Remote dosing method based on real-time monitoring of subject dosing and smart drug matching

Abstract

The application discloses a remote drug administration method based on real-time monitoring and intelligent drug matching of subject administration, and the method comprises the following steps: collecting image data information matched with a target detection requirement corresponding to a target user; sending the image data information to a management end, so that the management end performs image analysis on the image data information to generate first guidance information; receiving the first guidance information fed back by the management end, and performing data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information; and performing dynamic adjustment of drugs based on the second guidance information. Through the above technical adjustment, real-time distribution of remotely controlled drugs can be realized, and the real-time performance and reliability of drug distribution and monitoring are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital information transmission. More particularly, it relates to a remote drug administration method based on real-time monitoring of subject drug administration and intelligent drug matching. BACKGROUND

[0002] With the continuous development of medical technology, remote medical treatment has gradually become an important medical model. Traditional treatment methods mainly rely on patients going to the hospital for face-to-face diagnosis, and doctors prescribing drugs after observation and diagnosis. However, this method has many limitations: on the one hand, patients need to frequently go back and forth to the hospital, consuming a lot of time and effort; on the other hand, doctors cannot monitor the patient's condition changes and treatment effects in real time, making it difficult to adjust the treatment plan in a timely manner.

[0003] Although the existing remote medical technology can alleviate the inconvenience of patients going to the hospital to some extent, it still has deficiencies in drug administration monitoring and drug matching. At present, some remote medical systems can realize video calls between patients and doctors, and doctors can observe the scalp condition of patients through video, but this method cannot provide high-precision image data, making it difficult to meet the needs of precision medicine. Moreover, for drug matching, it mainly depends on the experience of doctors, lacks intelligent analysis and matching means, and it is difficult to dynamically adjust the drug plan according to the specific condition of the patient. SUMMARY

[0004] In order to solve the deficiencies in the prior art, the purpose of the present application is to solve the above-mentioned defects, and further to propose a remote drug administration method based on real-time monitoring of subject drug administration and intelligent drug matching.

[0005] According to an aspect of the present application, a remote drug administration method based on real-time monitoring of subject drug administration and intelligent drug matching is disclosed, comprising:

[0006] Collecting image data information matched with the target detection demand corresponding to the target user;

[0007] Sending the image data information to the management end to make the management end perform image analysis on the image data information and generate first guidance information;

[0008] Receiving the first guidance information fed back by the management end, and performing data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information;

[0009] Performing dynamic adjustment of drugs based on the second guidance information.

[0010] In some possible embodiments, receiving the first guidance information fed back by the management end, and performing data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information, comprises:

[0011] constructing a first directed graph corresponding to the data analysis model; wherein each calculation in the data analysis model corresponds to a node in the first directed graph;

[0012] splitting the first directed graph into a first part and a second part;

[0013] The management end performs data analysis on the first guidance information based on the first part of the data analysis model to obtain third data information; and transmits the encrypted third data information to the client;

[0014] The client performs data analysis on the third data information based on the second part of the data analysis model to obtain the second guidance information.

[0015] In some possible embodiments, based on the first guidance information, the node access order and the data analysis model, the second guidance information is generated, including:

[0016] determining a target first guidance information from the plurality of first guidance information sent by the management end based on the node access order;

[0017] Based on the retake parameter information and the retake area information contained in the target first guidance information, the update detection image is collected;

[0018] Based on the data analysis model, the update detection image and the remaining detection image are analyzed to determine the second guidance information; the remaining detection image is a detection image in the plurality of detection images, except for the to-be-retaken image.

[0019] In some possible embodiments, based on the data analysis model, the update detection image and the remaining detection image are analyzed to determine the second guidance information, including:

[0020] generating a multi-channel sequence based on the image parameter information of the update detection image and the remaining detection image;

[0021] Based on the multi-channel sequence and the data analysis model, spatial feature information and time sequence feature information corresponding to the target detection requirement are extracted;

[0022] Based on the spatial feature information and the time sequence feature information, feature fusion is performed to obtain feature fusion information; the feature fusion information has a mapping relationship with the drug candidate data set;

[0023] Based on the feature fusion information and the drug candidate data set, the second guidance information is determined.

[0024] In some possible embodiments, the method further comprises:

[0025] obtaining standard drug distribution information;

[0026] Based on the standard drug dispensing information and the second guidance information, the data analysis model is corrected.

[0027] A second aspect of the present application discloses another remote drug administration method based on real-time monitoring of subject drug administration and intelligent drug matching, comprising:

[0028] Receiving image data information collected by the client;

[0029] Image analysis is performed on the image data information to generate first guidance information;

[0030] The first guidance information is sent to the client, so that the client performs data analysis on the first guidance information based on a data analysis model to obtain second guidance information, and dynamically adjusts the drug based on the second guidance information.

