change slightly during extraction, and the current average extraction torque value calculation is more stable and accurate.

[0017] Preferably, it also includes an image acquisition module and an image analysis module; The data storage module pre-stores complete medicine bottle pictures at multiple angles; The image acquisition module captures picture information of the medicine bottle and the intelligent dispensing system, marks them as medicine bottle images and system images respectively, and transmits the medicine bottle images and system images to the image analysis module; The image analysis module calls the complete medicine bottle image stored in the data storage module, selects an area similar to any complete medicine bottle image in the system image, uses the selected area as the system medicine bottle image, searches for the closest complete medicine image corresponding to the medicine bottle image and the system medicine bottle image and associates them, and determines whether the medicine bottle image or the system medicine bottle image has cracks or damage by comparing it with the associated complete medicine image. If it is determined that the medicine bottle image or the system medicine bottle image has cracks or damage, a stop signal is sent to the dispensing stop module and an alarm is issued.

[0018] By adopting the above solution, the medicine bottles with serious damage can be identified through visual recognition technology, and the dispensing of medicine can be stopped in advance.

[0019] Preferably, it also includes a drug solution analysis module; The data storage module is preset with the total capacity of the medicine bottles; The liquid medicine analysis module calls the medicine bottle image and the system medicine bottle image of the image analysis module, determines the proportion of the shadow part in the medicine bottle image and the system medicine bottle image to the entire medicine bottle, calculates the remaining medicine amount in the medicine bottle according to the proportion of the shadow part to the entire medicine bottle, compares the remaining medicine amount in the medicine bottle image and the system medicine bottle image before liquid extraction with the medicine amount of the corresponding medicine type in the dispensing information, and if the remaining medicine amount is less than the medicine amount of the corresponding medicine type in the dispensing information, sends a stop signal to the dispensing stop module and issues an alarm, and according to the liquid extraction time of the system medicine bottle image in the liquid extraction, The amount of medicine to be extracted is calculated using the extraction torque, the amount of medicine to be extracted is subtracted from the amount of medicine corresponding to the type of medicine in the dispensing information to obtain the amount of medicine to be left, the remaining amount of medicine in the system medicine bottle image during extraction is compared with the amount of medicine to be left. If the remaining amount of medicine in the system medicine bottle image during extraction is less than the amount of medicine to be left, it is determined that the medicine bottle is leaking, a stop signal is sent to the dispensing stop module, and a leakage alarm is issued. If the remaining amount of medicine in the system medicine bottle image during extraction is greater than the amount of medicine to be left, it is determined that the medicine bottle is leaking, a stop signal is sent to the dispensing stop module, and a leakage alarm is issued.

[0020] By adopting the above solution, since there are obvious traces of liquid medicine in the medicine bottle, the visual recognition technology can also estimate the remaining amount of medicine in the medicine bottle, and can determine whether the liquid medicine in the medicine bottle is enough to prepare the medicine, and judge whether the medicine bottle is leaking or leaking by comparing the remaining amount of medicine in the process of extracting the liquid medicine with the amount of medicine that should be left. Leakage will make the remaining amount of medicine less than the amount of medicine that should be left, and leakage will make it difficult for the motor to extract the liquid medicine due to the air pressure problem in the medicine bottle, resulting in the remaining amount of medicine being greater than the amount of medicine that should be left, making the judgment result more accurate.

[0021] Preferably, it also includes a medicine bottle testing module and a speed adjustment module; After receiving the test instruction, the medicine bottle test controls the intelligent dispensing system to place the medicine bottle flat and then tilt it, and continuously collects image information of the medicine bottle at fixed time intervals, sorts all the continuously collected image information according to time, identifies the shadow part of the medicine bottle in each image information and marks the end of the shadow part closest to the bottle mouth, and transmits the marked image information to the speed adjustment module; The speed adjustment module estimates the centipoise value according to the movement process of the marking point in the image information, and adjusts the extraction torque of the motor according to the centipoise value until the actual liquid medicine extraction speed is equivalent to the set liquid medicine extraction speed.

[0022] By adopting the above solution, the approximate viscosity of the medicine liquid can be estimated by the flow rate of the medicine liquid in the medicine bottle, so that the extraction torque of the motor can be adjusted more conveniently.

