Children OSA diagnosis method and system based on facial image and clinical data features

Abstract

The embodiment of the invention provides a child OSA diagnosis method and system based on facial images and clinical data features. The method comprises the steps that the age, gender, height, weight, facial images and iconography detection data of a to-be-detected child are acquired; performing feature extraction on the facial image through a predetermined facial recognition model to obtain facial features of the to-be-detected child; determining a model level based on age, gender, height and weight; selecting a target child OSA diagnosis model corresponding to the model level from the plurality of child OSA diagnosis models according to the model level; based on the facial features and the iconography detection data, performing child OSA diagnosis through a target child OSA diagnosis model, and determining an initial diagnosis result of the to-be-detected child; and performing secondary diagnosis according to the pre-acquired clinical data features and the initial diagnosis result of the to-be-detected child to obtain a final diagnosis result of the to-be-detected child. According to the scheme, the applicability and convenience of OSA diagnosis of children can be improved.

Description

Technical Field

[0001] This application relates to the field of intelligent medical technology, and in particular to a diagnostic method and system for pediatric OSA based on facial image and clinical data features. Background Technology

[0002] Obstructive sleep apnea (OSA) is a common sleep disorder in children (2-18 years old), with a prevalence of approximately 1%-6%. Compared to adults, children with OSA have unique manifestations, such as hyperactivity, emotional difficulties, declining academic performance, and inattention. Furthermore, childhood OSA can cause a variety of serious complications, leading to cardiovascular disease, growth and developmental disorders, cognitive impairment, craniofacial developmental abnormalities, and secretory otitis media.

[0003] Currently, the diagnosis of childhood OSA relies on polysomnography (PSG). However, PSG testing has many limitations, such as requiring a specialized sleep laboratory, connecting multiple measurement channels, and requiring overnight parental care. In summary, PSG testing is complex, uncomfortable, time-consuming, costly, and often results in poor child cooperation, making it unsuitable for rapid or large-scale screening. Summary of the Invention

[0004] The embodiments of this application aim to provide a method and system for diagnosing pediatric OSA based on facial image and clinical data features, which can improve the applicability and convenience of pediatric OSA diagnosis.

[0005] The technical solution of this application is implemented as follows: In a first aspect, embodiments of this application provide a method for diagnosing obstructive sleep apnea (OSA) in children based on facial images and clinical data features, the method comprising: Acquire basic information, facial images, and imaging data of the child to be tested; wherein, the basic information includes the child's age, gender, height, and weight; The facial features of the child to be detected are obtained by extracting features from the facial image using a pre-determined facial recognition model. Based on the age, gender, height, and weight, the model level is determined; and according to the model level, a target child OSA diagnostic model corresponding to the model level is selected from multiple child OSA diagnostic models. Based on the facial features and the imaging data, OSA diagnosis of the child is performed using the target child OSA diagnostic model to determine the initial diagnostic result of the child to be tested; By using the pre-acquired clinical data characteristics of the child to be tested and the initial diagnostic results, a secondary diagnosis is performed to obtain the final diagnostic result of the child to be tested.

[0006] In the above scheme, the step of extracting features from the facial image using a pre-determined facial recognition model to obtain the facial features of the child to be detected includes: The facial recognition model is used to extract features from the facial image to determine the position coordinates of the eyes, nose, lips and jaw. Based on the position coordinates corresponding to the eyes, nose, lips, and jaw, calculate the first distance between the eyes, the second distance between the eyes and the nose, the third distance between the upper and lower lips, the fourth distance between the upper lip and the nose, and the fifth distance between the lower lip and the jaw. The first distance, the second distance, the third distance, the fourth distance, and the fifth distance are analyzed respectively to determine multiple analysis results; The multiple analysis results are fused to determine the facial features of the child to be tested.