[0031] In some possible embodiments, the image analysis on the image data information to generate the first guidance information comprises:

[0032] Based on the analysis index data, the image data information is analyzed to obtain an image analysis result;

[0033] In a case where the image analysis result indicates that, among the multiple detection images, analysis index data corresponding to at least one detection image does not satisfy a preset analysis condition, the at least one detection image is determined as a to-be-detected image;

[0034] The retake parameter information and the retake region information corresponding to the to-be-detected image are taken as the first guidance information.

[0035] A third aspect of the present application provides a remote drug administration device based on real-time monitoring of subject drug administration and intelligent drug matching, comprising:

[0036] The data acquisition module is configured to acquire image data information matched with a target detection requirement corresponding to a target user;

[0037] The first data sending module is configured to send the image data information to the management end, so that the management end performs image analysis on the image data information to generate first guidance information;

[0038] The first data receiving module is configured to receive the first guidance information fed back by the management end, and perform data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information;

[0039] The drug selection module is configured to dynamically adjust the drug based on the second guidance information.

[0040] A fourth aspect of the present application provides another remote drug administration device based on real-time monitoring of subject drug administration and intelligent drug matching, comprising:

[0041] The second data receiving module is configured to receive image data information collected by the client;

[0042] The image analysis module is configured to perform image analysis on the image data information to generate first guidance information;

[0043] The second data sending module is configured to send the first guidance information to the client, so that the client performs data analysis on the first guidance information based on the data analysis model to obtain second guidance information, and dynamically adjusts the medicine based on the second guidance information.

[0044] A fifth aspect of the present application discloses an electronic device, which comprises a processor and a memory, the memory storing at least one instruction and at least one program, the at least one instruction and the at least one program being loaded and executed by the processor to implement the remote medicine administration method based on real-time monitoring of subject medicine administration and intelligent medicine matching as above.

[0045] A sixth aspect of the present application discloses a computer storage medium, which stores at least one instruction and at least one program, the at least one instruction and the at least one program being loaded and executed by a processor to implement the remote medicine administration method based on real-time monitoring of subject medicine administration and intelligent medicine matching as above.

[0046] Compared with the prior art, the present application has the following advantages:

[0047] The remote medicine administration method based on real-time monitoring of subject medicine administration and intelligent medicine matching provided by the present application realizes efficient and accurate identification and feedback of the user's medicine administration state by collecting image data information matched with the detection needs of the target user and combining the collaborative analysis of the management end and the local end, thereby realizing remote adjustment of the medicine use and dosage adjustment of the target user and improving the convenience, real-time performance and efficiency of the adjustment. On the basis of the preliminary image analysis of the management end to generate first guidance information, further deep analysis of the image information by the data analysis model deployed locally by the client can avoid leakage of user privacy in the data transmission process and improve the safety of the medicine use and dosage adjustment of the target user. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 is a system architecture diagram corresponding to the remote medicine administration method based on real-time monitoring of subject medicine administration and intelligent medicine matching provided by the embodiments of the present application;

[0049] Figure 2 is a flowchart corresponding to the remote medicine administration method based on real-time monitoring of subject medicine administration and intelligent medicine matching provided by the embodiments of the present application;

[0050] Figure 3is a flowchart of another remote drug administration method for real-time monitoring of subject drug administration and intelligent drug matching provided by an embodiment of the present application;

[0051] Figure 4 is a structural diagram of a remote drug administration device for real-time monitoring of subject drug administration and intelligent drug matching provided by an embodiment of the present application;

[0052] Figure 5 is a structural diagram of another remote drug administration device for real-time monitoring of subject drug administration and intelligent drug matching provided by an embodiment of the present application. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0054] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product, or device.

[0055] Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numbers in the drawings represent functionally the same or similar elements. Although various aspects of the embodiments are illustrated in the drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

[0056] The word "exemplary" is used herein in the sense of being an example, illustration, or demonstration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

[0057] The term "and / or", used in the present document, only describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" in the present document means any one of a plurality of combinations or any combination of at least two of a plurality of combinations, for example, including at least one of A, B and C can mean including any one or more elements selected from the set consisting of A, B and C.

[0058] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the specific embodiments below. Those skilled in the art should understand that the present disclosure can also be implemented without some specific details. In some examples, methods, means, elements and circuits that are well known to those skilled in the art are not described in detail, in order to highlight the main idea of the present disclosure.