[0023] Preferably, it also includes a torque insurance module, which has a preset maximum torque value. When the extraction torque exceeds the maximum torque value, the extraction torque is stopped from being adjusted, and the extraction torque is maintained at the maximum torque value. The image information of the extracted medicine bottle is collected, and the final remaining liquid height of the medicine bottle is calculated according to the dispensing information. The remaining liquid height of the medicine bottle is calculated by the shaded part in the image information of the extracted medicine bottle, and when the calculated remaining liquid height of the medicine bottle is equal to the final remaining liquid height of the medicine bottle, a stop signal is sent to the dispensing stop module.

[0024] By adopting the above solution, visual recognition technology can be used to monitor from image information whether the amount of liquid medicine extracted meets the requirements of the dispensing information, and the dispensing is more accurate.

[0025] In summary, the present invention has the following beneficial effects: 1. The system can estimate whether the medicine bottle is damaged by comparing the motor's extraction torque with the torque target preset value. The current average extraction torque value is calculated because the motor torque will change slightly during extraction, and the current average extraction torque value calculation is more stable and accurate.

[0026] 2. The system can determine whether the liquid medicine in the medicine bottle is sufficient for dispensing, and determine whether the medicine bottle is leaking liquid or air based on the remaining amount of medicine during the liquid medicine extraction process and the expected remaining amount of medicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is an overall system block diagram of the second embodiment of the present application.

[0028] Figure 2 It is a module block diagram of the image acquisition module and the drug solution analysis module of the second embodiment of the present application.

[0029] Description of reference numerals: 1. Data storage module; 2. Dispensing analysis module; 3. Torque detection module; 4. Extraction calculation module; 5. Damage detection module; 6. Dispensing stop module; 7. Image acquisition module; 71. Image analysis module; 8. Liquid analysis module; 81. Medicine bottle testing module; 82. Speed ​​adjustment module; 9. Torque insurance module. DETAILED DESCRIPTION

[0030] Embodiment 1: This embodiment of the present application discloses a method for detecting and controlling the liquid extraction process of intelligent medication dispensing, and the specific steps are as follows: Set the liquid extraction speed, detection time, torque target preset value, total capacity of the medicine bottle, and maximum torque. Calculate the piston stroke of the pipette according to the extracted dose. Pre-stored complete medicine bottle pictures from multiple angles. The intelligent dispensing system includes a capillary that can be inserted into the medicine bottle and a motor that provides suction to the capillary.

[0031] Receive medication dispensing information and control the intelligent medication dispensing system to grab the medicine bottle according to the medication dispensing information. The medication dispensing information includes the type of medicine and the extracted dosage.

[0032] The extraction torque of each pipette is detected and the extraction torque value is stored. The extraction torque can be detected by detecting the force feedback of the motor through a torque sensor.

[0033] The current piston stroke of the pipette is detected, and the amount of liquid medicine extracted is calculated according to the current piston stroke of the pipette. The pipette pipette has a fixed cross-sectional area, and the amount of liquid medicine extracted can be obtained by multiplying the current piston stroke of the pipette with the pipe cross-sectional area.

[0034] When the current piston stroke of the pipette reaches the required piston stroke of the pipette, all extraction torque values ​​within the detection time are read to calculate the current average extraction torque value.

[0035] Compare the current average extracted torque value with the torque target preset value.

[0036] If the current average extraction torque value is less than the torque target preset value, the extraction of liquid medicine will be stopped and an alarm will be sounded. If the medicine bottle is damaged, even if the damage occurs on the liquid surface and the medicine bottle does not leak, the effect of the medicine liquid will decrease due to the possible reaction of the medicine liquid with the air, and the medicine bottle still needs to be replaced.

[0037] If the current average extraction torque value is greater than or equal to the torque target preset value, the detection process is repeated until the amount of liquid medicine extracted reaches the requirements of the dispensing information.

[0038] The pictures of the medicine bottle and the intelligent medicine dispensing system are marked as the medicine bottle image and the system image respectively.

[0039] In the system image information, a region similar to any complete medicine bottle image is selected, and the selected region is used as the system medicine bottle image.

[0040] The closest complete medicine picture corresponding to the medicine bottle image and the system medicine bottle image is found and associated, and the medicine bottle image or the system medicine bottle image is compared with the associated complete medicine picture to determine whether there are cracks or damages.

