Intelligent tongue diagnosis instrument based on multispectral imaging technology

Abstract

This application relates to the field of biomedical imaging technology, and more particularly to an intelligent tongue diagnosis device based on multispectral imaging technology. The device includes a multispectral imaging system, a data processing module, and an analysis module. The multispectral imaging system is configured to acquire multispectral images of the human tongue using multiple single-wavelength light sources. The data processing module is configured to extract tongue coating morphology features, vascular distribution information, and sublingual oxygen saturation information from the acquired multispectral images. The analysis module is configured to generate a diagnostic report and personalized health recommendations based on the tongue coating morphology features, vascular distribution information, and sublingual oxygen saturation information. Therefore, it is possible to accurately extract key information such as tongue coating morphology features, vascular distribution, and sublingual oxygen saturation from the acquired multispectral images, improving the accuracy and efficiency of diagnosis, and providing users with personalized preliminary dietary and health advice.

Description

Technical Field

[0001] This application relates to the field of biomedical imaging technology, and in particular to an intelligent tongue diagnosis instrument based on multispectral imaging technology. Background Technology

[0002] Tongue diagnosis in traditional Chinese medicine is an important component of TCM's visual diagnosis, playing a significant role in clinical diagnosis, constitution assessment, disease progression and prognosis, and prescription guidance. However, traditional tongue diagnosis is heavily influenced by the doctor's clinical experience and the surrounding environment, resulting in highly subjective diagnostic results that are prone to error, thus limiting the transmission of valuable TCM experience.

[0003] In recent years, with the development of computer technology, digital technology, new sensors, and nonlinear information processing methods, significant progress has been made in the objective study of tongue diagnosis in Traditional Chinese Medicine. Existing tongue image analysis equipment mainly consists of a digital tongue image acquisition system and an image feature processing system, typically using white light illumination to acquire color tongue photographs.

[0004] However, existing technologies still have shortcomings in terms of standardization and normalization, and the effective integration of new technologies with clinical needs needs to be strengthened. Furthermore, traditional white light imaging struggles to acquire functional information such as sublingual oxygen saturation, which is associated with a variety of diseases. Summary of the Invention

[0005] This application provides an intelligent tongue diagnosis instrument based on multispectral imaging technology, which can solve at least one of the technical problems in the background art to a certain extent.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, a smart tongue diagnosis instrument based on multispectral imaging technology is provided, comprising a multispectral imaging system, a data processing module, and an analysis module, wherein... The multispectral imaging system is configured to acquire multispectral images of the human tongue using multiple single-wavelength light sources; The data processing module is configured to extract tongue coating morphology features, blood vessel distribution information, and sublingual blood oxygen saturation information from the acquired multispectral images. The analysis module is configured to generate diagnostic reports and personalized health recommendations based on the tongue coating morphology, blood vessel distribution information, and sublingual blood oxygen saturation information.

[0008] Optionally, the multiple single-wavelength light sources include at least 530nm and 488nm wavelength light sources, where 530nm is the isoabsorption point of oxyhemoglobin and deoxyhemoglobin, and 488nm is the point of maximum absorption difference between oxyhemoglobin and deoxyhemoglobin.

[0009] Optionally, the multispectral imaging system includes: 488nm and 530nm lasers are used to provide illumination sources of the corresponding wavelengths; Beam splitters, laser beam expanders, and optical cameras are used to realize optical path transmission and image acquisition. A timing control unit is configured to control the timing of the laser to achieve synchronized image acquisition under different wavelengths of light.

[0010] Optionally, the data processing module is configured as follows: The acquired multispectral images are color corrected and the tongue is segmented to obtain the target image; Calculate the optical density ratio corresponding to the target image according to the Lambert-Beer law; The sublingual oxygen saturation is calculated based on the optical density ratio.

