Evaluation system for bilirubin level

Abstract

The invention relates to the technical field of jaundice detection, in particular to a bilirubin level evaluation system, comprising: an input module for inputting historical patient eye images and measured bilirubin value parameters; the camera module is used for collecting eye images of a patient and performing structured similarity comparison on the collected eye images of the patient and the historical eye images of the patient in the input module; the calling module is used for calling the two historical patient eye images with the optimal structured similarity comparison result in the camera module, and identifying the difference ratio of the bilirubin value of the patient to the eye images based on the two called historical patient eye images; different from the prior art, the bilirubin evaluation method has the advantages that the images of the eyes of the user are acquired by the machine vision technology, the bilirubin level of the user is evaluated by combining a large amount of prior data with the images of the eyes of the user, and the bilirubin evaluation method is simpler and quicker than existing blood detection and sensor detection and is more accurate than existing skin observation modes by machine vision.

Description

Technical Field

[0001] The invention relates to the technical field of jaundice detection, and in particular to an evaluation system for bilirubin levels. Background Art

[0002] Neonatal jaundice refers to a disease characterized by yellowing of the skin, mucous membranes and sclera due to abnormal bilirubin metabolism in the neonatal period. About 60% of newborns will have jaundice symptoms, of which pathological jaundice often lasts for more than 2 to 4 weeks. For newborns, if jaundice is not treated in time, it may cause visual impairment, the formation of bilirubin stones, poor nutrient absorption, impaired liver function, impaired immune function and other symptoms, thus affecting the baby's healthy growth.

[0003] The invention patent with application number 202010027602.1 discloses a jaundice monitoring sensor in contact with the skin, including: A) a decomposition light source that operates when the sensor light source is turned off B) a sensor light source that operates when the decomposition light source is turned off C) a detector that operates when the sensor light source is turned on, wherein, if turned on, the decomposition light source emits light, decomposes and removes fixed (tissue-attached) bilirubin from the skin in the light path of the sensor light source, and the sensor measures the decomposition rate of fixed (tissue-attached) bilirubin; when the sensor light source is turned on, the light received by the detector passes through mobile (in the blood) bilirubin rather than fixed (tissue-attached) bilirubin, generating a sensor signal proportional to the mobile (in the blood) bilirubin, thereby determining the blood bilirubin level.

[0004] The application aims to address the problem that "the problem with all current non-invasive methods for jaundice detection is that analyzing or imaging the yellow color of the skin indicates jaundice, but cannot directly reflect the bilirubin level in the blood and the changes in the bilirubin level in the blood."

[0005] However, although the above method can accurately measure the bilirubin level in the blood, obtaining the measurement results requires a professional sensor, and the measurement results are affected by the detection operation and the performance of the sensor itself. Once an operation error or a performance failure occurs, the detection result will be greatly biased or even impossible to detect.

[0006] To this end, a system for assessing bilirubin levels is proposed. Summary of the invention

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a system for evaluating bilirubin levels, which solves the technical problems raised in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A system for assessing bilirubin levels, comprising:

[0010] An input module is used to input historical patient eye images and measured bilirubin value parameters; a camera module is used to collect patient eye images, and use the collected patient eye images to perform structured similarity comparison with the historical patient eye images in the input module; a retrieval module is used to retrieve two historical patient eye images with the best structured similarity comparison results in the camera module, and identify the difference ratio between the patient's bilirubin value and the eye image based on the two retrieved historical patient eye images; an analysis module is used to analyze the similarity between the patient eye images collected by the camera module and the two patient eye images retrieved in the retrieval module, and determine the bilirubin value interval corresponding to the patient eye image collected by the camera module based on the similarity analysis results; a correction module is used to receive the bilirubin value interval corresponding to the patient eye image determined in the analysis module, and decide the output and correction of the bilirubin value interval based on the interval attributes; an output module is used to output the bilirubin value interval processed by the correction module.