[0041] If the image of the medicine bottle or the system medicine bottle image is determined to be cracked or damaged, an alarm will be issued and the dispensing of medicine will be stopped. The visual recognition technology can identify the medicine bottles with larger damage and stop the dispensing of medicine in advance.

[0042] The ratio of the shadow portion in the medicine bottle image and the system medicine bottle image to the entire medicine bottle is determined, and the remaining medicine amount in the medicine bottle is calculated according to the ratio of the shadow portion to the entire medicine bottle.

[0043] The remaining medicine amount in the medicine bottle image and the system medicine bottle image before liquid extraction is compared with the medicine amount of the corresponding medicine type in the dispensing information.

[0044] If the remaining medicine amount is less than the medicine amount of the corresponding medicine type in the dispensing information, an alarm is issued and dispensing is stopped.

[0045] The amount of medicine to be extracted is calculated based on the extraction time and extraction torque of the system medicine bottle image during extraction, the amount of medicine to be extracted is subtracted from the amount of medicine of the corresponding medicine type in the dispensing information to obtain the amount of medicine to be left, and the remaining amount of medicine in the system medicine bottle image during extraction is compared with the amount of medicine to be left.

[0046] If the remaining amount of medicine in the system medicine bottle image during extraction is less than the expected remaining amount of medicine, the medicine bottle is judged to be leaking, an alarm is issued and the dispensing of medicine is stopped.

[0047] If the remaining amount of the system medicine bottle image during the extraction is greater than the expected remaining amount of medicine, the medicine bottle is judged to be leaking and an alarm is issued. The internal pressure of a broken medicine bottle is different from that of a complete medicine bottle. When extracting, it is not necessary to overcome the sealing force of the medicine bottle itself, and the extraction force is also different, so the torque can accurately detect whether the medicine bottle is damaged.

[0048] After receiving the viscosity test instruction, the medicine bottle is laid flat and then tilted, and image information of the medicine bottle is continuously collected at fixed time intervals.

[0049] All the continuously acquired image information is sorted according to time, the shadow part of the medicine bottle in each image information is identified and the end of the shadow part closest to the bottle mouth is marked.

[0050] Estimate the centipoise value based on the movement process of the marker points in the image information.

[0051] Adjust the motor's extraction torque according to the centipoise value until the actual liquid extraction speed is equivalent to the set liquid extraction speed. By measuring the speed of the liquid flow in the medicine bottle, the approximate viscosity of the liquid can be estimated, making it easier to adjust the motor's extraction torque.

[0052] When the extraction torque exceeds the maximum torque value, the extraction torque adjustment is stopped to keep the extraction torque at the maximum torque value.

[0053] Call the image information of the extracted medicine bottle, calculate the final remaining liquid height of the medicine bottle according to the dispensing information, calculate the remaining liquid height of the medicine bottle through the shadow part in the image information of the extracted medicine bottle, and stop extracting liquid when the calculated remaining liquid height of the medicine bottle is equal to the final remaining liquid height of the medicine bottle.

[0054] The implementation principle of the intelligent dispensing liquid medicine extraction process detection and control method of the present application embodiment is as follows: during the dispensing process of the intelligent dispensing system, the capillary is inserted into the medicine bottle and the vacuum pump or the exhaust fan is driven by the motor to absorb the liquid medicine. The system can estimate whether the medicine bottle is damaged by comparing the extraction torque of the motor with the torque target preset value. The current average extraction torque value is calculated because the torque of the motor will change slightly during extraction, and the current average extraction torque value calculation is more stable and accurate.

[0055] Embodiment 2: This application embodiment discloses a detection and control system for the liquid medicine extraction process of intelligent medicine dispensing, such as Figure 1 and Figure 2 As shown, it includes a data storage module 1, a dispensing analysis module 2, a torque detection module 3, an extraction calculation module 4, a damage detection module 5, a dispensing stop module 6, an image acquisition module 7, an image analysis module 71, a liquid medicine analysis module 8, a medicine bottle testing module 81, a speed adjustment module 82, and a torque insurance module 9.