[0011] Optionally, the analysis module is configured as follows: Based on machine learning algorithms, the tongue coating morphology, blood vessel distribution, and sublingual oxygen saturation information are jointly analyzed to automatically identify lesion areas. Based on the knowledge graph of the lesion area and the field of traditional Chinese medicine, a structured TCM diagnostic report is generated.

[0012] Optionally, the step of using machine learning algorithms to jointly analyze the tongue coating morphology, vascular distribution information, and sublingual oxygen saturation information to automatically identify lesion areas includes: The tongue coating morphology features, blood vessel distribution information, and sublingual blood oxygen saturation information are input into a pre-trained convolutional neural network model, which outputs the semantic segmentation results of the tongue lesion area.

[0013] Optionally, a remote service module is also included, for: The collected multispectral images, diagnostic reports, and personalized health recommendations are uploaded to a cloud server to enable remote tongue diagnosis services via 5G smart IoT.

[0014] Optionally, it also includes portable structural components and a user boot module, wherein, The portable structural component integrates the multispectral imaging system using a cage-like optical structure. The user guidance module is configured to guide the test subject to adjust the position of their head and tongue through a display screen or voice prompts.

[0015] Optionally, the analysis module is further configured as follows: By linking user identity information and comparing historical tongue diagnosis data, a health trend analysis report is generated.

[0016] Optionally, the data processing module is further configured to: The multispectral image was corrected for white balance and color consistency using a standard color chart.

[0017] In this embodiment, an intelligent tongue diagnosis device based on multispectral imaging technology includes a multispectral imaging system, a data processing module, and an analysis module. The multispectral imaging system is configured to acquire multispectral images of the human tongue using multiple single-wavelength light sources. The data processing module is configured to extract tongue coating morphology features, blood vessel distribution information, and sublingual oxygen saturation information from the acquired multispectral images. The analysis module is configured to generate a diagnostic report and personalized health recommendations based on the tongue coating morphology features, blood vessel distribution information, and sublingual oxygen saturation information. Therefore, it is possible to accurately extract key information such as tongue coating morphology features, blood vessel distribution, and sublingual oxygen saturation from the acquired multispectral images, improving the accuracy and efficiency of diagnosis and providing users with personalized preliminary dietary and health advice.

[0018] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0019] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a schematic diagram of the structure of the intelligent tongue diagnostic instrument based on multispectral imaging technology provided in the embodiments of this application; Figure 2 This is the absorption spectrum of oxyhemoglobin (HbO2) and deoxyhemoglobin (Hb); Figure 3 This is a schematic diagram of the prototype of the multispectral tongue diagnosis imaging system and its illumination mode provided in an embodiment of the present invention. Detailed Implementation

[0020] The embodiments of the technical solutions of this application will now be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of this application, and are therefore merely examples and should not be used to limit the scope of protection of this application. When the following description relates to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. Various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but can be changed as will become apparent upon understanding this disclosure, except for operations that must be performed in a specific order. Furthermore, for clarity and conciseness, descriptions of features known in the art may be omitted.

[0021] The embodiments described in the following examples of this disclosure are not representative of all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0022] See Figure 1 This is a schematic diagram of the structure of an intelligent tongue diagnostic instrument based on multispectral imaging technology provided in an embodiment of this application. The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] like Figure 1 As shown, the intelligent tongue diagnostic instrument based on multispectral imaging technology provided in this embodiment mainly includes a multispectral imaging system, a data processing module, and an analysis module. Additionally, a remote service module, portable structural components, and a user guidance module can also be configured. These modules work collaboratively to complete multispectral image acquisition, data processing, intelligent analysis, diagnostic output, and remote transmission of the tongue, achieving non-contact tongue diagnostic testing.

[0024] Multispectral Imaging Technology (MIT) is an imaging technology that uses multiple single-wavelength light sources to image a target, capture the target's characteristic information in different spectral bands, and then extract the target's details and functional information.

