Intelligent medicine taking identification system and method based on machine vision

Abstract

The invention relates to the technical field of health-related information systems and cloud platforms, in particular to an intelligent medicine taking recognition system and method based on machine vision. Collecting medicine taking data, an actual medicine taking operation sequence of a medicine taking robot, an actual medicine taking period of the actual medicine taking operation sequence, a historical medicine taking operation sequence associated with the medicine taking data and a historical medicine taking period of the historical medicine taking operation sequence; a candidate medicine taking operation sequence is filtered out from the actual medicine taking operation sequence; analyzing whether an optimization sequence associated with the candidate medicine taking operation sequence exists or not, and outputting an analysis log; when the analysis log represents that an optimization sequence associated with the candidate medicine taking operation sequence exists, the cloud platform generates a medicine taking sequence associated with a plurality of medicine taking patients based on the optimization sequence and the candidate medicine taking operation sequence; when the medicine taking robot takes medicine from the medicine taking patient, identity verification is carried out. According to the method, when the medicine taking sequence can be planned in an area with complex patient state conditions, the planned sequence is closer to the actual sequence, more time is saved, and the medicine taking efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of health-related information systems and cloud platforms, and in particular to an intelligent medicine-taking recognition system and method based on machine vision. Background Art

[0002] With the development of computer technology, patients' demand for medication is increasing.

[0003] In the current drug dispensing scheduling system, drug dispensing personnel input multiple target patient wards and beds into the drug dispensing scheduling system, and the drug dispensing scheduling system can generate the order of arriving at the corresponding target patient wards and beds. Compared with the past when there was no drug dispensing scheduling system, it not only facilitates drug dispensing personnel, but also improves the efficiency of drug dispensing.

[0004] However, when the drug dispensing scheduling system under the existing technology plans the order of drug dispensing in areas with complex patient conditions, the planned order will be farther and more time-consuming than the actual order, thereby reducing the efficiency of drug dispensing. Summary of the invention

[0005] To achieve the above objectives, this application provides the following technical solutions: According to the first aspect of the present invention, the present invention claims protection for an intelligent medication identification method based on machine vision, which is applied to a cloud platform, and the method comprises: The cloud platform collects medication collection data, and the medication collection data includes multiple medication collection patients; The cloud platform collects the actual drug taking operation sequence of the drug taking robot, the actual drug taking cycle of the actual drug taking operation sequence, the historical drug taking operation sequence associated with the drug taking data, and the historical drug taking cycle of the historical drug taking operation sequence; The cloud platform filters out candidate drug fetching operation sequences from the actual drug fetching operation sequence, wherein the sequence deviation between the candidate drug fetching operation sequence and the historical drug fetching operation sequence is not less than a preset deviation value threshold, and the candidate drug fetching cycle of the candidate drug fetching operation sequence is less than the historical drug fetching cycle; The cloud platform analyzes whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs an analysis log; When the parsed log representation has an optimized sequence associated with the candidate medication pickup running sequence, the cloud platform generates a medication pickup sequence associated with the plurality of medication pickup patients based on the optimized sequence and the candidate medication pickup running sequence; The medicine dispensing robot performs identity verification when dispensing medicine for the patient.

[0006] Furthermore, the analysis determines whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs a analysis log, including: When the candidate medication fetching operation sequence includes a preset patient in the preset ward where the POI is located, outputting a parsing log having an optimized sequence associated with the candidate medication fetching operation sequence; When the candidate medication fetching operation sequence includes a preset ward where the POI is located but there is no preset patient, a parsing log having no optimized sequence associated with the candidate medication fetching operation sequence is output.

[0007] Further, the analysis determines whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs a analysis log, including: The first patient of the medicine taking robot and the second patient of the medicine taking box are collected in the process of the medicine taking robot and the medicine taking box taking medicine according to the candidate medicine taking operation sequence; When the medication similarity between the first patient and the second patient exceeds a preset medication similarity threshold, outputting a parsing log having an optimized sequence associated with the candidate medication running sequence; When the medication similarity between the first patient and the second patient does not exceed a preset medication similarity threshold, a parsing log having no optimized sequence associated with the candidate medication running sequence is output.

[0008] Furthermore, the method further comprises: When the parsed log representation does not have an optimized sequence associated with the candidate medication pickup operation sequence, the candidate medication pickup operation sequence is used as the medication pickup operation sequence associated with the multiple medication pickup patients.

[0009] Furthermore, the method further comprises: Based on the medicine picking sequence, guidance information is pushed to the processor so that the processor provides animation information associated with the optimized sequence to the medicine picking robot based on the guidance information.

[0010] Further, after outputting the parsing log having the optimized sequence associated with the candidate medicine taking operation sequence, the method further includes: transmitting verification information to the processor, so that the processor informs the medicine taking robot to verify the parsing log based on the verification information, and feeds back the verification log to the cloud platform; When the parsed log representation has an optimized sequence associated with the candidate medication pickup running sequence, the cloud platform generates medication pickup sequences associated with the plurality of medication pickup patients based on the optimized sequence and the candidate medication pickup running sequence, including: When both the parsing log and the checking log indicate an optimized sequence associated with the candidate medication pickup running sequence, a medication pickup sequence associated with the plurality of medication pickup patients is generated based on the optimized sequence and the candidate medication pickup running sequence.