[0011] Furthermore, the input module is internally provided with submodules, including:

[0012] A binding unit, used for monitoring the patient's eye image and the measured bilirubin value parameter in the input module, and binding the continuously input patient's eye image and the measured bilirubin value parameter;

[0013] The camera module is provided with submodules at the lower level, including:

[0014] A preprocessing unit, used for receiving the patient's eye image acquired by the camera module, converting the patient's eye image into a grayscale image, and outputting the grayscale image;

[0015] A comparison unit is used to receive the patient's eye image converted into a grayscale image output by the preprocessing unit, and to perform structural similarity comparison with the received grayscale image by taking each patient's eye image stored in the input module as a comparison target;

[0016] After the input module inputs and binds the historical patient eye images and the measured bilirubin value parameters based on the binding unit, the input module synchronously sorts and stores the mutually bound patient eye images and bilirubin value parameters based on the input time sequence. The historical patient eye images and the measured bilirubin value parameters input in the input module are derived from the EMR system backend. When the retrieval module retrieves the patient eye images in the input module, the bilirubin value parameters corresponding to the bound patient eye images are synchronously retrieved.

[0017] When the comparison unit runs and applies the patient's eye image stored in the entry module as the comparison target, the patient's eye image stored in the entry module is compared sequentially based on the time sequence. When the patient's eye image in the entry module is applied to the comparison operation, it is synchronously processed by the preprocessing unit and converted into a grayscale image before performing the comparison operation.

[0018] Furthermore, the structured similarity comparison logic between the patient's eye image and the grayscale image in the comparison unit is expressed as:

[0019] Setting grayscale value intervals representing the pupil and the iris, setting grayscale value intervals representing the sclera, segmenting the pupil, iris region images and the sclera region images in the patient's eye image and the grayscale image based on the grayscale value intervals representing the pupil and the iris and the grayscale value intervals representing the sclera, and capturing the center points of the pupil, iris region images and the sclera region image;

[0020] ;

[0021] Where: is the structural similarity between the patient's eye image and the grayscale image; The coordinates of the center point of the pupil and iris area image in the grayscale image are determined based on the pixel position; The coordinates of the center point of the pupil and iris area image in the patient's eye image are determined based on the pixel position; is the coordinates of the center point of the sclera area image in the grayscale image determined based on the pixel position; The coordinates of the center point of the sclera area image in the patient's eye image determined based on the pixel position;

[0022] Among them, the structural similarity between the patient's eye image and the grayscale image The smaller the value, the higher the structural similarity between the patient's eye image and the grayscale image. Conversely, the lower the structural similarity between the patient's eye image and the grayscale image. Based on the above formula, the structural similarity between the grayscale image and each patient's eye image is calculated. The two historical patient eye images with the best comparison results retrieved by the module are The patient's eye image corresponding to the two calculation results with the smallest value.

[0023] Furthermore, the recognition logic of the difference ratio between the patient's bilirubin value and the eye image in the retrieval module is expressed as:

[0024] ;

[0025] Where: is the difference ratio between the patient's bilirubin value and the eye image; The similarity of the two patient eye images retrieved by the retrieval module in terms of color distribution; Bilirubin values ​​corresponding to two patient eye images retrieved by the retrieval module; , , The hue segmentation interval, the saturation segmentation interval, and the lightness segmentation interval are equally divided; is the pixel frequency of the patient eye image A and the patient eye image B in the i-th tone interval; is the pixel frequency of the patient eye image A and the patient eye image B in the jth saturation interval; The patient eye image A and the patient eye image B are in the The frequency of pixels in each brightness interval;

[0026] Among them, the difference ratio between the patient's bilirubin value and the eye image The unit is , The unit is percentage;

[0027] The analysis logic of the similarity between the patient eye image collected by the camera module in the analysis module and the two patient eye images retrieved in the retrieval module is the same as The calculation logic is the same, so the analysis result is recorded as , C represents the patient's eye image collected by the current camera module, and further determines The size relationship of the three;

[0028] In The bilirubin value range corresponding to the patient's eye image collected by the camera module is: ;

[0029] Greater than , the bilirubin value range corresponding to the patient's eye image collected by the camera module is:

[0030] Less than , the bilirubin value range corresponding to the patient's eye image collected by the camera module is: .