[0025] The multispectral imaging system is the imaging component of the intelligent tongue diagnostic instrument. Its function is to acquire multispectral images of the human tongue using multiple single-wavelength light sources, providing raw image data for subsequent data processing and analysis. This system can simultaneously acquire tongue morphological and functional information.

[0026] Optionally, the multiple single-wavelength light sources include at least 530nm and 488nm wavelength light sources, where the 530nm wavelength is the isoabsorption point of oxyhemoglobin and deoxyhemoglobin, and the 488nm wavelength is the point of maximum absorption difference between oxyhemoglobin and deoxyhemoglobin.

[0027] Optional, multispectral imaging systems include: 488nm and 530nm lasers are used to provide illumination sources of the corresponding wavelengths; Beam splitters, laser beam expanders, and optical cameras are used to realize optical path transmission and image acquisition. The timing control unit is configured to control the timing of the laser to achieve synchronized image acquisition under different wavelengths of light.

[0028] Optionally, the multiple single-wavelength light sources include at least 530 nm and 488 nm wavelength light sources, wherein the 530 nm wavelength is the isoabsorption point of oxyhemoglobin (HbO2) and deoxyhemoglobin (Hb), and the 488 nm wavelength is the point of maximum absorption difference between oxyhemoglobin and deoxyhemoglobin.

[0029] Oxyhemoglobin refers to hemoglobin bound to oxygen and is the primary carrier of oxygen transport in the blood. Deoxyhemoglobin refers to hemoglobin that is not bound to oxygen. The isosbestic point refers to the point at which the molar absorptivity of two substances is equal at a specific wavelength; at this point, the absorption spectra of the two substances intersect, and this point can be used to calculate the concentration ratio of the two substances. The molar absorptivity (MA) refers to a substance's ability to absorb light at a specific wavelength and is an important parameter for measuring a substance's light absorption characteristics.

[0030] Optionally, the multispectral imaging system specifically includes a 488nm laser (Laser Diode, LD), a 530nm laser (Laser Diode, LD), a beam splitter, a laser beam expander, an optical camera, and a timing control unit. The functions and connections of each component are as follows: The 488nm and 530nm lasers provide monochromatic illumination at their respective wavelengths. The 488nm laser outputs monochromatic light at 488nm, while the 530nm laser outputs monochromatic light at 530nm. These two light sources operate independently, illuminating the tongue separately to capture images of the tongue under different spectra. A beam splitter separates the light path, precisely guiding the monochromatic light output from the laser to the tongue detection area and simultaneously guiding the reflected light signal from the tongue to the optical camera, ensuring the stability and accuracy of the light path transmission. A laser beam expander amplifies the laser beam, increasing the illumination range and ensuring uniform illumination of the tongue detection area, avoiding image acquisition deviations caused by uneven local illumination. The optical camera captures the reflected light signals from the tongue under different wavelengths of illumination, converting them into digital images. The image resolution is tailored to the detailed tongue capture requirements, clearly capturing subtle features such as tongue texture and blood vessel distribution. The timing control unit controls the timing of the laser to achieve synchronous image acquisition under different wavelengths of light. Specifically, it precisely controls the on and off timing of the 488nm and 530nm lasers to ensure that the two wavelength light sources provide alternating illumination. The optical camera synchronously acquires tongue images at the corresponding wavelengths, avoiding interference between the two light source signals and ensuring the quality of multispectral image acquisition.

[0031] Optionally, the optical path design of the multispectral imaging system adopts a standardized optical structure, with all components integrated into a portable structural assembly to ensure system stability. During prototype construction, the focus was on resolving the placement of the optical camera to ensure the image acquisition angle is adapted to tongue imaging, reducing image distortion. Simultaneously, methods for fixing the subject's head and tongue were explored to reduce motion artifacts and further improve image acquisition quality. Future optimization directions include replacing the laser beam expander with a simplified lens system and designing the optical path using fiber optics to achieve system miniaturization and portability upgrades.