[0011] Furthermore, when the preset patient is a traditional Chinese medicine patient, it at least includes: Patients with phlegm obstruction, heart blood stasis, qi stagnation in the chest, or cold stagnation in the heart meridians; When the preset patients are western medicine patients, they at least include: rhinitis patients, sinusitis patients, and pharyngitis patients.

[0012] Furthermore, the medicine dispensing robot performs identity verification on the patient when dispensing medicine, and further includes: When the medicine-taking robot stops in front of the patient taking medicine, it scans the facial information of the patient taking medicine; The medicine dispensing robot compares the facial information with the information in the patient database to obtain the medicine prescription information of the corresponding patient; The drug-taking patient obtains the drug from the drug-taking robot by himself, and when the drug-taking robot detects that the drug taken by the drug-taking patient is inconsistent with the drug prescription information of the corresponding patient, an early warning is issued to the cloud platform; The medicine dispensing robot recycles the medicine dispensing drugs.

[0013] According to the second aspect of the present invention, the present invention claims protection for an intelligent medication identification system based on machine vision, characterized in that the system is used to execute the intelligent medication identification method based on machine vision, and comprises: A collection unit is used to collect medication data, wherein the medication data includes a plurality of medication patients, and collects an actual medication operation sequence of the medication robot, an actual medication cycle of the actual medication operation sequence, a historical medication operation sequence associated with the medication data, and a historical medication cycle of the historical medication operation sequence; A filtering unit, used to filter out a candidate drug fetching operation sequence from the actual drug fetching operation sequence, wherein the sequence deviation between the candidate drug fetching operation sequence and the historical drug fetching operation sequence is not less than a preset deviation value threshold, and the candidate drug fetching cycle of the candidate drug fetching operation sequence is less than the historical drug fetching cycle; A parsing unit, used for parsing whether there is an optimized sequence associated with the candidate drug taking operation sequence, and outputting a parsing log; The optimization unit is used for generating a medication taking sequence associated with the plurality of medication taking patients based on the optimization sequence and the candidate medication taking sequence when the parsed log representation has an optimization sequence associated with the candidate medication taking sequence.

[0014] The present application relates to the field of intelligent medicine collection technology, and in particular to an intelligent medicine collection recognition system and method based on machine vision, which collects medicine collection data and the actual medicine collection operation sequence of the medicine collection robot, the actual medicine collection cycle of the actual medicine collection operation sequence, the historical medicine collection operation sequence associated with the medicine collection data, and the historical medicine collection cycle of the historical medicine collection operation sequence; filters out candidate medicine collection operation sequences from the actual medicine collection operation sequence; analyzes whether there is an optimized sequence associated with the candidate medicine collection operation sequence, and outputs a parsing log; when the parsing log indicates that there is an optimized sequence associated with the candidate medicine collection operation sequence, the cloud platform generates a plurality of medicine collection sequences associated with the medicine collection patients based on the optimized sequence and the candidate medicine collection operation sequence; the medicine collection robot performs identity verification when collecting medicine for the medicine collection patient. The present invention can plan the medicine collection sequence in areas with complex patient status, and the planned sequence is closer to the actual sequence, saves more time, and improves the efficiency of medicine collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of an application scenario of an intelligent medication identification method based on machine vision claimed in an embodiment of the present application; Figure 2 A schematic diagram of a second application scenario of a machine vision-based intelligent medication identification method claimed in an embodiment of the present application; Figure 3 A workflow diagram of an intelligent drug-taking identification method based on machine vision as claimed in an embodiment of the present application; Figure 4 This is a structural module diagram of a machine vision-based intelligent medicine dispensing recognition system for which protection is requested in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative patient relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). When the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0018] Reference to "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present application. The appearance of the phrase in various patients in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0019] like Figure 1 As shown, this figure is a schematic diagram of an application scenario of an intelligent medicine picking identification method based on machine vision provided by an embodiment of the present application. In this application scenario, the patients who pick up medicine include a first patient 101, a second patient 102, a third patient 103, and a fourth patient 104. Based on the above patients, the medicine picking scheduling system can plan the forward sequence, such as R1, R2, and R3 in the figure. When the medicine picking robot picks up medicine and runs the medicine box to the first patient 101 to complete the delivery, it can pick up medicine and run the medicine box to the second patient 102 (the medicine picking operation sequence is as follows Figure 1 As shown in R1 in the figure), after the medicine is taken, the medicine box for taking medicine comes to the third patient 103 for taking medicine (the order of taking medicine is as follows Figure 1 As shown in R2 in the figure), after the medicine is taken, the medicine box for taking medicine comes to the fourth patient 104 for taking medicine (the order of taking medicine is as follows Figure 1 ).

[0020] from Figure 1It can be seen from the shown medication dispensing operation sequence that this medication dispensing operation sequence has overlapping situations. For example, after the medication dispensing robot completes dispensing medicine for the third patient 103, due to the baffle 202 set in the center of the ward corridor, the medicine box needs to go around to the end of the corridor and then turn right to reach the fourth patient 104. From the detour to the right turn, the medication dispensing robot has already moved forward once when dispensing medicine for the second patient 102, and the medication dispensing operation sequence planning is unreasonable.