[0031] Furthermore, They respectively represent taking the minimum value and the maximum value in the brackets.

[0032] Furthermore, the output and correction logic of the bilirubin value interval based on interval attribute decision in the correction module is:

[0033] Interval attribute decision is to determine whether the bilirubin value interval contains ;

[0034] The bilirubin value range includes , the bilirubin value interval is sent to the output module, and the output module performs the output operation of the bilirubin value interval;

[0035] Bilirubin value range does not include , correct the bilirubin value interval;

[0036] The logic for correcting the bilirubin value interval in the correction module is expressed as:

[0037] Recognize the structural similarity between the patient eye image captured by the camera module and the patient eye image A and the patient eye image B;

[0038] The patient's eye image captured by the camera module has a higher structural similarity with the patient's eye image A, then:

[0039] Corrected to ;

[0040] The patient's eye image captured by the camera module has a higher structural similarity with the patient's eye image B, then:

[0041] Corrected to ;

[0042] Among them, the ± selection of the bilirubin value interval follows: , Take + from ±, Middle ± Take ﹣; ;

[0043] Take the middle ±, Take + from ±;

[0044] in, is a constant, which is defined by the system user, and >1, and constant Make sure that the corrected bilirubin value interval is still within the bilirubin value interval before correction.

[0045] Furthermore, the entry module is interactively connected to a binding unit via a wireless network, the entry module is interactively connected to a camera module via a wireless network, the camera module is interactively connected to a preprocessing unit and a comparison unit via a wireless network, the camera module is interactively connected to a retrieval module via a wireless network, the retrieval module is interactively connected to the entry module via a wireless network, and the retrieval module is interactively connected to an analysis module, a correction module and an output module via a wireless network.

[0046] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0047] The present invention provides a system for assessing bilirubin levels. During operation, the system, unlike the prior art, uses machine vision technology to collect binocular images of a user, and assesses the user's bilirubin level based on a large amount of prior data combined with the binocular images of the user. Compared with the existing blood test and sensor test, the system is simpler and faster, and more accurate than the existing machine vision skin observation method. In addition, the system can continuously improve the system assessment accuracy based on the continuous accumulation of prior data, and use the bilirubin value interval method to feedback to the system end user, which is more convenient for the system end user to make a diagnosis, so as to formulate and decide on a treatment plan more quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 A schematic diagram of the structure of a system for assessing bilirubin levels. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The present invention will be further described below in conjunction with the embodiments.

[0052] Example:

[0053] A system for assessing bilirubin levels in this embodiment, such as Figure 1 As shown, including:

[0054] An input module is used to input historical patient eye images and measured bilirubin value parameters;

[0055] The input module is internally provided with submodules, including:

[0056] A binding unit, used for monitoring the patient's eye image and the measured bilirubin value parameter in the input module, and binding the continuously input patient's eye image and the measured bilirubin value parameter;

[0057] Among them, after the input module inputs and binds the historical patient eye images and the measured bilirubin value parameters based on the binding unit, the eye images and bilirubin value parameters of each mutually bound patient are synchronously sorted and stored based on the input time sequence. The historical patient eye images and the measured bilirubin value parameters input in the input module are derived from the EMR system background. When the retrieval module retrieves the patient eye images in the input module, the bilirubin value parameters corresponding to the bound patient eye images are synchronously retrieved;

[0058] The camera module is used to collect the patient's eye image, and compare the collected patient's eye image with the historical patient's eye image in the input module for structured similarity;

[0059] The camera module is equipped with submodules, including:

[0060] A preprocessing unit, used for receiving the patient's eye image acquired by the camera module, converting the patient's eye image into a grayscale image, and outputting the grayscale image;

[0061] A comparison unit is used to receive the patient's eye image converted into a grayscale image output by the preprocessing unit, and to perform structural similarity comparison with the received grayscale image by taking each patient's eye image stored in the input module as a comparison target;

[0062] When the comparison unit runs and applies the patient eye images stored in the input module as the comparison target, the patient eye images stored in the input module are compared in sequence based on the time sequence. When the patient eye images in the input module are applied to the comparison operation, they are synchronously processed by the preprocessing unit and converted into grayscale images before the comparison operation is performed.

[0063] A retrieval module is used to retrieve two historical patient eye images with the best structural similarity comparison results in the camera module, and identify the difference ratio between the patient's bilirubin value and the eye image based on the two retrieved historical patient eye images;

[0064] The structured similarity comparison logic between the patient's eye image and the grayscale image in the comparison unit is expressed as:

[0065] Setting grayscale value intervals representing the pupil and the iris, setting grayscale value intervals representing the sclera, segmenting the pupil, iris region images and the sclera region images in the patient's eye image and the grayscale image based on the grayscale value intervals representing the pupil and the iris and the grayscale value intervals representing the sclera, and capturing the center points of the pupil, iris region images and the sclera region image;

[0066] ;

[0067] Where: is the structural similarity between the patient's eye image and the grayscale image; The coordinates of the center point of the pupil and iris area image in the grayscale image are determined based on the pixel position; The coordinates of the center point of the pupil and iris area image in the patient's eye image are determined based on the pixel position; is the coordinates of the center point of the sclera area image in the grayscale image determined based on the pixel position; The coordinates of the center point of the sclera area image in the patient's eye image determined based on the pixel position;

[0068] Among them, the structural similarity between the patient's eye image and the grayscale image The smaller the value, the higher the structural similarity between the patient's eye image and the grayscale image. Conversely, the lower the structural similarity between the patient's eye image and the grayscale image. Based on the above formula, the structural similarity between the grayscale image and each patient's eye image is calculated. The two historical patient eye images with the best comparison results retrieved by the module are The patient's eye image corresponding to the two calculation results with the smallest value;

[0069] The structural similarity between the patient's eye image and the grayscale image is calculated by the above logic formula, which provides necessary operation data support for the subsequent module operation of the system in this embodiment.

[0070] The recognition logic of the difference ratio between the patient's bilirubin value and the eye image in the retrieval module is expressed as:

[0071] ;

[0072] Where: is the difference ratio between the patient's bilirubin value and the eye image; The similarity of the two patient eye images retrieved by the retrieval module in terms of color distribution; Bilirubin values ​​corresponding to two patient eye images retrieved by the retrieval module; , , The hue segmentation interval, the saturation segmentation interval, and the lightness segmentation interval are equally divided; is the pixel frequency of the patient eye image A and the patient eye image B in the i-th tone interval; is the pixel frequency of the patient's eye image A and the patient's eye image B in the jth saturation interval; The patient eye image A and the patient eye image B are in the The frequency of pixels in each brightness interval;

[0073] Among them, the difference ratio between the patient's bilirubin value and the eye image The unit is , The unit is percentage; the meaning of the unit of K can be regarded as the change in bilirubin value corresponding to each percentage similarity, that is, H here represents the bilirubin value.

[0074] The above logic formula is used to calculate the difference ratio between the patient's bilirubin value and the eye image, providing necessary data support for the subsequent determination of the bilirubin value interval by the system in this embodiment.