[0056] like Figure 1 As shown, the data storage module 1 receives and stores data. The data storage module 1 pre-stores complete medicine bottle pictures at multiple angles. The data storage module 1 is preset with the total capacity of the medicine bottle. The dispensing analysis module 2 receives the dispensing information input from the outside and transmits it to the data storage module 1. The dispensing information includes the type of medicine and the extraction dose. The piston stroke of the pipette is calculated according to the extraction dose. The intelligent dispensing system is controlled to grab the corresponding medicine bottle according to the dispensing information, and the pipette is inserted into the grabbed medicine bottle, the motor is started to extract the medicine solution, and the start signal and the piston stroke of the pipette are transmitted to the torque detection module 3 and the extraction calculation module 4.

[0057] like Figure 1 As shown, the torque detection module 3 includes a torque sensor arranged on the motor, and the torque sensor is used to detect the feedback torque of the motor, which is the extraction torque of the capillary. After receiving the start signal, the torque detection module 3 starts to detect the extraction torque of each pipette and the current piston stroke of the pipette. When the current piston stroke of the pipette reaches the piston stroke that the pipette should travel, all extraction torque values ​​within the detection time are read, the current average extraction torque value is calculated, and the current average extraction torque value is transmitted to the damage detection module 5. The current piston stroke of the pipette is transmitted to the extraction calculation module 4.

[0058] like Figure 1 As shown, the extraction calculation module 4 is preset with a stroke medicine amount calculation formula. After receiving the start signal, the extraction calculation module 4 receives the current piston stroke of the pipette, calculates the amount of liquid medicine extracted according to the current piston stroke of the pipette and the stroke medicine amount calculation formula, and transmits the amount of liquid medicine extracted to the dispensing stop module 6. The damage detection module 5 is preset with a torque target preset value. The damage detection module 5 compares the current average extraction torque value with the torque target preset value. If the current average extraction torque value is less than the torque target preset value, a stop signal is sent to the dispensing stop module 6 and an alarm is issued. The alarm notifies the user that the medicine bottle may be damaged.

[0059] like Figure 1 As shown, after receiving the stop signal, the dispensing stop module 6 controls the intelligent dispensing system to stop dispensing. The dispensing stop module 6 calls the dispensing information of the data storage module 1, compares the received amount of liquid medicine extracted with the extracted dose in the dispensing information, and when the received amount of liquid medicine extracted is equal to the extracted dose in the dispensing information, controls the intelligent dispensing system to stop dispensing.

[0060] like Figure 1As shown, the torque insurance module 9 is preset with a maximum torque value. When the extraction torque exceeds the maximum torque value, the extraction torque is stopped from being adjusted, the extraction torque is kept at the maximum torque value, the image information of the extracted medicine bottle is collected, the final remaining liquid height of the medicine bottle is calculated according to the dispensing information, the remaining liquid height of the medicine bottle is calculated by the shadow part in the image information of the extracted medicine bottle, and when the calculated remaining liquid height of the medicine bottle is equal to the final remaining liquid height of the medicine bottle, a stop signal is sent to the dispensing stop module 6. At the same time, the torque insurance module 9 prevents the motor from being damaged due to excessive extraction torque.

[0061] like Figure 2 As shown, the image acquisition module 7 includes a plurality of industrial cameras arranged on the intelligent dispensing system. The image acquisition module 7 captures the picture information of the medicine bottle and the intelligent dispensing system, marks them as medicine bottle images and system images respectively, and transmits the medicine bottle images and system images to the image analysis module 71. The image analysis module 71 calls the complete medicine bottle image stored in the data storage module 1, selects the area similar to any complete medicine bottle image in the system image, takes the selected area as the system medicine bottle image, searches for the closest complete medicine image corresponding to the medicine bottle image and the system medicine bottle image and associates them, and determines whether the medicine bottle image or the system medicine bottle image is cracked or damaged by comparing it with the associated complete medicine image. If it is determined that the medicine bottle image or the system medicine bottle image is cracked or damaged, a stop signal is sent to the dispensing stop module 6 and an alarm is issued. The visual recognition technology can be used to identify medicine bottles with larger damage, and dispensing can be stopped in advance.