[0021] In fact, after the medicine dispensing robot finishes dispensing medicine for the second patient 102, it can go through the green channel 201 to dispensing medicine for the third patient 103, then walk back to the medicine dispensing box through the green channel, and then dispensing medicine and run the medicine dispensing box to the end of the front channel, turn right, and dispensing medicine for the fourth patient 104.

[0022] It can be seen that the order planned by the current drug-taking scheduling system has the problem of wasting time, thereby reducing the efficiency of drug-taking.

[0023] In view of this, an embodiment of the present application provides an intelligent drug taking identification method based on machine vision, which is applied to a cloud platform. The method includes: the cloud platform collects drug taking data, and the drug taking data includes multiple drug taking patients. The cloud platform can also collect the actual drug taking operation sequence of the drug taking robot, the actual drug taking cycle of the actual drug taking operation sequence, the historical drug taking operation sequence associated with the drug taking data, and the historical drug taking cycle of the historical drug taking operation sequence, and then filter out candidate drug taking operation sequences from the actual drug taking operation sequence. Compared with the historical drug taking operation sequence, the sequence deviation value of the candidate drug taking operation sequence is not less than a preset deviation value threshold, and the candidate drug taking cycle of the candidate drug taking operation sequence is less than the historical drug taking cycle. Then, it is analyzed whether there is an optimized sequence associated with the candidate drug taking operation sequence, and a parsing log is output. When the parsing log indicates that there is an optimized sequence associated with the candidate drug taking operation sequence, a drug taking sequence associated with multiple drug taking patients is generated based on the optimized sequence and the candidate drug taking operation sequence. After outputting the medicine picking sequence, the medicine picking robot can pick up medicines based on the actual medicine picking operation sequence in the medicine picking sequence, and can pick up medicines without carrying a medicine box based on the optimized sequence in the medicine picking sequence, thereby achieving higher medicine picking efficiency.

[0024] like Figure 2 As shown in FIG. 1 , this figure is a schematic diagram of another application scenario of an intelligent drug taking identification method based on machine vision provided in an embodiment of the present application. The intelligent drug taking identification method based on machine vision in an embodiment of the present application is used to modify the original drug taking operation sequence ( Figure 1 After optimizing the drug taking operation sequence R1, R2 and R3 in the output Figure 2The medicine taking sequence shown includes medicine taking operation sequences J1 and J2 and optimization sequences F1 and F2. The optimization sequence may be a non-medication taking operation sequence.

[0025] Among them, the medicine taking robot first takes the medicine and runs the medicine taking box according to the medicine taking operation sequence J1 from the first medicine taking patient 101 to the second medicine taking patient 102. After completing the delivery to the second medicine taking patient 102, it walks according to the optimization sequence F1 and passes through the green channel 201 to reach the third medicine taking patient 103. After completing the delivery to the third medicine taking patient 103, it walks according to the optimization sequence F2 and passes through the green channel 201 back to the medicine taking box. Then it takes the medicine and runs the medicine taking box according to the medicine taking operation sequence J2 to reach the fourth medicine taking patient 104, and delivers the medicine to the fourth medicine taking patient 104.

[0026] It can be seen that the second patient 102 can walk to the third patient 103 without carrying the medicine box to the end of the corridor, and there is no need to carry the medicine box to go through the original ward corridor again. For specific patients (for example, the ward corridor has a green channel), taking medicine in a specific way (such as walking) can reduce the medicine collection cycle and improve the efficiency of medicine collection.

[0027] In order to make the technical solution of the present application clearer and easier to understand, the following is an introduction to the intelligent drug taking recognition method based on machine vision provided by the embodiment of the present application in conjunction with the accompanying drawings. Figure 3 As shown in the figure, this figure is a flow chart of a method for intelligent drug-taking recognition based on machine vision provided in an embodiment of the present application. The method can be executed by a cloud platform or by a cloud platform and a processor in collaboration. The following takes the collaborative execution of a cloud platform and a processor as an example to introduce the method. The method includes: S301. The cloud platform collects medication collection data.

[0028] The medication collection data includes multiple patients who need to pick up medication. In some examples, a merchant can transmit an order request to the cloud platform. The order request can carry the merchant's geographical patients and the required goods. After receiving the order request, the cloud platform can parse the order request and then output the medication collection data, that is, multiple patients who need to pick up medication.

[0029] S302, the cloud platform collects the actual medicine picking operation sequence of the medicine picking robot, the actual medicine picking cycle of the actual medicine picking operation sequence, the historical medicine picking operation sequence associated with the medicine picking data, and the historical medicine picking cycle of the historical medicine picking operation sequence.