[0059] Figure 1 The system architecture diagram corresponding to the remote drug delivery method based on real-time monitoring of subject drug delivery and intelligent drug matching provided by the embodiments of the present application. The remote drug delivery method based on real-time monitoring of subject drug delivery and intelligent drug matching disclosed in the present application can be applied in a multi-end interaction process. Specifically, as shown in Figure 1 The management end is used to record and store image data information, and to track and record the entire disease and drug delivery process of a single patient in real time, so as to remotely guide and summarize the case, and form first guidance information. The first guidance information is used to guide the client to use LED light sources of different frequency bands to retake the local monitoring area of the target object, so as to obtain more accurate raw data.

[0060] Among them, the management end can adopt a star-shaped hierarchical structure, that is, a central management end uniformly coordinates the rest of the management ends as slave nodes to participate in data exchange, and the specific network topology structure is determined according to the actual deployment environment. Since the management end adopts a star-shaped hierarchical structure, the disease course tracking chain needs to ensure the consistency of data in each slave management end.

[0061] The client can be an application program loaded on a mobile terminal, or a multispectral light source acquisition terminal. The multispectral light source acquisition terminal includes a multispectral LED light source array and a communication unit. The multispectral LED light source array is used to provide different LED light sources in the image data information acquisition process. The communication unit supports wired or wireless transmission protocols, and is used for data communication with the management end, processing the received data, and feeding back second guidance information to the target user.

[0062] The multispectral LED light source array can include white light LED, red light LED, blue light LED, and the like; and the multispectral LED light source array is arranged in a matrix. Each LED unit can be independently controlled to ensure imaging detail optimization under different wavebands. The light intensity value of the LED is set according to specific detection requirements. For example, the white light LED simulates natural light under a color temperature of 6500K according to the D65 standard (daylight standard), and the brightness is controlled at 500-1500 lm / m²; the output power of the ultraviolet LED (365-405 nm) should be strictly controlled to ensure radiation safety. The current industry standard recommends that the ultraviolet radiation does not exceed 15 µW / cm² to ensure human safety.

[0063] In the embodiment of the present application, the collection area corresponding to the target detection requirement of the target user is taken as the scalp as an example. Specifically, the white light LED simulates standard daylight to ensure the overall brightness and color restoration of the image, which facilitates the subsequent algorithm to distinguish the color boundary between the scalp and the hair; the red light LED (620-750 nm) is used to compare the bleeding phenomenon of the scalp and the hair base; the blue light LED (450-495 nm) can enhance the surface details and is suitable for distinguishing the exposed part of the scalp and the covered hair; the green light LED (500-570 nm) is extremely sensitive to the fiber structure, color and diameter of the hair and is used to realize accurate detection of the hair and scalp boundary and hair density; the ultraviolet LED (365-405 nm) is used to enhance the imaging of the hair surface, the horny layer and the fine cracks (such as the hair scale) under low power state; the near-infrared LED is used for subscalp hair follicle and microcirculation monitoring. The near-infrared light has strong penetration ability and is sensitive to the internal tissue spectrum response of the hair.

[0064] In the embodiment, a dual-band near-infrared design can be used. The short waveband (750-850 nm) is suitable for detecting the subscalp structure; and the long waveband (850-950 nm) can penetrate the hair fiber to form an image and distinguish the hair shaft and the hair follicle connection part.

[0065] Figure 2 is a process schematic diagram of a remote dosing method for real-time monitoring of subject dosing and intelligent drug matching provided by the embodiment of the present application. The execution subject is a client, such as Figure 2 As shown in the remote dosing method for real-time monitoring of subject dosing and intelligent drug matching, the method comprises the following steps:

[0066] Step S201: Collect image data information matched with the target detection requirement corresponding to the target user.

[0067] Specifically, the target detection requirement can be related to the disease of the target user, and the physiological change indicators that can be detected and the corresponding detection regions. In this embodiment, the detection region in the target detection requirement is taken as the scalp region of the target user as an example, and the physiological change indicators detected can include hair follicle opening density, hair scale arrangement state, hair follicle entrance contour, etc.

[0068] The image data information can include multiple detection images collected based on the client, or can be a video image, etc. The multiple detection images can correspond to different detection regions. In this embodiment, the client selects a multi-spectral light source including a white light LED, a red light LED, and a blue light LED, etc. to collect the terminal.

[0069] Step S202: sending the image data information to the management end, so that the management end performs image analysis on the image data information to generate first guidance information.