[0062] like Figure 2 As shown, the liquid medicine analysis module 8 calls the medicine bottle image and the system medicine bottle image of the image analysis module 71, determines the proportion of the shadow part in the medicine bottle image and the system medicine bottle image to the entire medicine bottle, calculates the remaining medicine amount in the medicine bottle according to the proportion of the shadow part to the entire medicine bottle, compares the remaining medicine amount in the medicine bottle image and the system medicine bottle image before liquid extraction with the medicine amount of the corresponding medicine type in the dispensing information, and if the remaining medicine amount is less than the medicine amount of the corresponding medicine type in the dispensing information, sends a stop signal to the dispensing stop module 6 and issues an alarm, and according to the liquid extraction time of the system medicine bottle image during liquid extraction, The amount of medicine to be extracted is calculated by using the time and extraction torque, the amount of medicine to be extracted is subtracted from the amount of medicine of the corresponding medicine type in the dispensing information to obtain the amount of medicine to be left, the remaining amount of medicine in the system medicine bottle image during the extraction is compared with the amount of medicine to be left. If the remaining amount of medicine in the system medicine bottle image during the extraction is less than the amount of medicine to be left, it is determined that the medicine bottle is leaking, a stop signal is sent to the dispensing stop module 6 and a leakage alarm is issued. If the remaining amount of medicine in the system medicine bottle image during the extraction is greater than the amount of medicine to be left, it is determined that the medicine bottle is leaking, a stop signal is sent to the dispensing stop module 6 and a leakage alarm is issued.

[0063] like Figure 2As shown, after receiving the test instruction, the medicine bottle test controls the intelligent dispensing system to lay the medicine bottle flat and then tilt it, and at the same time continuously collects the image information of the medicine bottle at fixed time intervals. The image information of the medicine bottle can be taken by a separate industrial camera or by calling the image information taken by the image acquisition module 7. All the continuously collected image information is sorted by time, the shadow part of the medicine bottle in each image information is identified and the end of the shadow part closest to the bottle mouth is marked, and the marked image information is transmitted to the speed adjustment module 82. The speed adjustment module 82 estimates the centipoise value according to the movement process of the marked point in the image information, and adjusts the extraction torque of the motor according to the centipoise value until the actual liquid medicine extraction speed is equivalent to the set liquid medicine extraction speed. By measuring the speed of the liquid medicine flowing in the medicine bottle, the approximate viscosity of the liquid medicine can be estimated, so that the extraction torque of the motor can be adjusted more conveniently.

[0064] The embodiments of this specific implementation method are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting and controlling the liquid extraction process of intelligent drug dispensing, characterized in that: The following steps are involved: Set the liquid extraction speed, extraction dose and torque target preset values, and calculate the piston stroke of the pipette according to the extraction dose; Receive medication dispensing information, and control the intelligent medication dispensing system to grab the medicine bottle according to the medication dispensing information; Control the intelligent dispensing system to add the pipette into the grabbed medicine bottle and extract the medicine liquid; Detect the extraction torque of each pipette and store the extraction torque value; Detecting the current piston stroke of the pipette, and calculating the amount of liquid medicine extracted according to the current piston stroke of the pipette; When the current piston stroke of the pipette reaches the required piston stroke of the pipette, all extraction torque values ​​within the detection time are read to calculate the current average extraction torque value; Compare the current average extracted torque value with the torque target preset value; If the current average extraction torque value is less than the torque target preset value, the extraction of liquid medicine will be stopped and an alarm will be issued; If the current average extraction torque value is greater than or equal to the torque target preset value, the detection process is repeated until the amount of liquid medicine extracted reaches the requirements of the dispensing information.

2. The method for detecting and controlling the liquid medicine extraction process of intelligent medicine dispensing according to claim 1 is characterized in that: The following steps are also included: Pre-stored complete medicine bottle pictures from multiple angles; Take pictures of the medicine bottle and the intelligent medicine dispensing system, and mark them as medicine bottle image and system image respectively; In the system image information, a region similar to any complete medicine bottle image is selected, and the selected region is used as the system medicine bottle image; Find the closest complete medicine picture corresponding to the medicine bottle image and the system medicine bottle image and associate them, and compare the medicine bottle image or the system medicine bottle image with the associated complete medicine picture to determine whether there are cracks or damages; If it is determined that the medicine bottle image or the system medicine bottle image is cracked or damaged, an alarm will be issued and dispensing will be stopped.