[0030] The actual medication operation sequence refers to the order in which the medication robot moves forward with the medicine box during the process of taking medicine from the above-mentioned multiple medication patients. The actual medication cycle refers to the cycle required for the medication robot to take medicine from the multiple medication patients according to the actual medication operation sequence. The historical medication operation sequence refers to the order in which the medication robot moves forward with the medicine box during the process of taking medicine from the above-mentioned multiple medication patients in historical time periods. The historical medication cycle refers to the cycle required for the historical medication robot to take medicine from the multiple medication patients according to the historical medication operation sequence. When there are multiple historical medication operation sequences, the sequence with the shortest historical medication cycle can be selected from the multiple historical medication operation sequences.

[0031] In some examples, the actual medication dispensing operation sequence, the actual medication dispensing cycle, the historical medication dispensing operation sequence, and the historical medication dispensing cycle can be stored in the cloud platform to facilitate the cloud platform to collect these data.

[0032] It should be noted that the present application does not specifically limit the execution order of S301 and S302. In some examples, S302 may be executed first and then S301, or S301 and S302 may be executed simultaneously.

[0033] S303: The cloud platform filters out candidate drug fetching operation sequences from the actual drug fetching operation sequence.

[0034] The sequence deviation value between the candidate drug fetching operation sequence and the historical drug fetching operation sequence is not less than a preset deviation value threshold, and the candidate drug fetching cycle of the candidate drug fetching operation sequence is less than the historical drug fetching cycle.

[0035] The actual drug fetching operation sequence can be the current drug fetching operation sequence, or the current drug fetching operation sequence and the previous drug fetching operation sequence. That is, the actual drug fetching operation sequence can be one sequence or multiple sequences. After the cloud platform collects the actual drug fetching operation sequence and the historical drug fetching operation sequence, it can filter out the candidate drug fetching operation sequence from the actual drug fetching operation sequence.

[0036] In some examples, the cloud platform can compare the actual drug pickup operation order and the historical drug pickup operation order in sequence, and then find the order in which the sequence deviation value between the actual drug pickup operation order and the historical drug pickup operation order is not less than a preset deviation value threshold, and the order in which the actual drug pickup cycle is less than the historical drug pickup cycle, as a candidate drug pickup operation order.

[0037] The cloud platform can perform feature extraction on the actual medication fetching operation sequence and the historical medication fetching operation sequence respectively, output the actual sequence vector and the historical sequence vector, and then calculate the similarity between the actual sequence vector and the historical sequence vector, characterize the sequence deviation value by the similarity, and characterize the preset deviation value threshold by the preset similarity threshold. When the similarity is not less than the preset similarity threshold, it is considered that the sequence deviation value is not less than the preset deviation value threshold.

[0038] S304: The cloud platform analyzes whether there is an optimized sequence associated with the candidate drug extraction operation sequence.

[0039] The cloud platform analyzes whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs the analysis log. When it is determined that the log representation has an optimized sequence associated with the candidate drug extraction operation sequence, S305 is executed, and when the analysis log representation does not have an optimized sequence associated with the candidate drug extraction operation sequence, S306 is executed.

[0040] The embodiments of the present application provide a plurality of ways of analyzing whether there is an optimized order, and the first one is introduced below.

[0041] The cloud platform can parse whether the preset ward where the POI is located, included in the candidate medication pickup operation sequence, has a preset patient. When the candidate medication pickup operation sequence includes that there is a preset patient in the preset ward where the POI is located, a parsing log with an optimized order associated with the candidate medication pickup operation sequence is output. When the candidate medication pickup operation sequence includes that there is no preset patient in the preset ward where the POI is located, a parsing log without an optimized order associated with the candidate medication pickup operation sequence is output.

[0042] Among them, the POI in the candidate medication operation sequence can be a patient who picks up medicine, and the preset ward can be customized. For example, the preset ward can be within a preset number of meters with the POI as the center of the circle. The preset patient may not be convenient for taking the medicine box, but it is convenient for walking. For example, the preset patient can be a patient with phlegm obstruction, a patient with blood stasis in the heart, a patient with qi stagnation in the heart and chest, or a patient with cold stagnation in the heart meridian; the preset patient can also be a Western medicine patient, including at least: rhinitis patients, sinusitis patients, and pharyngitis patients.

[0043] For example, Figure 2 As shown, the candidate medication operation sequence can be sequence J1 and sequence J2. In the overall sequence composed of sequence J1 and sequence J2, POIs include the first medication patient 101, the second medication patient 102, and the fourth medication patient 104. The cloud platform can analyze in sequence whether there are patients with phlegm turbidity obstruction, heart blood stasis, qi stagnation in the chest, or cold coagulation in the heart within 30 meters of the first medication patient 101. Obviously, there are no such preset patients within 30 meters of the first medication patient 101. Then, it is analyzed whether there are patients with phlegm turbidity obstruction, heart blood stasis, qi stagnation in the chest, or cold coagulation in the heart within 30 meters of the second medication patient 102. Figure 2 It can be seen that there is a patient with heart blood stasis within 30 meters of the second medication patient 102. Therefore, the cloud platform believes that there is an optimized sequence associated with the candidate medication operation sequence, and the optimized sequence is sequence F1 and sequence F2. Then the cloud platform continues to analyze the subsequent medication patients to analyze whether the preset ward where the medication patient is located includes the preset patients, which will not be repeated here.