[0032] The data processing module is configured to extract tongue coating morphology features, blood vessel distribution information, and sublingual oxygen saturation information from the acquired multispectral images. The data processing module is electrically connected to the multispectral imaging system, receives multispectral images acquired by the system, preprocesses and extracts features from the images, and ultimately obtains the tongue coating morphology features, blood vessel distribution information, and sublingual oxygen saturation information.

[0033] Optionally, the data processing module is configured as follows: Color correction and tongue segmentation are performed on the acquired multispectral images to obtain the target image; Calculate the optical density ratio of the target image according to the Lambert-Beer law; Calculate sublingual oxygen saturation based on optical density ratio.

[0034] Optionally, the data processing module is also configured as follows: White balance and color consistency correction were performed on multispectral images using a standard color chart.

[0035] Among them, the morphological characteristics of tongue coating refer to quantifiable morphological parameters such as the thickness, color, texture, and distribution range of tongue coating.

[0036] Among them, vascular distribution information refers to quantitative information such as the direction, density, and thickness of blood vessels in the tongue (especially under the tongue), which can reflect the blood circulation status of the tongue.

[0037] Among them, sublingual oxygen saturation refers to the percentage of oxyhemoglobin in the sublingual blood relative to the total hemoglobin.

[0038] Optionally, the data processing module is specifically configured to perform the following steps to complete data processing: The acquired multispectral images were subjected to color correction and tongue segmentation to obtain the target image. Color correction employed a standard color chart for white balance and color consistency correction, eliminating the influence of factors such as light intensity and ambient light on image color and ensuring color consistency across different detection scenarios. Tongue segmentation used an image segmentation algorithm to accurately segment the tongue region from the multispectral image, removing irrelevant areas such as the mouth, teeth, and lips to reduce interference from subsequent feature extraction. The optical density ratio (ODR) of the target image was calculated according to the Lambert-Beer Law. The Lambert-Beer Law is a fundamental law describing the degree of light absorption by a substance, and its formula is as follows:

[0039] in, and They represent the incident light and the outgoing light, respectively. It is the molar absorption coefficient. Represents the concentration of blood components. This represents the optical path length. There is an approximately linear relationship between sublingual oxygen saturation and optical density ratio (ODR), as shown in the following formula:

[0040]

[0041] in, and It is a correction factor. It is the optical density at 530nm. It is the optical density at 488nm.

[0042] The analysis module is configured to generate diagnostic reports and personalized health recommendations based on tongue coating morphology, vascular distribution information, and sublingual oxygen saturation. Electrically connected to the data processing module, its core function is to receive the tongue coating morphology, vascular distribution, and sublingual oxygen saturation information from the data processing module, and generate diagnostic reports and personalized health recommendations through intelligent analysis, thus achieving intelligent and objective tongue diagnosis. The analysis module incorporates a traditional Chinese medicine knowledge graph and machine learning model, enabling automatic identification of lesion areas, TCM diagnostic analysis, and the generation of health recommendations.

[0043] Optionally, the analysis module can be configured as follows: Based on machine learning algorithms, the morphological characteristics of tongue coating, vascular distribution information and sublingual blood oxygen saturation information are jointly analyzed to automatically identify lesion areas. Based on the knowledge graph of the lesion area and the field of traditional Chinese medicine, a structured TCM diagnostic report is generated.

[0044] Optionally, based on machine learning algorithms, the morphological characteristics of the tongue coating, vascular distribution information, and sublingual oxygen saturation information are jointly analyzed to automatically identify lesion areas, including: The morphological features of the tongue coating, the distribution of blood vessels, and the oxygen saturation of blood under the tongue are input into a pre-trained convolutional neural network model, which outputs the semantic segmentation results of the lesion area of ​​the tongue.