[0007] In the above scheme, determining the model level based on the age, gender, height, and weight includes: Based on the age and gender, determine the standard height and weight range for the child to be tested; The height and weight are compared with the standard height and weight range to obtain the comparison results; If the comparison results indicate that the height and weight are normal, then the model level is determined based on the age; If the comparison result indicates that the height and / or weight is abnormal, then the weight values ​​of the age, height and weight are determined respectively; and the age, height and weight are weighted according to the weight values ​​to obtain the corrected age, and the model level is determined based on the corrected age.

[0008] In the above scheme, the model levels include a first level, a second level, a third level, a fourth level, a fifth level, and a sixth level; wherein, the first level corresponds to 0-28 days, the second level corresponds to 1-12 months, the third level corresponds to 1-3 years, the fourth level corresponds to 4-6 years, the fifth level corresponds to 7-12 years, and the sixth level corresponds to 13-18 years. The step of selecting a pediatric OSA diagnostic model corresponding to the model level from multiple pediatric OSA diagnostic models includes: Based on one of the first level, the second level, the third level, the fourth level, the fifth level, and the sixth level, a target child OSA diagnostic model corresponding to the model level is selected from multiple child OSA diagnostic models; wherein, the first level, the second level, the third level, the fourth level, the fifth level, and the sixth level each correspond to a child OSA diagnostic model.

[0009] In the above scheme, the imaging data includes at least one of nasopharyngeal lateral radiographs, CT scans, and MRI scans. The process of diagnosing OSA in children based on the facial features and imaging data using the target child OSA diagnostic model, and determining the initial diagnostic result for the child to be tested, includes: Based on at least one of the nasopharyngeal lateral radiograph data, the CT examination data, and the MRI examination data, characteristic parameters of pediatric OSA are extracted; wherein, the characteristic parameters include adenoid parameters, airway parameters, and maxillofacial structure parameters; The facial features, adenoid parameters, airway parameters, and maxillofacial structure parameters are input into the target child OSA diagnostic model to perform child OSA diagnosis and determine the initial diagnostic result of the child to be tested.

[0010] In the above scheme, the clinical data characteristics include snoring frequency, mouth breathing information, tonsil grading, and adenoid hypertrophy degree; The process of performing a secondary diagnosis based on the pre-acquired clinical data characteristics of the child to be tested and the initial diagnostic results to obtain the final diagnostic result of the child to be tested includes: Determine the first assessment value corresponding to the initial diagnostic result; Based on the snoring frequency, the mouth breathing information, the tonsil grading, and the degree of adenoid hypertrophy, a secondary diagnosis of OSA in children is performed to obtain a second assessment value. The final evaluation value is obtained by superimposing the first evaluation value and the second evaluation value. The final assessment value is compared with the preset assessment value to determine the final diagnosis result of the child to be tested.

[0011] In the above scheme, before determining the model level based on the age, gender, height, and weight; and before selecting the target child OSA diagnostic model corresponding to the model level from multiple child OSA diagnostic models according to the model level, the method further includes: The study obtained historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data of multiple subjects; the age distribution of these subjects ranged from 0 days to 18 years. Based on the historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data, univariate, minimum absolute contraction, and selection operator regression analyses were performed to determine independent predictors of childhood OSA. Based on the independent predictive factors, a multivariate logistic regression was performed to construct a basic OSA model for children. The subjects were divided into multiple age groups based on their ages. Sub-independent predictive factors were determined for each of the six age groups based on their historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data. The multiple age groups included six age groups: 0-28 days, 1-12 months, 1-3 years, 4-6 years, 7-12 years, and 13-18 years. By adjusting the parameters of the basic pediatric OSA model using the aforementioned independent predictive factors, the multiple pediatric OSA diagnostic models are determined.