[0044] It should be noted that the specific numerical values ​​in the above examples are merely illustrative and may be other numerical values ​​in other examples.

[0045] The following describes a second method for analyzing whether there is an optimized order associated with the candidate drug extraction operation order.

[0046] The cloud platform can collect the data of the first patient of the medicine picking robot and the second patient of the medicine picking box during the process of the medicine picking robot picking medicine according to the candidate medicine picking operation sequence, and then calculate the medicine picking similarity between the first patient and the second patient. When the medicine picking similarity between the first patient and the second patient exceeds the preset medicine picking similarity threshold, a parsing log with an optimized order associated with the candidate medicine picking operation sequence is output; when the medicine picking similarity between the first patient and the second patient does not exceed the preset medicine picking similarity threshold, a parsing log without an optimized order associated with the candidate medicine picking operation sequence is output.

[0047] Among them, when the medicine-taking robot takes medicine and the medicine-taking box takes medicine according to the candidate medicine-taking operation sequence, the cloud platform can monitor the first patient of the medicine-taking robot and the second patient of the medicine-taking box. For example, the wearable device, handheld device or other device capable of collecting the patient information of the medicine-taking robot transmits the patient information of the medicine-taking robot to the cloud platform, and the medicine-taking box is equipped with a positioning system such as GPS to transmit the patient information of the medicine-taking box to the cloud platform. The preset medicine-taking similarity threshold can be 30% or 40%, and those skilled in the art can set it based on actual needs.

[0048] For example, Figure 2 As shown, the candidate drug picking operation sequences can be sequence J1 and sequence J2. When the drug picking robot picks up drugs in sequence J1 and sequence J2, the cloud platform can monitor the first patient of the drug picking robot and the second patient of the medicine box. After the drug picking robot carries the medicine box in sequence J1 from the first drug picking patient 101 to the second drug picking patient 102 and completes picking up drugs for the second drug picking patient 102, the drug picking robot can walk through the green channel 201 to the third drug picking patient opposite in sequence F1 to pick up drugs. At this time, the drug picking similarity between the first patient of the drug picking robot and the second patient of the medicine box is 50%, which exceeds the preset drug picking similarity threshold of 30%. Then the cloud platform outputs a parsing log with an optimized sequence associated with the candidate drug picking operation sequence.

[0049] It should be noted that as long as the cloud platform verifies that the medication similarity between the first patient and the second patient exceeds the preset medication similarity threshold, it is considered that there is an optimized order.

[0050] Then, after the medicine robot completes the medicine collection for the third patient 103, it returns to the medicine collection box according to the sequence F2, and continues to carry the medicine collection box to the fourth patient 104. At the fourth patient 104, the medicine robot never leaves the medicine collection box. For example, the medicine collection similarity between the third patient 103 and the fourth patient 104 is 10%, which does not exceed the preset medicine collection similarity threshold of 30%. It can be seen that there is no optimized order in the process of collecting medicine for the fourth patient 104, and there is an optimized order in the process of collecting medicine for the third patient 103.

[0051] Among them, in this embodiment, the method for calculating the similarity of medicine collection is to segment the Chinese medicine prescription string of the medicine collection prescription to form a short string vector, obtain the degree of mutual correlation between the short string vectors; and calculate the prescription similarity based on the degree of mutual correlation between the short string vectors.

[0052] For example, the prescription of the third patient who takes medicine includes Codonopsis pilosula, Gardenia jasminoides, Bupleurum chinense, Citrus aurantium, Platycladus orientalis, Atractylodes macrocephala, Paeonia suffruticosa, Leonurus japonicus, White peony root and Pollen Typhae, and the prescription of the fourth patient who takes medicine includes Licorice, Angelica sinensis, Rubia cordifolia, Astragalus membranaceus, Codonopsis pilosula, Chuanxiong, Curcuma aromatica, Scorpio, Sodium sulfate and Camphor. The similarity of the medicine taken by the third patient 103 and the fourth patient 104 is 10% (only Codonopsis pilosula is the same), which does not exceed the preset threshold of 30% of the similarity of medicine taken. It can be seen that there is no optimized order in the process of taking medicine for the fourth patient 104, and there is an optimized order in the process of taking medicine for the third patient 103.

[0053] For example, when the first patient is a patient with rhinitis, his prescription includes budesonide nasal spray, loratadine, tranilast, oxymetazoline nasal drops, and bencyclidine nasal spray; When the second patient was a sinusitis patient, his prescriptions included budesonide nasal spray, cetirizine, azelastine nasal spray, tranilast, and oxymetazoline nasal drops; The similarity between the first patient and the second patient in the medicine box is 60% (Budesonide Nasal Spray, Tranilast, Oxymetazoline Nasal Drops), which exceeds the preset similarity threshold of 30%. The cloud platform then outputs a parsing log with an optimized order associated with the candidate drug collection run order. S305. The cloud platform generates a medication collection sequence associated with multiple medication collection patients based on the optimization sequence and the candidate medication collection operation sequence.

[0054] After outputting the parsing log with the optimized sequence associated with the candidate medication collection operation sequence, a medication collection sequence associated with multiple medication collection patients can be generated based on the optimized sequence and the candidate medication collection operation sequence, wherein the optimized sequence may be a non-medication collection operation sequence.