[0045] Among them, the Traditional Chinese Medicine Knowledge Graph is a database that integrates knowledge of TCM diagnostic theories, the relationship between symptoms and tongue appearance characteristics, and the rules of syndrome differentiation and treatment in a structured form, providing theoretical support for intelligent diagnosis.

[0046] Among them, the machine learning model is an algorithm model that learns patterns through training data, thereby achieving data classification, recognition, and prediction. In this embodiment, it is used for the identification of lesion areas on the tongue.

[0047] Optionally, the analysis module can be configured to perform the following steps to complete the analysis: Based on machine learning algorithms, this method jointly analyzes tongue coating morphology, vascular distribution, and sublingual oxygen saturation to automatically identify lesion areas on the tongue. Specifically, the morphology, vascular distribution, and sublingual oxygen saturation are input into a pre-trained convolutional neural network (CNN) model, which outputs semantic segmentation results of the lesion areas. A convolutional neural network is a deep learning model specifically designed for image recognition and segmentation. Semantic segmentation is an image processing technique that assigns each pixel in an image to a corresponding category (such as normal area or lesion area), accurately locating the position and extent of lesions.

[0048] It should be noted that the training process of the convolutional neural network model is as follows: A large number of multispectral image samples of the tongue are collected, and the lesion areas in the samples (such as abnormal tongue coating areas, abnormal blood vessel areas, etc.) are labeled. Simultaneously, the corresponding tongue coating morphology features, blood vessel distribution information, and sublingual blood oxygen saturation information are extracted to construct a training dataset. The training dataset is then input into the convolutional neural network model for iterative training, optimizing the model parameters until the model's lesion recognition accuracy reaches the preset standard, thus completing model training. The trained model can quickly and accurately identify lesion areas of the tongue, improving diagnostic efficiency and accuracy.

[0049] Furthermore, based on the lesion area identification results and the TCM knowledge graph, a structured TCM diagnostic report can be generated. This report includes a summary of quantitative indicators of tongue appearance, a description of the lesion area, and TCM syndrome differentiation results (such as Yin deficiency, Yang deficiency, damp-heat, etc.), clearly presenting the tongue diagnosis analysis results and providing doctors with objective and quantitative diagnostic evidence, while also facilitating users' understanding of their own tongue appearance. Combining the tongue diagnosis analysis results, TCM syndrome differentiation theory, and basic user information (such as age, gender, and dietary habits), personalized dietary and health recommendations can be generated. For example, for users with a thick, greasy tongue coating and slightly low sublingual oxygen saturation, a light diet is recommended, reducing the intake of greasy and spicy foods, while also suggesting appropriate exercise to improve blood circulation; for users with a thin tongue coating and insufficient Qi and blood, it is recommended to consume more Qi-nourishing and blood-tonifying foods and maintain a regular sleep schedule.

[0050] Optionally, the analysis module can also be configured as follows: By linking user identity information and comparing historical tongue diagnosis data, a health trend analysis report is generated.

[0051] Specifically, the analysis module is also configured to link user identity information, perform longitudinal comparisons of past tongue diagnosis data, and generate a health trend analysis report. This report clearly displays the changing trends of the user's tongue image indicators, helping users and doctors understand changes in their physical condition and providing a reference for disease prevention and treatment.

[0052] Optionally, a remote service module is also included, for: The collected multispectral images, diagnostic reports, and personalized health recommendations are uploaded to a cloud server to enable remote tongue diagnosis services via 5G smart IoT.

[0053] The remote service module is electrically connected to the analysis module and the multispectral imaging system. Its function is to enable remote tongue diagnosis services, expanding the application scope of the intelligent tongue diagnosis instrument, especially to meet the medical needs of remote areas with relatively underdeveloped medical conditions. Specifically, the acquired multispectral images, diagnostic reports generated by the analysis module, and personalized health suggestions are uploaded to a cloud server via 5G Internet of Things (5G IoT) technology. Doctors can log in to the cloud server through terminal devices (such as computers and mobile phones) to view user tongue diagnosis data and diagnostic results, enabling remote diagnosis and guidance, and extending high-quality medical resources to grassroots and remote areas.