[0070] In some embodiments, the client sends the collected image data information to the management end through the communication unit, so that the management end receives the image data information and performs image analysis, and generates first guidance information. The first guidance information is used to represent that there are images to be retaken in the multiple detection images, and includes retake parameter information and retake region information corresponding to the images to be retaken.

[0071] Specifically, the multiple detection images in the image data information can have low collection accuracy or have deviations in the collection region, etc., so that the detection images cannot complete the reaction of the real physiological state of the target user. Therefore, the first guidance information can be generated by respectively performing image analysis on the multiple detection images, so as to determine which regions in the detection images need to be retaken.

[0072] Step S203: receiving the first guidance information fed back by the management end, and performing data analysis on the image data information based on the data analysis model and the first guidance information to obtain second guidance information.

[0073] In some embodiments, since there is a certain data leakage risk in the process of data interaction between the management end and the client, the first guidance information generated by the management end cannot directly indicate how to dynamically adjust the use of the drug, but the client combines the first guidance information and the image data information, and uses the data analysis model to process them to obtain the second guidance information. The second guidance information is used to dynamically adjust the drug, which can include drug selection information and drug amount adjustment information, for example.

[0074] In one specific embodiment, the first guidance information can not correspond to the second guidance information because the first guidance information corresponds to a different calculation Lenz than the second guidance information, thereby causing an error in the dynamic allocation of the drug. To solve this problem, i.e., to improve the accuracy of the generation of the second guidance information, the following method can be used:

[0075] That is, receiving the first guidance information fed back by the management end, and performing data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information, comprising:

[0076] Constructing a first directed graph corresponding to the data analysis model; wherein each calculation unit in the data analysis model corresponds to a node in the first directed graph;

[0077] Splitting the first directed graph into a first part and a second part;

[0078] The management end performs data analysis on the first guidance information based on the first part of the data analysis model to obtain third data information; and transmits the encrypted third data information to the client;

[0079] The client performs data analysis on the third data information based on the second part of the data analysis model to obtain the second guidance information.

[0080] Since the data analysis model needs to capture both temporal features and spatial features, in some embodiments, the data analysis model can select a ConvLSTM network. Since the ConvLSTM network is an existing algorithm, its core idea will not be described in detail. In addition, it should be understood that the first directed graph is essentially the data analysis model itself. It can be understood that the data analysis model is essentially a string of code, and the process of drawing the internal calculation units (equivalent to nodes) and their connection relationships (equivalent to edges) into the first directed graph is the process of construction.

[0081] It can be understood that the third data information is essentially a collection of data information of all output nodes of the first part. However, the data information of different output nodes may not be consistent in terms of temporal sequence, which may lead to instability of the final second guidance information.

[0082] Based on this, in one specific embodiment, when the first directed graph is constructed, the loops in it are replaced with loop nodes to obtain a loop-free second directed graph. Assuming that there are edges: {a1->a2; a2->a3; a3->a1; a4->a2; a3->a5}, {a1, a2, a3} is replaced with loop node A1, while inheriting its connection relationship with other nodes, which is transformed into {a4->A1; A1->a5}.

[0083] The node corresponding to the original data is taken as a source point to perform a depth-first search. It can be understood that, since the original data is multi-dimensional data (equivalent to multiple source nodes), a total source node can be virtually set to point to the multiple source nodes, and then the total source node is taken as a source point.

[0084] It is assumed that the second directed graph contains n nodes in total. In the process of the depth-first search, each node is accessed exactly twice, and a corresponding access sequence is generated in order. In the access sequence, the numbers correspond to the numbers of the nodes, and the positions thereof determine the order of access, i.e., the order of node access.

[0085] Further, the access sequence is split according to the order of access of the ring nodes, so as to determine the first part and the second part. Specifically, it is assumed that n = 11, and it is assumed that node 2, node 7 and node 9 are ring nodes, and it is assumed that the access sequence is [1, 2, 3, 3, 2, 1, 4, 5, 5, 4, 6, 7, 7, 8, 8, 6, 9, 10, 10, 11, 11, 9]. Then the nodes involved in the second part can be 3, 10 and 11. Among them, the ring nodes can only be in the second part (for example, nodes 10 and 11 are wrapped by the ring node 9; node 3 is wrapped by the ring node 2); the nodes wrapped by the ring nodes and other nodes can only be in the first part. The nodes "wrapped" can also be in the first part.

[0086] Through the setting of the above technical features, the third data information is taken as a substitute for transmission, so as to ensure data privacy.