[0055] In some examples, after determining the optimized sequence associated with the candidate medication collection operation sequence, the cloud platform can determine the POI with the optimized sequence, such as the second medication collection patient 102. The staff can go to the second medication collection patient 102, record the sequence information of the optimized sequence, and store it in the cloud platform to facilitate the subsequent generation of the medication collection sequence. It is also possible to generate guidance information for the optimized sequence for the optimized sequence and store it in the cloud platform.

[0056] In some embodiments, the cloud platform may also transmit verification information to the processor, and the verification information is used to inform the medicine-taking robot to check the parsing log, for example, to check whether it has an optimized order. After receiving the verification information, the processor may provide a verification interface to the medicine-taking robot, and the medicine-taking robot may input "has an optimized order" or "does not have an optimized order" in the verification interface, and the processor feeds back the verification log of the medicine-taking robot to the cloud platform.

[0057] After the cloud platform receives the verification log fed back by the processor, it can comprehensively analyze the log and the verification log to analyze whether there is an optimized sequence, and then decide whether to remind the staff to go to the site for investigation. For example, if there is, go, otherwise don’t go. When both the analysis log and the verification log indicate an optimized sequence associated with the candidate medication collection operation sequence, a medication collection sequence associated with multiple medication collection patients is generated based on the optimized sequence and the candidate medication collection operation sequence. Through the double verification method, the rigor of the optimized medication collection sequence is increased to avoid misoptimization.

[0058] S306, the medicine dispensing robot performs identity verification on the patient when dispensing medicine; When the medicine-taking robot stops in front of the patient taking medicine, it scans the facial information of the patient taking medicine; The medicine dispensing robot compares the facial information with the information in the patient database to obtain the medicine prescription information of the corresponding patient; The drug-taking patient obtains the drug from the drug-taking robot by himself, and when the drug-taking robot detects that the drug taken by the drug-taking patient is inconsistent with the drug prescription information of the corresponding patient, an early warning is issued to the cloud platform; The medicine dispensing robot recycles the medicine dispensing drugs.

[0059] S307. The cloud platform uses the candidate medication collection operation sequence as the medication collection operation sequence associated with multiple medication collection patients.

[0060] In the case that the cloud platform analysis does not have an optimized sequence associated with the candidate drug collection operation sequence, it indicates that the candidate drug collection operation sequence is better than the historical drug collection operation sequence and takes less time to collect drugs. Therefore, the candidate drug collection operation sequence can be used as the drug collection operation sequence associated with multiple drug collection patients. Subsequent drug collection robots collect drugs according to the candidate drug collection operation sequence, which can reduce the time consumption of drug collection and improve the efficiency of drug collection.

[0061] S308. The cloud platform pushes guidance information to the processor based on the order of taking medicines.

[0062] After outputting the medicine fetching sequence, the cloud platform can push guidance information to the processor based on the medicine fetching sequence. The guidance information includes guidance information for portable medicine fetching boxes and guidance information for non-portable medicine fetching boxes, wherein the guidance information for portable medicine fetching boxes is generated by the candidate medicine fetching operation sequence, and the guidance information for non-portable medicine fetching boxes is generated by the optimized sequence.

[0063] S309. The processor provides animation information associated with the optimized sequence to the medicine-taking robot based on the guidance information.

[0064] After receiving the guidance information, the processor can provide the medicine-picking robot with animation information associated with the optimized sequence. The animation information can be a video, graphic, etc. The animation information is used to introduce the optimized sequence and how to pick up medicines according to the optimized sequence. After seeing the animation information, the medicine-picking robot can pick up medicines from the goods according to the optimized sequence, thereby reducing the time spent on picking up medicines and improving the efficiency of picking up medicines.

[0065] Based on the above description, the present application provides an intelligent drug collection identification method based on machine vision, which can be executed by a cloud platform. In this method, the cloud platform collects drug collection data, which includes multiple drug collection patients, collects the actual drug collection operation sequence, actual drug collection cycle, historical drug collection operation sequence, and historical drug collection cycle of the drug collection robot, and then the actual drug collection operation sequence can be compared with the historical drug collection operation sequence, and the sequence with a deviation value not less than a preset deviation value threshold is output, and then the sequence with an actual drug collection cycle less than the historical drug collection cycle is filtered out from these sequences, that is, the candidate drug collection operation sequence is output. Of course, the sequence with a short cycle can also be filtered first, and then the sequence with a large deviation value can be filtered, and then the candidate drug collection operation sequence is output. After the cloud platform determines the candidate drug collection operation sequence, it analyzes whether there is an optimized sequence associated with the candidate drug collection operation sequence. If it has, the drug collection sequence associated with the multiple drug collection patients is generated based on the optimized sequence and the candidate drug collection operation sequence. Therefore, the subsequent medication personnel can pick up medicines for these multiple patients in the order of picking up medicines, that is, pick up medicines by running the medicine box according to the actual order of picking up medicines, thereby reducing the medicine picking cycle of carrying the medicine box, and pick up medicines by picking up medicines without carrying the medicine box according to the optimized order, thereby reducing the overall medicine picking cycle. It can be seen that this method can reduce the medicine picking cycle and improve the efficiency of medicine picking.