[0054] Optionally, it also includes portable structural components and a user boot module, wherein, The portable structural components integrate a multispectral imaging system using a cage-like optical structure; The user guidance module is configured to guide the test subject to adjust the position of their head and tongue through a display screen or voice prompts.

[0055] The portable structural component adopts a cage-like optical structure to integrate all components of the multispectral imaging system, including the laser, beam splitter, laser beam expander, optical camera, and timing control unit. The cage-like optical structure is characterized by its compact structure, high stability, and easy assembly and disassembly, enabling the miniaturization and portability of the intelligent tongue diagnostic instrument. This facilitates its transport to community health centers, pharmacies, and other primary healthcare facilities, while also reducing the space occupied by the device.

[0056] Among them, the cage-type optical structure is a standardized optical integrated structure composed of metal rods, connectors and optical supports. It can accurately fix various optical components, ensure optical path stability, and is widely used in the integration of small optical devices.

[0057] The user guidance module is configured to guide the subject to adjust their head and tongue position via a display screen or voice prompts, ensuring the tongue is in the optimal acquisition area, reducing motion artifacts, and improving image acquisition quality. For example, the display screen can show a schematic diagram of the tongue acquisition area, and the voice prompt can tell the subject to "keep their head still, extend their tongue naturally, and remain motionless," while simultaneously displaying the tongue position in real time. After guiding the subject to adjust to the appropriate position, the multispectral imaging system is triggered to acquire the image.

[0058] Figure 2 This is the absorption spectrum of oxyhemoglobin (HbO2) and deoxyhemoglobin (Hb). The figure shows the absorption characteristics of Hb and HbO2 at different wavelengths, with 530 nm being the isoabsorption point and 488 nm being the point of maximum absorption difference, providing a theoretical basis for wavelength selection in multispectral imaging.

[0059] Figure 3 The following is a schematic diagram of the prototype and illumination mode of the multispectral tongue diagnosis imaging system provided in an embodiment of the present invention, which is described in detail below: Figure 3 (A) is a structural diagram of the system prototype, showing the layout of the core optical components of the multispectral tongue diagnosis imaging system. The functions of each component are as follows: 530nm laser diode (530nm laser): Outputs 530nm monochromatic laser light, which is the isoabsorption point of oxyhemoglobin and deoxyhemoglobin, and is used to acquire spectral images of the tongue at the isoabsorption wavelength.

[0060] 488nm laser diode (488nm laser): Outputs 488nm monochromatic laser, which is the wavelength of maximum absorption difference between oxyhemoglobin and deoxyhemoglobin, and is used to acquire spectral images of the tongue at the wavelength of maximum absorption difference.

[0061] A dichrooscope enables the combining and separating of two wavelengths of laser light, precisely guiding the laser to the tongue detection area while simultaneously guiding the reflected light from the tongue to the subsequent imaging unit.

[0062] Laser beam expander: Expands the laser beam to increase the illumination range, ensuring uniform illumination of the tongue detection area and avoiding image deviations caused by uneven local illumination.

[0063] Figure 3 (B) shows the illumination effect when the 488nm laser diode is turned on alone. The tongue detection area shows a light spot. At this wavelength, the tongue tissue has the greatest difference in absorption between oxyhemoglobin and deoxyhemoglobin, which can highlight the functional characteristics of blood oxygenation in the tongue. Figure 3 (C) shows the illumination effect when the 530nm laser diode is turned on alone. A light spot appears in the tongue detection area. At this wavelength, the absorption coefficients of oxyhemoglobin and deoxyhemoglobin are equal, which can be used to correct the concentration ratio in the blood oxygen saturation calculation and improve the accuracy of blood oxygen detection.