[0087] Further, the second guidance information is generated based on the first guidance information, the node access order and the data analysis model, and includes:

[0088] The target first guidance information is determined from the multiple first guidance information sent by the management end based on the node access order;

[0089] The update detection image is collected based on the retake parameter information and the retake area information contained in the target first guidance information;

[0090] The second guidance information is determined by performing data analysis on the update detection image and the remaining detection images based on the data analysis model; the remaining detection images are the detection images other than the to-be-retaken image in the multiple detection images.

[0091] In a specific embodiment, the target first guidance information is first determined according to the node access order, so as to ensure that there is relevance between the second guidance information generated based on the first target guidance information.

[0092] Specifically, the first target guidance information includes the retouching parameter information and the retouching area information, and then the client updates the acquisition of the detection image based on the retouching parameter information and the retouching area information. In a multi-spectral retouching, the system automatically identifies and quantifies the following image features:

[0093] Hair follicle opening density: less than 20 / cm² on average in the visible area;

[0094] UV-induced fluorescence: under UV (365 nm), the significantly thickened keratin layer area, and the irregular arrangement of hair scales;

[0095] Red light bleeding signs: under 620-750 nm red light, some areas appear fine blood vessel bleeding signals;

[0096] Blue light surface texture: under 450-495 nm blue light, the depth of the fine lines on the scalp surface increases, and the outline of the hair follicle entrance is blurred;

[0097] Near-infrared transmission degree: 750-850 nm short waveband shows local insufficient microcirculation perfusion around the hair follicle.

[0098] When the above image features meet their corresponding requirements, the updated detection image and the remaining detection image are input into the data analysis model to generate the second guidance information.

[0099] Based on the data analysis model, the updated detection image and the remaining detection image are analyzed to determine the second guidance information, including:

[0100] Generating a multi-channel sequence based on the image parameter information of the updated detection image and the remaining detection image;

[0101] Based on the multi-channel sequence and the data analysis model, spatial feature information and time sequence feature information corresponding to the target detection requirement are extracted;

[0102] Based on the spatial feature information and the time sequence feature information, feature fusion is performed to obtain feature fusion information; the feature fusion information has a mapping relationship with the drug candidate data set;

[0103] Based on the feature fusion information and the drug candidate data set, the second guidance information is determined.

[0104] In one specific embodiment, the image parameter information can include the image acquisition time. According to the acquisition time of each detection image, the detection images of different wavebands (white light, red light, blue light, green light, near-infrared, etc.) are integrated into a multi-channel sequence.

[0105] According to the order in the multi-channel sequence, each frame of the detection image is filtered and down-sampled using a convolutional layer to extract spatial feature information of the hair and scalp. The spatial feature information includes spatial texture information and structural feature information. The spatial feature information is input into a time sequence network unit in the data analysis model to extract time sequence feature information. The time sequence feature information represents the dynamic changes of the scalp condition at different time points.

[0106] The spatial feature information and the time sequence feature information are fused to generate feature fusion information through a fully connected layer mapping to a drug candidate data set. The optimal drug and its dose recommendation are output as the second guidance information through a regression layer.

[0107] In a specific embodiment, in the multi-channel sequence, each detection image can be pre-processed. For example, the image pixel values are normalized to the range of [0, 1] or [-1, 1] through normalization processing to speed up network training and improve numerical stability, or the generalization ability of the model is enhanced through data enhancement processing, and the processed detection image is organized into multi-channel input according to different wavebands, such as white light, red light, blue light, etc.

[0108] Further, each frame of image is input into a convolutional layer, which includes multiple convolutional kernels. Each convolutional kernel slides on the image to perform convolution operation. The size of the convolutional kernel can be selected as 3x3, 5x5, etc., and the convolution step is set to determine the step distance of the convolution kernel sliding on the image. Smaller convolutional kernels can capture more fine local features, and larger convolutional kernels can capture more extensive context information.

[0109] The spatial feature information extracted through spatial feature extraction is input into a time sequence network unit. The time sequence network unit can capture the time dependence in the sequence data, remember long-term information, and thus capture the detection change state at different time points. The time sequence feature information output by the time sequence feature unit is fused with the corresponding spatial feature information. Usually, a concatenate operation is used to concatenate the two in the channel dimension to generate the fused feature representation, i.e., the fused feature information.

[0110] The fused feature information output by the fully connected layer can generate the optimal drug and the corresponding dose recommendation through a Softmax layer or a regression layer. The Softmax layer can convert the output of the fully connected layer into a probability distribution, representing the probability of selecting different drugs. The drug with the highest probability is selected as the recommended drug. The regression layer can directly output the continuous value of the drug dose, which is suitable for the case where the drug dose needs to be accurately controlled.

[0111] Step S204: dynamically adjusting the drug based on the second guidance information.