[0066] Combination of the above Figures 1 to 3 The intelligent drug dispensing identification method based on machine vision provided in the embodiment of the present application is introduced in detail. The system and equipment provided in the embodiment of the present application will be introduced in conjunction with the accompanying drawings.

[0067] like Figure 4 As shown, this figure is a schematic diagram of an intelligent drug taking recognition system based on machine vision provided in an embodiment of the present application, and the system includes: The collection unit 401 is used to collect medication data, wherein the medication data includes multiple medication patients, and collects an actual medication operation sequence of the medication robot, an actual medication cycle of the actual medication operation sequence, a historical medication operation sequence associated with the medication data, and a historical medication cycle of the historical medication operation sequence; A filtering unit 402 is used to filter out a candidate drug fetching operation sequence from the actual drug fetching operation sequence, wherein the sequence deviation between the candidate drug fetching operation sequence and the historical drug fetching operation sequence is not less than a preset deviation value threshold, and the candidate drug fetching cycle of the candidate drug fetching operation sequence is less than the historical drug fetching cycle; The parsing unit 403 is used to analyze whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and output a parsing log; The optimization unit 404 is configured to generate a medication taking sequence associated with the plurality of medication taking patients based on the optimization sequence and the candidate medication taking sequence when the parsed log representation has an optimization sequence associated with the candidate medication taking sequence.

[0068] In some possible implementations, the parsing unit 403 is specifically used to output a parsing log having an optimized order associated with the candidate medication pickup operation sequence when the candidate medication pickup operation sequence includes a preset patient in the preset ward where the POI is located; and output a parsing log having no optimized order associated with the candidate medication pickup operation sequence when the candidate medication pickup operation sequence includes a preset patient in the preset ward where the POI is located.

[0069] In some possible implementations, the parsing unit 403 is specifically used to collect data of a first patient of the medicine picking robot and a second patient of the medicine picking box during the process of the medicine picking robot running and picking up medicines in accordance with the candidate medicine picking operation sequence; when the medicine picking similarity between the first patient and the second patient exceeds a preset medicine picking similarity threshold, a parsing log with an optimized order associated with the candidate medicine picking operation sequence is output; when the medicine picking similarity between the first patient and the second patient does not exceed the preset medicine picking similarity threshold, a parsing log without an optimized order associated with the candidate medicine picking operation sequence is output.

[0070] In some possible implementations, the optimization unit 404 is further configured to use the candidate medication taking operation sequence as the medication taking operation sequence associated with the multiple medication taking patients when the parsed log representation does not have an optimized sequence associated with the candidate medication taking operation sequence.

[0071] In some possible implementations, the optimization unit 404 is further used to push guidance information to the processor based on the medicine picking sequence, so that the processor provides the animation information associated with the optimization sequence to the medicine picking robot based on the guidance information.

[0072] In some possible implementations, the system also includes a transmission unit, which is used to transmit verification information to the processor after outputting a parsing log having an optimized order associated with the candidate drug taking operation order, so that the processor informs the drug taking robot to verify the parsing log based on the verification information, and feeds back the verification log to the cloud platform; the optimization unit 404 is specifically used to generate a drug taking order associated with the multiple drug taking patients based on the optimized order and the candidate drug taking operation order when both the parsing log and the verification log represent an optimized order associated with the candidate drug taking operation order.

[0073] In some possible implementations, the preset patients include patients with phlegm obstruction, patients with heart blood stasis, patients with qi stagnation in the heart and chest, or patients with cold stagnation in the heart meridians.

[0074] The intelligent drug-taking identification system based on machine vision according to the embodiment of the present application can be associated with the method described in the embodiment of the present application, and the above-mentioned other operations and / or functions of each unit / unit of the intelligent drug-taking identification system based on machine vision are to achieve Figure 3 For the sake of brevity, the corresponding processes of each method in the illustrated embodiment are not described in detail here.

[0075] The present application also provides a computing device. The computing device may be a cloud platform, which is specifically used to implement the following Figure 4 The functions of the intelligent medicine dispensing recognition system based on machine vision in the illustrated embodiment.

[0076] In the several embodiments provided in the present application, it should be understood that the disclosed systems, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or units, which can be electrical, mechanical or other forms.

[0077] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be physically provided separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent protection of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection of the present application.

[0078] The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. An intelligent drug-taking identification method based on machine vision, characterized in that: Applied to a cloud platform, the method includes: The cloud platform collects medication collection data, and the medication collection data includes multiple medication collection patients; The cloud platform collects the actual drug taking operation sequence of the drug taking robot, the actual drug taking cycle of the actual drug taking operation sequence, the historical drug taking operation sequence associated with the drug taking data, and the historical drug taking cycle of the historical drug taking operation sequence; The cloud platform filters out candidate drug fetching operation sequences from the actual drug fetching operation sequence, wherein the sequence deviation between the candidate drug fetching operation sequence and the historical drug fetching operation sequence is not less than a preset deviation value threshold, and the candidate drug fetching cycle of the candidate drug fetching operation sequence is less than the historical drug fetching cycle; The cloud platform analyzes whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs an analysis log; When the parsed log representation has an optimized sequence associated with the candidate medication pickup running sequence, the cloud platform generates a medication pickup sequence associated with the plurality of medication pickup patients based on the optimized sequence and the candidate medication pickup running sequence; The medicine dispensing robot performs identity verification when dispensing medicine for the patient.