[0064] In this embodiment, the workflow of the intelligent tongue diagnostic instrument is as follows: 1. Turn on the intelligent tongue diagnostic instrument. Each module completes self-testing. The portable structural components ensure the stability of the multispectral imaging system components. The user guidance module starts up, and the display screen shows the operation instructions. 2. The subject adjusts the position of his head and tongue according to the display screen or voice prompts of the user guidance module until the tongue is in the optimal acquisition area. After the user guidance module detects the appropriate position, it issues a acquisition prompt. 3. The timing control unit controls the 488nm laser and the 530nm laser to turn on alternately. The beam splitter guides the laser to the tongue, the laser beam expander expands the laser beam, and the optical camera simultaneously acquires multispectral images of the tongue at the two wavelengths. After the acquisition is completed, the image data is transmitted to the data processing module. 4. The data processing module performs color correction and tongue segmentation on the multispectral image to obtain the target image; calculates the optical density ratio according to Lambert-Beer's law, and calculates the sublingual blood oxygen saturation using the correction formula; at the same time, it extracts the morphological features of the tongue coating and the information on blood vessel distribution, and standardizes all quantitative data. 5. The analysis module receives the quantitative data output by the data processing module and automatically identifies the lesion areas of the tongue through a convolutional neural network model; combined with the TCM knowledge graph, it generates a structured TCM diagnostic report and personalized health advice; if the health trend analysis function is enabled, it associates user identity information and generates a health trend analysis report. 6. Diagnostic reports, personalized health advice, and health trend analysis reports are displayed on the screen and can also be printed. If a remote service module is configured, multispectral images, diagnostic reports, and health advice can be uploaded to a cloud server to enable remote tongue diagnosis services. 7. The device automatically turns off the light source, prompts the subject that the test is complete, and saves the tongue diagnosis data for later query and longitudinal comparison.

[0065] Compared to existing technologies, the technical advantages of this solution are: (1) Functional imaging capability: Existing tongue diagnosis equipment mainly uses white light illumination to acquire color tongue photographs, which can only provide macroscopic information such as the shape and color of the tongue. This invention uses multispectral imaging technology, which can acquire sublingual blood oxygen saturation information, which is impossible with traditional white light imaging. Blood oxygen saturation is an important physiological and biochemical indicator, and is associated with a variety of diseases (such as severe sepsis, diabetic nephropathy, severe H1N1 influenza, dengue fever, etc.). Therefore, this invention can provide deeper functional diagnostic information.

[0066] (2) Non-contact measurement and safety: This invention uses multispectral imaging technology to perform non-contact measurement of the tongue, avoiding direct contact between the doctor and the patient in traditional tongue diagnosis and reducing the risk of cross-infection. At the same time, multispectral imaging uses visible light and near-infrared light, which is safer and has no radiation risk than some medical imaging technologies that require ionizing radiation such as X-rays, and is non-invasive to patients.

[0067] (3) Objectification and Quantification of Diagnosis: Traditional Chinese medicine tongue diagnosis relies heavily on the doctor's experience, and the diagnostic results are highly subjective. This invention obtains quantitative indicators such as tongue coating morphology, blood vessel distribution, and blood oxygen saturation through multispectral imaging, and analyzes them in conjunction with data processing programs, transforming subjective experience into objective data. This provides a quantitative basis and scientific evidence for Chinese medicine tongue diagnosis, and improves the accuracy and repeatability of diagnosis.

[0068] (4) Intelligent analysis and assisted diagnosis: This invention combines machine learning algorithms to segment lesion areas and performs TCM analysis based on quantitative data, which can provide doctors with more accurate assisted diagnostic information and even give preliminary dietary advice. This not only reduces the workload of doctors, but also improves diagnostic efficiency and the level of personalized service.