[0112] In some embodiments, after the second guidance information is generated, the type of the drug used by the target user, the amount of the drug used, and the like can be adjusted according to the second guidance information, so as to improve the real-time performance and accuracy of the drug adjustment.

[0113] The method further includes:

[0114] Obtaining standard drug dispensing information;

[0115] Based on the standard drug dispensing information and the second guidance information, the data analysis model is corrected.

[0116] In some embodiments, after the drug dispensing is adjusted according to the second guidance information, the data analysis model can be corrected in combination with the standard drug dispensing information provided by the medical staff and the second guidance information, so as to adjust the parameters of the data analysis model and improve the accuracy of the data analysis model, thereby improving the accuracy of the dynamic adjustment of the drug.

[0117] Figure 3 is another process schematic diagram corresponding to the remote drug dispensing method based on real-time monitoring of subject drug dispensing and intelligent drug matching provided by the embodiments of the present application. The execution subject is a management end, such as a server. Figure 3 The remote drug dispensing method based on real-time monitoring of subject drug dispensing and intelligent drug matching includes:

[0118] Step S301: receiving image data information collected by a client;

[0119] Step S302: performing image analysis on the image data information to generate first guidance information;

[0120] Each detection image corresponds to analysis index data matched with the target detection requirement;

[0121] The image analysis on the image data information to generate the first guidance information includes:

[0122] Based on the analysis index data, the image analysis on the image data information is performed to obtain an image analysis result;

[0123] In a case where the image analysis result indicates that, among the multiple detection images, the analysis index data corresponding to at least one detection image does not meet a preset analysis condition, the at least one detection image is determined as a to-be-detected image;

[0124] The retake parameter information and the retake region information corresponding to the to-be-detected image are taken as the first guidance information.

[0125] Step S303: sending the first guidance information to the client, so that the client performs data analysis on the first guidance information based on a data analysis model to obtain second guidance information, and performs dynamic adjustment of a drug based on the second guidance information.

[0126] In some embodiments, data communication can be achieved between the management end and the client to transmit image data information and the first guidance information.

[0127] Taking a scalp image as an example, the detection image included in the image data information is subjected to image index analysis to determine whether the detection image meets the accuracy requirements, position requirements, etc. for subsequent data processing. Specifically, different target detection requirements can correspond to different analysis index data. The preset analysis condition can be set according to different analysis index data. For example, when the detection area needs to be adjusted, the preset analysis condition can be the position information corresponding to the preset detection area. When the LED light intensity parameter in the detection process needs to be adjusted, the preset analysis condition can be the light intensity range.

[0128] In some embodiments, the first guidance data can correspond to different parameters under different cases. For example: for alopecia areata cases, it is recommended to locally increase the blue light output (blue light brightness is adjusted to about 350 lm / m²) to enhance the local texture and hair follicle details of the scalp. For diffuse hair loss cases, it is recommended to appropriately reduce the ultraviolet LED output and enhance the red light LED (red light output is adjusted to a medium brightness level) to avoid local high reflection interference. In addition, the specific parameter settings of each LED should also be specified in the guidance data information, for example: for sparse hair areas, it is recommended to increase the green light LED light intensity by 10%-20% to more clearly capture the hair details.

[0129] In summary, in the present application, the remote administration method based on real-time monitoring of subject administration and intelligent drug matching is provided. By collecting image data information matched with the target user's detection requirements and combining the collaborative analysis of the management end and the local end, efficient and accurate identification and feedback of the user's administration state are realized, and remote adjustment of the target user's drug use and dosage adjustment is realized, and the convenience, real-time and efficiency of the adjustment are improved. On the basis of the preliminary image analysis of the management end to generate the first guidance information, further combining the data analysis model deployed locally by the client to deeply analyze the image information can avoid the leakage of user privacy in the data transmission process and improve the safety of the target user's drug use and dosage adjustment.

[0130] The present application also provides a remote administration device based on real-time monitoring of subject administration and intelligent drug matching, as described above. Figure 4 The device comprises:

[0131] The data acquisition module 410 is configured to acquire image data information matched with the target user's detection requirements.

[0132] The first data sending module 420 is configured to send image data information to the management end, so that the management end performs image analysis on the image data information and generates first guidance information.

[0133] The first data receiving module 430 is configured to receive the first guidance information fed back by the management end, and perform data analysis on the image data information based on the data analysis model and the first guidance information, to obtain second guidance information.

[0134] The drug selection module 440 is configured to perform dynamic adjustment of a drug based on the second guidance information.

[0135] In some possible embodiments, the first data receiving module 430 further includes:

[0136] The first directed graph construction module is configured to construct a first directed graph corresponding to the data analysis model, wherein each calculation unit in the data analysis model corresponds to a node in the first directed graph.