[0075] An analysis module, used for analyzing the similarity between the patient eye image acquired by the camera module and the two patient eye images retrieved by the retrieval module, and determining the bilirubin value interval corresponding to the patient eye image acquired by the camera module based on the similarity analysis result;

[0076] The analysis logic and similarity between the patient eye images collected by the camera module in the analysis module and the two patient eye images retrieved in the retrieval module The calculation logic is the same, so the analysis result is recorded as , C represents the patient's eye image collected by the current camera module, and further determines The size relationship of the three;

[0077] In The bilirubin value range corresponding to the patient's eye image collected by the camera module is: ;

[0078] Greater than , the bilirubin value range corresponding to the patient's eye image collected by the camera module is:

[0079] Less than , the bilirubin value range corresponding to the patient's eye image collected by the camera module is: ;

[0080] A correction module, used for receiving the bilirubin value interval corresponding to the patient's eye image determined in the analysis module, and determining the output and correction of the bilirubin value interval based on the interval attribute;

[0081] The output and correction logic of the bilirubin value interval based on interval attribute decision in the correction module is:

[0082] Interval attribute decision is to determine whether the bilirubin value interval contains ;

[0083] The bilirubin value range includes , the bilirubin value interval is sent to the output module, and the output module performs the output operation of the bilirubin value interval;

[0084] Bilirubin value range does not include , correct the bilirubin value interval;

[0085] The logic for correcting the bilirubin value interval in the correction module is expressed as:

[0086] Recognize the structural similarity between the patient eye image captured by the camera module and the patient eye image A and the patient eye image B;

[0087] The patient's eye image captured by the camera module has a higher structural similarity with the patient's eye image A, then:

[0088] Corrected to ;

[0089] The patient's eye image captured by the camera module has a higher structural similarity with the patient's eye image B, then:

[0090] Corrected to ;

[0091] Among them, the ± selection of the bilirubin value interval follows: , Take + from ±, Middle ± Take ﹣; ;

[0092] Take the middle ±, Take + from ±;

[0093] in, is a constant, which is defined by the system user, and >1, and constant Make the corrected bilirubin value interval still within the pre-corrected bilirubin value interval;

[0094] Through the above logic formula, the bilirubin value range assessed by the user is determined and further corrected, so that the system finally outputs a more accurate bilirubin value range.

[0095] An output module, used for outputting the bilirubin value interval processed by the correction module;

[0096] The input module is interactively connected to a binding unit via a wireless network, the input module is interactively connected to a camera module via a wireless network, the camera module is interactively connected to a preprocessing unit and a comparison unit via a wireless network, the camera module is interactively connected to a retrieval module via a wireless network, the retrieval module is interactively connected to the input module via a wireless network, and the retrieval module is interactively connected to an analysis module, a correction module and an output module via a wireless network.

[0097] In this embodiment, the input module operates to input historical patient eye images and measured bilirubin value parameters, the binding unit synchronously monitors the patient eye images and the measured bilirubin value parameters in the input module, and binds the continuously input patient eye images and the measured bilirubin value parameters, the camera module further collects patient eye images, and uses the collected patient eye images to perform a structured similarity comparison with the historical patient eye images in the input module, the preprocessing unit synchronously receives the patient eye images collected by the camera module, converts the patient eye images into grayscale images, and outputs them, the comparison unit receives the patient eye images converted into grayscale images output by the preprocessing unit in real time, and uses each patient eye image stored in the input module as a comparison target with the received The grayscale image is compared for structured similarity, and then the retrieval module retrieves the two historical patient eye images with the best structured similarity comparison results in the camera module, and the difference ratio between the patient's bilirubin value and the eye image is identified based on the two retrieved historical patient eye images. The analysis module performs post-operation analysis on the similarity between the patient eye image captured by the camera module and the two patient eye images retrieved in the retrieval module, and the bilirubin value interval corresponding to the patient eye image captured by the camera module is determined based on the similarity analysis result. The correction module further receives the bilirubin value interval corresponding to the patient eye image determined in the analysis module, decides the output and correction of the bilirubin value interval based on the interval attribute, and finally outputs the bilirubin value interval processed by the correction module through the output module.

[0098] Through the system in the above embodiment, a large amount of prior data is used as a reference to collect user eye images to assess the user's bilirubin level. Compared with the existing technology, it is faster, relatively more accurate, and has good robustness.