2. The intelligent drug-taking identification method based on machine vision as claimed in claim 1, characterized in that: The analysis determines whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs a analysis log, including: When the candidate medication fetching operation sequence includes a preset patient in the preset ward where the POI is located, outputting a parsing log having an optimized sequence associated with the candidate medication fetching operation sequence; When the candidate medication fetching operation sequence includes a preset ward where the POI is located but there is no preset patient, a parsing log having no optimized sequence associated with the candidate medication fetching operation sequence is output.

3. The intelligent drug-taking identification method based on machine vision as claimed in claim 1, characterized in that: The analysis determines whether there is an optimized sequence associated with the candidate drug extraction operation sequence, and outputs a analysis log, including: The first patient of the medicine taking robot and the second patient of the medicine taking box are collected in the process of the medicine taking robot and the medicine taking box taking medicine according to the candidate medicine taking operation sequence; When the medication similarity between the first patient and the second patient exceeds a preset medication similarity threshold, outputting a parsing log having an optimized sequence associated with the candidate medication running sequence; When the medication similarity between the first patient and the second patient does not exceed a preset medication similarity threshold, a parsing log having no optimized sequence associated with the candidate medication running sequence is output.

4. The intelligent drug-taking identification method based on machine vision as claimed in claim 1, characterized in that: The method further comprises: When the parsed log representation does not have an optimized sequence associated with the candidate medication pickup operation sequence, the candidate medication pickup operation sequence is used as the medication pickup operation sequence associated with the multiple medication pickup patients.

5. The intelligent drug-taking identification method based on machine vision as claimed in claim 1, characterized in that: The method further comprises: Based on the medicine picking sequence, guidance information is pushed to the processor so that the processor provides animation information associated with the optimized sequence to the medicine picking robot based on the guidance information.

6. The intelligent drug-taking identification method based on machine vision as claimed in claim 2 or 3, characterized in that: After outputting the parsing log having the optimized sequence associated with the candidate medicine taking operation sequence, the method further includes: transmitting verification information to the processor, so that the processor informs the medicine taking robot to verify the parsing log based on the verification information, and feeds back the verification log to the cloud platform; When the parsed log representation has an optimized sequence associated with the candidate medication pickup running sequence, the cloud platform generates medication pickup sequences associated with the plurality of medication pickup patients based on the optimized sequence and the candidate medication pickup running sequence, including: When both the parsing log and the checking log indicate an optimized sequence associated with the candidate medication pickup running sequence, a medication pickup sequence associated with the plurality of medication pickup patients is generated based on the optimized sequence and the candidate medication pickup running sequence.

7. The method according to claim 2, characterized in that When the pre-set patient is a traditional Chinese medicine patient, it at least includes: Patients with phlegm obstruction, heart blood stasis, qi stagnation in the chest, or cold stagnation in the heart meridians; When the preset patients are western medicine patients, they at least include: rhinitis patients, sinusitis patients, and pharyngitis patients.

8. The method according to claim 1, characterized in that When the medicine dispensing robot dispensing medicine to the patient, identity verification is performed, and the following steps are also included: When the medicine-taking robot stops in front of the patient taking medicine, it scans the facial information of the patient taking medicine; The medicine dispensing robot compares the facial information with the information in the patient database to obtain the medicine prescription information of the corresponding patient; The drug-taking patient obtains the drug from the drug-taking robot by himself, and when the drug-taking robot detects that the drug taken by the drug-taking patient is inconsistent with the drug prescription information of the corresponding patient, an early warning is issued to the cloud platform; The medicine dispensing robot recycles the medicine dispensing drugs.

9. An intelligent drug-taking recognition system based on machine vision, characterized in that: A system for executing a machine vision-based intelligent medication identification method as described in any one of claims 1 to 8, the system comprising: A collection unit is used to collect medication data, wherein the medication data includes a plurality of medication patients, and collects an actual medication operation sequence of the medication robot, an actual medication cycle of the actual medication operation sequence, a historical medication operation sequence associated with the medication data, and a historical medication cycle of the historical medication operation sequence; A filtering unit, used to filter out a candidate drug fetching operation sequence from the actual drug fetching operation sequence, wherein the sequence deviation between the candidate drug fetching operation sequence and the historical drug fetching operation sequence is not less than a preset deviation value threshold, and the candidate drug fetching cycle of the candidate drug fetching operation sequence is less than the historical drug fetching cycle; A parsing unit, used for parsing whether there is an optimized sequence associated with the candidate drug taking operation sequence, and outputting a parsing log; The optimization unit is used for generating a medication taking sequence associated with the plurality of medication taking patients based on the optimization sequence and the candidate medication taking sequence when the parsed log representation has an optimization sequence associated with the candidate medication taking sequence.

Images