[0137] The second directed graph construction module is configured to replace ring nodes in the first directed graph to obtain a second directed graph.

[0138] The order determination module is configured to perform deep search with a node corresponding to the image data information in the second directed graph as a source point, to determine a node access order.

[0139] The first information generation module is configured to generate the second guidance information based on the first guidance information, the node access order, and the data analysis model.

[0140] In some possible embodiments, the first information generation module further includes:

[0141] The information determination module is configured to determine target first guidance information from a plurality of first guidance information sent by the management end based on the node access order.

[0142] The image updating module is configured to collect an updated detection image based on the retake parameter information and the retake region information included in the target first guidance information.

[0143] The second information generation module is configured to perform data analysis on the updated detection image and a remaining detection image based on the data analysis model, to determine second guidance information; the remaining detection image is a detection image other than the to-be-retaken image in the plurality of detection images.

[0144] In some possible embodiments, the second information generation module further includes:

[0145] The sequence construction module is configured to generate a multi-channel sequence based on image parameter information of the updated detection image and the remaining detection image.

[0146] The feature extraction module is configured to extract spatial feature information and time sequence feature information corresponding to the target detection requirement based on the multi-channel sequence and the data analysis model.

[0147] The feature fusion module is configured to perform feature fusion based on the spatial feature information and the time sequence feature information to obtain feature fusion information.

[0148] The third information generation module is configured to determine second guidance information based on the feature fusion information and the drug candidate data set.

[0149] In some embodiments, the apparatus further comprises:

[0150] The allocation information acquisition module is configured to acquire standard drug allocation information.

[0151] The model correction module is configured to correct the data analysis model based on the standard drug allocation information and the second guidance information.

[0152] The embodiments of the present application also provide another remote drug delivery device based on real-time monitoring of subject drug delivery and intelligent drug matching, as shown in Figure 5 The second data receiving module 510 is configured to receive image data information collected by the client.

[0153] The second data sending module 530 is configured to send the first guidance information to the client, so that the client performs data analysis on the first guidance information based on the data analysis model to obtain second guidance information, and performs dynamic drug adjustment based on the second guidance information.

[0154] The image analysis module 520 is configured to perform image analysis on the image data information to generate first guidance information.

[0155] The second data sending module 530 is configured to send the first guidance information to the client, so that the client performs data analysis on the first guidance information based on the data analysis model to obtain second guidance information, and performs dynamic drug adjustment based on the second guidance information.

[0156] In some embodiments, the image analysis module 520 further comprises:

[0157] The index analysis module is configured to perform image analysis on the image data information based on the analysis index data to obtain image analysis results.

[0158] The image determination module is configured to determine at least one detection image as a to-be-detected image in a case where the image analysis results indicate that the analysis index data corresponding to at least one detection image in the plurality of detection images does not satisfy the preset analysis condition.

[0159] The fourth information generation module is configured to take the retake parameter information and the retake region information corresponding to the to-be-detected image as the first guidance information.

[0160] With regard to the apparatus in the above-described embodiments, wherein a specific manner has been described in which the respective means perform the operations, the embodiments of the method are likewise applicable to the apparatuses.

[0161] Embodiments of the present application further provide an electronic device, the device comprising: a processor and a memory, the memory having stored therein at least one instruction, at least one program, a code set or an instruction set, the at least one instruction, the at least one program, the code set or the instruction set being loaded and executed by the processor to implement the method of any of the method embodiments.

[0162] Embodiments of the present application further provide a storage medium, the computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium can be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium include the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch cards or punched tape, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0163] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0164] Computer readable program instructions for carrying out operations of the present disclosure can be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate array (FPGA), or programmable logic array (PLA) can execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0165] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0166] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0167] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer, other programmable data processing apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0168] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer, other programmable data processing apparatus, or other device implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0169] Finally, it should be noted that the above-described embodiments are merely intended for describing and illustrating, but not limiting, the technical solutions of the present application. Although the present application has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application.