[0099] like Figure 1 As shown, They respectively represent taking the minimum value and the maximum value in the brackets.

[0100] In summary, during operation, the system in the above embodiment is different from the existing technology. It uses machine vision technology to collect images of the user's eyes, and assesses the user's bilirubin level based on a large amount of prior data combined with the user's eyes. Compared with the existing blood test and sensor test, it is simpler and faster, and more accurate than the existing machine vision skin observation method. In addition, the system can continuously improve the system assessment accuracy based on the continuous accumulation of prior data, and use the bilirubin value range to provide feedback to the system end user, which is more convenient for the system end user to make a diagnosis and formulate and decide on treatment plans more quickly.

[0101] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A system for assessing bilirubin levels, characterized in that: include: An input module is used to input historical patient eye images and measured bilirubin value parameters; The camera module is used to collect the patient's eye image, and compare the collected patient's eye image with the historical patient's eye image in the input module for structured similarity; A retrieval module is used to retrieve two historical patient eye images with the best structural similarity comparison results in the camera module, and identify the difference ratio between the patient's bilirubin value and the eye image based on the two retrieved historical patient eye images; An analysis module, used for analyzing the similarity between the patient eye image acquired by the camera module and the two patient eye images retrieved by the retrieval module, and determining the bilirubin value interval corresponding to the patient eye image acquired by the camera module based on the similarity analysis result; A correction module, used for receiving the bilirubin value interval corresponding to the patient's eye image determined in the analysis module, and determining the output and correction of the bilirubin value interval based on the interval attribute; The output module is used to output the bilirubin value interval processed by the correction module.

2. A system for assessing bilirubin levels according to claim 1, characterized in that: The input module is internally provided with submodules, including: A binding unit, used for monitoring the patient's eye image and the measured bilirubin value parameter in the input module, and binding the continuously input patient's eye image and the measured bilirubin value parameter; Among them, after the input module inputs and binds the historical patient eye images and the measured bilirubin value parameters based on the binding unit, the eye images and bilirubin value parameters of each mutually bound patient are synchronously stored based on the input time sequence. The historical patient eye images and the measured bilirubin value parameters entered in the input module are derived from the EMR system background. When the retrieval module retrieves the patient eye image in the input module, the bilirubin value parameters corresponding to the bound patient eye image are synchronously retrieved.

3. A system for assessing bilirubin levels according to claim 1, characterized in that: The camera module is provided with submodules at the lower level, including: A preprocessing unit, used for receiving the patient's eye image acquired by the camera module, converting the patient's eye image into a grayscale image, and outputting the grayscale image; A comparison unit is used to receive the patient's eye image converted into a grayscale image output by the preprocessing unit, and to perform structural similarity comparison with the received grayscale image by taking each patient's eye image stored in the input module as a comparison target; Among them, when the comparison unit runs and applies the patient's eye image stored in the entry module as the comparison target, the patient's eye image stored in the entry module is compared sequentially based on the time sequence. When the patient's eye image in the entry module is applied to the comparison operation, it is synchronously processed by the preprocessing unit and converted into a grayscale image before performing the comparison operation.

4. A system for assessing bilirubin levels according to claim 3, characterized in that: The structured similarity comparison logic of the patient's eye image and the grayscale image in the comparison unit is expressed as follows: setting grayscale value intervals representing the pupil and the iris, setting grayscale value intervals representing the sclera, segmenting the pupil, iris area image and sclera area image in the patient's eye image and the grayscale image based on the grayscale value intervals representing the pupil and the iris and the grayscale value intervals representing the sclera, and capturing the center points of the pupil, iris area images and the center point of the sclera area image; ; Where: is the structural similarity between the patient's eye image and the grayscale image; The coordinates of the center point of the pupil and iris area image in the grayscale image are determined based on the pixel position; The coordinates of the center point of the pupil and iris area image in the patient's eye image are determined based on the pixel position; is the coordinates of the center point of the sclera area image in the grayscale image determined based on the pixel position; is the coordinate of the center point of the sclera area image in the patient's eye image based on the pixel position; wherein the structural similarity between the patient's eye image and the grayscale image The smaller the value, the higher the structural similarity between the patient's eye image and the grayscale image. Conversely, the lower the structural similarity between the patient's eye image and the grayscale image. Based on the above formula, the structural similarity between the grayscale image and each patient's eye image is calculated. The two historical patient eye images with the best comparison results retrieved by the module are The patient's eye image corresponding to the two calculation results with the smallest value.