[0069] (5) Wide range of applications and potential for telemedicine: This invention aims to develop a miniaturized, low-cost, and easy-to-operate intelligent tongue diagnosis instrument, which can be promoted to community health centers and pharmacies in the future, so that more people can enjoy convenient tongue diagnosis services. In addition, with the help of 5G smart Internet of Things technology and cloud computing, this tongue diagnosis instrument has the prospect of telemedicine, which can deliver high-quality medical resources to remote areas with relatively poor medical conditions.

[0070] If the integrated unit is implemented as a software functional unit and used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / electronic device, a recording medium, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0071] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0072] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0073] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0074] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0075] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An intelligent tongue diagnostic instrument based on multispectral imaging technology, characterized in that, It includes a multispectral imaging system, a data processing module, and an analysis module, among which, The multispectral imaging system is configured to acquire multispectral images of the human tongue using multiple single-wavelength light sources; The data processing module is configured to extract tongue coating morphology features, blood vessel distribution information, and sublingual blood oxygen saturation information from the acquired multispectral images. The analysis module is configured to generate diagnostic reports and personalized health recommendations based on the tongue coating morphology, blood vessel distribution information, and sublingual blood oxygen saturation information.

2. The intelligent tongue diagnosis instrument according to claim 1, characterized in that, The multiple single-wavelength light sources include at least 530nm and 488nm wavelength light sources, where 530nm is the isoabsorption point of oxyhemoglobin and deoxyhemoglobin, and 488nm is the point of maximum absorption difference between oxyhemoglobin and deoxyhemoglobin.

3. The intelligent tongue diagnosis instrument according to claim 2, characterized in that, The multispectral imaging system includes: 488nm and 530nm lasers are used to provide illumination sources of the corresponding wavelengths; Beam splitters, laser beam expanders, and optical cameras are used to realize optical path transmission and image acquisition. A timing control unit is configured to control the timing of the laser to achieve synchronized image acquisition under different wavelengths of light.

4. The intelligent tongue diagnosis instrument according to claim 2, characterized in that, The data processing module is configured as follows: The acquired multispectral images are color corrected and the tongue is segmented to obtain the target image; Calculate the optical density ratio corresponding to the target image according to the Lambert-Beer law; The sublingual oxygen saturation is calculated based on the optical density ratio.

5. The intelligent tongue diagnosis instrument according to claim 2, characterized in that, The analysis module is configured as follows: Based on machine learning algorithms, the tongue coating morphology, blood vessel distribution, and sublingual oxygen saturation information are jointly analyzed to automatically identify lesion areas. Based on the knowledge graph of the lesion area and the field of traditional Chinese medicine, a structured TCM diagnostic report is generated.

6. The intelligent tongue diagnosis instrument according to claim 5, characterized in that, The method, based on machine learning algorithms, jointly analyzes the tongue coating morphology, vascular distribution information, and sublingual oxygen saturation information to automatically identify lesion areas, including: The tongue coating morphology features, blood vessel distribution information, and sublingual blood oxygen saturation information are input into a pre-trained convolutional neural network model, which outputs the semantic segmentation results of the tongue lesion area.

7. The intelligent tongue diagnosis instrument according to claim 5, characterized in that, It also includes a remote service module, used for: The collected multispectral images, diagnostic reports, and personalized health recommendations are uploaded to a cloud server to enable remote tongue diagnosis services via 5G smart IoT.

8. The intelligent tongue diagnosis instrument according to claim 1, characterized in that, It also includes portable structural components and a user boot module, among which, The portable structural component integrates the multispectral imaging system using a cage-like optical structure. The user guidance module is configured to guide the test subject to adjust the position of their head and tongue through a display screen or voice prompts.

9. The intelligent tongue diagnosis instrument according to claim 5, characterized in that, The analysis module is also configured to: By linking user identity information and comparing historical tongue diagnosis data, a health trend analysis report is generated.

10. The intelligent tongue diagnosis instrument as described in claim 1, characterized in that, The data processing module is also configured to: The multispectral image was corrected for white balance and color consistency using a standard color chart.

Images