[0012] Secondly, embodiments of this application provide a pediatric OSA diagnostic system based on facial image and clinical data features, comprising: an acquisition unit, a determination unit, and a diagnostic unit; wherein, The acquisition unit is used to acquire basic information, facial images, and imaging detection data of the child to be tested; wherein, the basic information includes the age, gender, height, and weight of the child to be tested; and the facial features of the child to be tested are obtained by extracting features from the facial images using a pre-determined facial recognition model. The determining unit is configured to determine the model level based on the age, gender, height, and weight; and select a target child OSA diagnostic model corresponding to the model level from multiple child OSA diagnostic models according to the model level. The diagnostic unit is used to perform OSA diagnosis on the child based on the facial features and the imaging data using the pediatric OSA diagnostic model, and determine the initial diagnostic result of the child to be tested; and to perform a secondary diagnosis using the pre-acquired clinical data features of the child to be tested and the initial diagnostic result, to obtain the final diagnostic result of the child to be tested.

[0013] Thirdly, embodiments of this application provide a pediatric OSA diagnostic device based on facial image and clinical data features, the pediatric OSA diagnostic device based on facial image and clinical data features comprising: a processor and a memory; wherein, The memory is used to store computer programs; The processor is configured to call and run the computer program from the memory to perform the method as described in the first aspect.

[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing executable instructions for causing a processor to perform the method described in the first aspect.

[0015] This application provides a method and system for diagnosing obstructive sleep apnea (OSA) in children based on facial image and clinical data features. The method includes: acquiring basic information, facial images, and imaging data of the child to be tested; wherein the basic information includes the child's age, gender, height, and weight; extracting features from the facial images using a pre-defined facial recognition model to obtain the child's facial features; determining a model level based on the age, gender, height, and weight; selecting a target OSA diagnostic model corresponding to the model level from multiple OSA diagnostic models according to the model level; performing OSA diagnosis using the target model based on the facial features and imaging data to determine an initial diagnosis result for the child; and performing a secondary diagnosis using the pre-acquired clinical data features of the child and the initial diagnosis result to obtain a final diagnosis result for the child. The above-described scheme determines the model level based on age, gender, height, and weight. Then, based on the model level, a target OSA diagnostic model corresponding to the chosen model level is selected from multiple pediatric OSA diagnostic models. OSA diagnosis is performed using this target model based on facial features and imaging data. Matching the target model to the child improves the accuracy of the diagnosis. Furthermore, the initial diagnosis is corrected based on the child's clinical data characteristics, further enhancing accuracy. Simultaneously, since the diagnosis only uses the child's basic information, facial images, and imaging data, without involving polysomnography, and these data are readily available, the applicability and convenience of pediatric OSA diagnosis are improved. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the technical solutions of this application. Obviously, the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0018] Figure 1 This application provides an optional flowchart illustrating a pediatric OSA diagnostic method based on facial image and clinical data features. Figure 1 ; Figure 2 This application provides an optional flowchart illustrating a pediatric OSA diagnostic method based on facial image and clinical data features. Figure 2 ; Figure 3 This application provides a schematic diagram of the structure of a pediatric OSA diagnostic system based on facial image and clinical data features. Figure 4 This application provides a schematic diagram of the structure of a pediatric OSA diagnostic device based on facial image and clinical data features. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the specific technical solutions of this application will be further described in detail below with reference to the accompanying drawings of the embodiments of this application. The following embodiments are used to illustrate this application, but are not intended to limit the scope of this application.

[0020] 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 belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] In the following description, references to "some embodiments," "this embodiment," "this application embodiment," and examples, etc., describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subset of all possible embodiments and may be combined with each other without conflict.

[0022] If the application documents contain similar descriptions such as "first / second", the following explanation shall be added: In the following description, the terms "first / second / third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first / second / third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0023] This application provides a method for diagnosing obstructive sleep apnea (OSA) in children based on facial images and clinical data features. Figure 1 This application provides an optional flowchart illustrating a pediatric OSA diagnostic method based on facial image and clinical data features. Figure 1 , will combine Figure 1 The steps shown are explained.

[0024] S101. Obtain basic information, facial images, and imaging data of the child to be tested; among which, the basic information includes the child's age, gender, height, and weight.

[0025] In some embodiments of this application, the pediatric OSA diagnostic method based on facial image and clinical data features is applicable to pediatric OSA assisted diagnostic scenarios.