5. A system for assessing bilirubin levels according to claim 1, characterized in that: The recognition logic of the difference ratio between the patient's bilirubin value and the eye image in the retrieval module is expressed as: ； Where: is the difference ratio between the patient's bilirubin value and the eye image; The similarity of the two patient eye images retrieved by the retrieval module in terms of color distribution; Bilirubin values ​​corresponding to two patient eye images retrieved by the retrieval module; , , The hue segmentation interval, the saturation segmentation interval, and the lightness segmentation interval are equally divided; is the pixel frequency of the patient eye image A and the patient eye image B in the i-th tone interval; is the pixel frequency of the patient's eye image A and the patient's eye image B in the jth saturation interval; The patient eye image A and the patient eye image B are in the The frequency of pixels in each brightness interval; Among them, the difference ratio between the patient's bilirubin value and the eye image The unit is , The unit is percentage.

6. A system for assessing bilirubin levels according to claim 5, characterized in that: The analysis logic of the similarity between the patient eye image collected by the camera module in the analysis module and the two patient eye images retrieved in the retrieval module is the same as The calculation logic is the same, so the analysis result is recorded as , C represents the patient's eye image collected by the current camera module, and further determines The size relationship of the three; In The bilirubin value range corresponding to the patient's eye image collected by the camera module is: ; Greater than , the bilirubin value range corresponding to the patient's eye image collected by the camera module is: Less than , the bilirubin value range corresponding to the patient's eye image collected by the camera module is: .

7. A system for assessing bilirubin levels according to claim 5 or 6, characterized in that: They respectively represent taking the minimum value and the maximum value in the brackets.

8. A system for assessing bilirubin levels according to claim 6, characterized in that: The logic of outputting and correcting the bilirubin value interval based on interval attribute decision in the correction module is: Interval attribute decision is to determine whether the bilirubin value interval contains ; The bilirubin value range includes , the bilirubin value interval is sent to the output module, and the output module performs the output operation of the bilirubin value interval; Bilirubin value range does not include , and correct the bilirubin value interval.

9. A system for assessing bilirubin levels according to claim 6, characterized in that: The logic for correcting the bilirubin value interval in the correction module is expressed as: Recognize the structural similarity between the patient eye image captured by the camera module and the patient eye image A and the patient eye image B; The patient's eye image captured by the camera module has a higher structural similarity with the patient's eye image A, then: Corrected to ; The patient's eye image captured by the camera module has a higher structural similarity with the patient's eye image B, then: Corrected to ; Among them, the ± selection of the bilirubin value interval follows: , Take + from ±, Middle ± Take ﹣; ; Take the middle ±, Take + from ±; in, is a constant, which is defined by the system user, and >1, and constant Make sure that the corrected bilirubin value interval is still within the bilirubin value interval before correction.

10. A system for assessing bilirubin levels according to claim 1, characterized in that: The entry module is interactively connected to a binding unit via a wireless network, the entry module is interactively connected to a camera module via a wireless network, the camera module is interactively connected to a preprocessing unit and a comparison unit via a wireless network, the camera module is interactively connected to a retrieval module via a wireless network, the retrieval module is interactively connected to the entry module via a wireless network, and the retrieval module is interactively connected to an analysis module, a correction module and an output module via a wireless network.

Images