Claims

1. A remote dosing method based on real-time monitoring of subject dosing and smart drug matching, characterized in that, The method comprises: Collecting image data information matched with a target detection requirement corresponding to a target user; Sending the image data information to a management end to enable the management end to perform image analysis on the image data information to generate first guidance information; the first guidance information is used to guide the client to use LED light sources of different frequency bands to retake a local monitoring area of a target object; a multi-spectrum LED light source array comprises a white light LED, a red light LED and a blue light LED; Receiving the first guidance information fed back by the management end, and performing data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information; Performing dynamic adjustment of a drug based on the second guidance information; The receiving of the first guidance information fed back by the management end and the data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information comprises: Constructing a first directed graph corresponding to the data analysis model; each calculation unit in the data analysis model corresponds to a node in the first directed graph; Splitting the first directed graph into a first part and a second part; The management end performs data analysis on the first guidance information based on the first part of the data analysis model to obtain third data information; and transmits the third data information to the client after encryption; The client performs data analysis on the third data information based on the second part of the data analysis model to obtain the second guidance information; The data analysis model is a ConvLSTM network.

2. The remote dosing method for real-time monitoring and smart medicine matching based on subject dosing as claimed in claim 1, wherein, The image data information comprises a plurality of detection images corresponding to a plurality of detection regions respectively, and the first guidance information is used to represent that there is a to-be-retaken image in the plurality of detection images, and comprises retake parameter information and retake region information corresponding to the to-be-retaken image; Based on the first guidance information, the node access order and the data analysis model, the second guidance information is generated, which comprises: Determining target first guidance information from a plurality of first guidance information sent by the management end based on the node access order; Collecting updated detection images based on the retake parameter information and the retake region information contained in the target first guidance information; Performing data analysis on the updated detection images and remaining detection images based on the data analysis model to determine the second guidance information; the remaining detection images are detection images other than the to-be-retaken image in the plurality of detection images.

3. The remote dosing method for real-time monitoring and smart medicine matching based on subject dosing as claimed in claim 2, wherein, The data analysis on the updated detection images and the remaining detection images based on the data analysis model to determine the second guidance information comprises: Generating a multi-channel sequence based on image parameter information of the updated detection images and the remaining detection images; Extracting spatial feature information and time sequence feature information corresponding to the target detection requirement based on the multi-channel sequence and the data analysis model; Performing feature fusion based on the spatial feature information and the time sequence feature information to obtain feature fusion information; the feature fusion information has a mapping relationship with a drug candidate data set; determine the second guidance information based on the feature fusion information and the drug candidate data set.

4. The remote dosing method for real-time monitoring and smart medicine matching based on subject dosing as claimed in claim 1, wherein, The method further comprises: acquiring standard drug dispensing information; correcting the data analysis model based on the standard drug dispensing information and the second guidance information.

5. The remote dosing method for real-time monitoring and smart medicine matching based on subject dosing as claimed in claim 1, wherein, The method comprises: receiving image data information collected by a client; performing image analysis on the image data information to generate first guidance information; sending the first guidance information to the client, so that the client performs data analysis on the first guidance information based on a data analysis model to obtain second guidance information, and dynamically adjusts the drug based on the second guidance information.

6. The remote dosing method for real-time monitoring and smart medicine matching based on subject dosing as claimed in claim 5, wherein, The image data information comprises a plurality of detection images, and each detection image corresponds to analysis index data matched with a target detection requirement. The image analysis on the image data information to generate first guidance information comprises: performing image analysis on the image data information based on the analysis index data to obtain image analysis results; in a case where the image analysis results indicate that the analysis index data corresponding to at least one detection image in the plurality of detection images does not satisfy a preset analysis condition, determining the at least one detection image as a to-be-detected image; taking the retake parameter information and the retake region information corresponding to the to-be-detected image as the first guidance information.

7. A remote dosing device for performing the method of any one of claims 1-6, based on real-time monitoring of subject dosing and smart drug matching, characterized in that, The device comprises: a data collection module configured to collect image data information matched with a target detection requirement corresponding to a target user; a first data sending module configured to send the image data information to a management end, so that the management end performs image analysis on the image data information to generate first guidance information; a first data receiving module configured to receive the first guidance information fed back by the management end, and perform data analysis on the image data information based on a data analysis model and the first guidance information to obtain second guidance information; a drug selection module configured to dynamically adjust the drug based on the second guidance information.

8. The remote drug administration device for real-time monitoring and smart drug matching based on subject administration according to claim 7, wherein, The device comprises: a second data receiving module configured to receive image data information collected by a client; an image analysis module configured to perform image analysis on the image data information to generate first guidance information; a second data sending module configured to send the first guidance information to the client, so that the client performs data analysis on the first guidance information based on a data analysis model to obtain second guidance information, and dynamically adjusts the drug based on the second guidance information. 9.An electronic device, the device comprising a processor and a memory, the memory having stored therein at least one instruction and at least one program, the at least one instruction and the at least one program being loaded and executed by the processor to implement the remote drug administration method based on real-time monitoring of subject administration and intelligent drug matching according to any one of claims 1-6.

Images