3. The method for detecting and controlling the liquid medicine extraction process of intelligent medicine dispensing according to claim 2 is characterized in that: The following steps are also included: The total capacity of the medicine bottles is pre-stored; Determine the ratio of the shadow part in the medicine bottle image and the system medicine bottle image to the entire medicine bottle, and calculate the remaining medicine amount in the medicine bottle according to the ratio of the shadow part to the entire medicine bottle; comparing the remaining drug amount in the drug bottle image and the system drug bottle image before the extraction with the drug amount of the corresponding drug type in the dispensing information; If the remaining medicine amount is less than the medicine amount of the corresponding medicine type in the dispensing information, an alarm is issued and dispensing is stopped; Calculate the amount of medicine to be extracted according to the extraction time and extraction torque of the system medicine bottle image in the extraction process, subtract the amount of medicine to be extracted from the amount of medicine of the corresponding medicine type in the dispensing information to obtain the remaining amount of medicine, and compare the remaining amount of medicine in the system medicine bottle image in the extraction process with the remaining amount of medicine to be extracted; If the remaining amount of the system medicine bottle image during the extraction is less than the expected remaining amount of medicine, the medicine bottle is judged to be leaking, an alarm is issued and the dispensing is stopped; If the remaining amount of medicine in the system medicine bottle image during extraction is greater than the expected remaining amount of medicine, the medicine bottle is judged to be leaking and an alarm is issued.

4. The method for detecting and controlling the liquid medicine extraction process of intelligent medicine dispensing according to claim 1 is characterized in that: The following steps are also included: The medicine bottle is laid flat and then tilted, and image information of the medicine bottle is continuously collected at fixed time intervals; All the continuously acquired image information is sorted according to time, the shadow part of the medicine bottle in each image information is identified and the end of the shadow part closest to the bottle mouth is marked; Estimate the centipoise value according to the movement process of the marking point in the image information; Adjust the motor's extraction torque according to the centipoise value until the actual liquid extraction speed is equivalent to the set liquid extraction speed.

5. The method for detecting and controlling the liquid medicine extraction process of intelligent medicine dispensing according to claim 4 is characterized in that: The following steps are also included: Preset maximum torque value; When the extraction torque exceeds the maximum torque value, stop adjusting the extraction torque and keep the extraction torque at the maximum torque value; Collect image information of the extracted medicine bottle; The final remaining height of the medicine bottle is calculated according to the dispensing information, and the remaining height of the medicine bottle is calculated through the shadow part in the image information of the extracted medicine bottle. When the calculated remaining height of the medicine bottle is equal to the final remaining height of the medicine bottle, the extraction is stopped.

6. A method for detecting and controlling the liquid extraction process of intelligent medicine dispensing, characterized in that: It comprises a data storage module (1), a medicine dispensing analysis module (2), a torque detection module (3), an extraction calculation module (4), a damage detection module (5), and a medicine dispensing stop module (6); The data storage module (1) receives and stores data; The dispensing analysis module (2) receives dispensing information input from the outside and transmits it to the data storage module (1), the dispensing information including the type of medicine and the extraction dosage, calculates the piston stroke that the pipette should travel according to the extraction dosage, controls the intelligent dispensing system to grab the corresponding medicine bottle according to the dispensing information, extends the pipette into the grabbed medicine bottle, starts the motor to extract the medicine liquid, and transmits the start signal and the piston stroke that the pipette should travel to the torque detection module (3) and the extraction calculation module (4); After receiving the start signal, the torque detection module (3) starts to detect the extraction torque of each pipette and the current piston stroke of the pipette. When the current piston stroke of the pipette reaches the piston stroke that the pipette should travel, all extraction torque values ​​within the detection time are read, the current average extraction torque value is calculated, the current average extraction torque value is transmitted to the damage detection module (5), and the current piston stroke of the pipette is transmitted to the extraction calculation module (4); The extraction calculation module (4) is preset with a stroke medicine volume calculation formula. After receiving the start signal, the extraction calculation module (4) receives the current piston stroke of the pipette, calculates the amount of liquid medicine extracted according to the current piston stroke of the pipette and the stroke medicine volume calculation formula, and transmits the amount of liquid medicine extracted to the dispensing stop module (6); The damage detection module (5) is preset with a torque target preset value. The damage detection module (5) compares the current average extracted torque value with the torque target preset value. If the current average extracted torque value is less than the torque target preset value, a stop signal is sent to the dispensing stop module (6) and an alarm is issued. The dispensing stop module (6) controls the intelligent dispensing system to stop dispensing after receiving the stop signal. The dispensing stop module (6) calls the dispensing information of the data storage module (1), compares the received amount of liquid medicine extracted with the extracted dose in the dispensing information, and controls the intelligent dispensing system to stop dispensing when the received amount of liquid medicine extracted is equal to the extracted dose in the dispensing information.

7. The intelligent drug dispensing liquid extraction process detection control system and method according to claim 6 is characterized by: It also includes an image acquisition module (7) and an image analysis module (71); The data storage module (1) pre-stores complete medicine bottle pictures at multiple angles; The image acquisition module (7) captures image information of the medicine bottle and the intelligent dispensing system, marks them as medicine bottle images and system images respectively, and transmits the medicine bottle images and system images to the image analysis module (71); The image analysis module (71) calls the complete medicine bottle image stored in the data storage module (1), selects an area in the system image that is similar to any complete medicine bottle image, uses the selected area as the system medicine bottle image, searches for the closest complete medicine image corresponding to the medicine bottle image and the system medicine bottle image, and associates them. The medicine bottle image or the system medicine bottle image is compared with the associated complete medicine image to determine whether there are cracks or damages. If it is determined that the medicine bottle image or the system medicine bottle image has cracks or damages, a stop signal is sent to the dispensing stop module (6) and an alarm is issued.

8. The intelligent drug dispensing liquid extraction process detection control system and method according to claim 7 is characterized by: It also includes a drug solution analysis module (8); The data storage module (1) is preset with the total capacity of the medicine bottles; The liquid medicine analysis module (8) calls the medicine bottle image and the system medicine bottle image of the image analysis module (71), determines the proportion of the shadow part in the medicine bottle image and the system medicine bottle image to the entire medicine bottle, calculates the remaining medicine amount in the medicine bottle according to the proportion of the shadow part to the entire medicine bottle, compares the remaining medicine amount in the medicine bottle image and the system medicine bottle image before liquid extraction with the medicine amount of the corresponding medicine type in the dispensing information, and if the remaining medicine amount is less than the medicine amount of the corresponding medicine type in the dispensing information, sends a stop signal to the dispensing stop module (6) and issues an alarm, and according to the liquid extraction of the system medicine bottle image during liquid extraction, The amount of medicine to be extracted is calculated using time and extraction torque, the amount of medicine to be extracted is subtracted from the amount of medicine corresponding to the type of medicine in the dispensing information to obtain the amount of medicine remaining, the remaining amount of medicine in the system medicine bottle image in the extracted liquid is compared with the amount of medicine remaining, if the remaining amount of medicine in the system medicine bottle image in the extracted liquid is less than the amount of medicine remaining, it is determined that the medicine bottle is leaking, a stop signal is sent to the dispensing stop module (6) and a leakage alarm is issued, if the remaining amount of medicine in the system medicine bottle image in the extracted liquid is greater than the amount of medicine remaining, it is determined that the medicine bottle is leaking, a stop signal is sent to the dispensing stop module (6) and a leakage alarm is issued.

9. The intelligent drug dispensing liquid extraction process detection control system and method according to claim 6 is characterized by: It also includes a medicine bottle testing module (81) and a speed adjustment module (82); After receiving the test instruction, the medicine bottle test controls the intelligent dispensing system to place the medicine bottle flat and then tilt it, and at the same time continuously collects image information of the medicine bottle at fixed time intervals, sorts all the continuously collected image information according to time, identifies the shadow part of the medicine bottle in each image information and marks the end of the shadow part closest to the bottle mouth, and transmits the marked image information to the speed adjustment module (82); The speed adjustment module (82) estimates the centipoise value according to the movement process of the marked point in the image information, and adjusts the extraction torque of the motor according to the centipoise value until the actual liquid medicine extraction speed is equivalent to the set liquid medicine extraction speed.

10. The intelligent drug dispensing liquid extraction process detection control system and method according to claim 9, characterized in that: It also includes a torque insurance module (9), which is preset with a maximum torque value. When the extraction torque exceeds the maximum torque value, the extraction torque is stopped from being adjusted, and the extraction torque is maintained at the maximum torque value. The image information of the extracted medicine bottle is collected, and the final remaining medicine liquid height of the medicine bottle is calculated according to the dispensing information. The remaining medicine liquid height of the medicine bottle is calculated by the shadow part in the image information of the extracted medicine bottle. When the calculated remaining medicine liquid height of the medicine bottle is equal to the final remaining medicine liquid height of the medicine bottle, a stop signal is sent to the dispensing stop module (6).

Images