[0026] In some embodiments of this application, the subject of the pediatric OSA diagnosis method based on facial image and clinical data features is a pediatric OSA diagnosis device based on facial image and clinical data features.

[0027] S102. Extract features from the facial image using a pre-determined facial recognition model to obtain the facial features of the child to be detected.

[0028] In some embodiments of this application, a facial recognition model is used to extract features from a facial image to determine the position coordinates of the eyes, nose, lips, and jaw. Based on the position coordinates of the eyes, nose, lips, and jaw, a first distance between the eyes, a second distance between the eyes and nose, a third distance between the upper and lower lips, a fourth distance between the upper lip and nose, and a fifth distance between the lower lip and jaw are calculated. The first, second, third, fourth, and fifth distances are analyzed separately to determine multiple analysis results. The multiple analysis results are then fused to determine the facial features of the child to be detected.

[0029] S103. Determine the model level based on age, gender, height, and weight; and select the target child OSA diagnostic model corresponding to the model level from multiple child OSA diagnostic models according to the model level.

[0030] In some embodiments of this application, the standard height and weight range of the child to be tested is determined based on age and gender; the height and weight are compared with the standard height and weight range to obtain the comparison result; if the comparison result indicates that the height and weight are normal, the model level is determined based on age; if the comparison result indicates that the height and / or weight are abnormal, the weight values ​​of age, height and weight are determined respectively; and the age, height and weight are weighted according to the weight values ​​to obtain the corrected age, and the model level is determined based on the corrected age.

[0031] In some embodiments of this application, the model levels include a first level, a second level, a third level, a fourth level, a fifth level, and a sixth level; wherein, the first level corresponds to 0-28 days, the second level corresponds to 1-12 months, the third level corresponds to 1-3 years, the fourth level corresponds to 4-6 years, the fifth level corresponds to 7-12 years, and the sixth level corresponds to 13-18 years. In some embodiments of this application, a target child OSA diagnostic model corresponding to a model level is selected from multiple child OSA diagnostic models based on one of the following levels: first level, second level, third level, fourth level, fifth level, and sixth level; wherein, the first level, second level, third level, fourth level, fifth level, and sixth level each correspond to a child OSA diagnostic model.

[0032] S104. Based on facial features and imaging data, perform OSA diagnosis on children using the target child OSA diagnostic model to determine the initial diagnostic result for the child to be tested.

[0033] In some embodiments of this application, the imaging data includes at least one of nasopharyngeal lateral radiographs, CT scans, and MRI scans.

[0034] In some embodiments of this application, feature parameters of pediatric OSA are extracted based on at least one of nasopharyngeal lateral radiographs, CT scans, and MRI scans; wherein the feature parameters include adenoid parameters, airway parameters, and maxillofacial structural parameters; the facial features, adenoid parameters, airway parameters, and maxillofacial structural parameters are input into a target pediatric OSA diagnostic model to diagnose pediatric OSA and determine the initial diagnostic result of the child to be tested.

[0035] For example, lateral nasopharyngeal radiographs include adenoid thickness to nasopharyngeal cavity ratio and adenoid hypertrophy; CT scans include upper airway narrowing, retropharyngeal space, and intertonsillar distance; MRI scans include airway changes. S105. Based on the pre-acquired clinical data characteristics and initial diagnostic results of the child to be tested, a secondary diagnosis is performed to obtain the final diagnostic result of the child to be tested.

[0036] In some embodiments of this application, clinical data features include snoring frequency, mouth breathing information, tonsil grading, and adenoid hypertrophy.

[0037] In some embodiments of this application, a first assessment value corresponding to the initial diagnostic result is determined; a second assessment value is obtained by performing a secondary diagnosis of OSA in children based on snoring frequency, mouth breathing information, tonsil grading, and adenoid hypertrophy; the first and second assessment values ​​are superimposed to obtain a final assessment value; and the final assessment value is compared with a preset assessment value to determine the final diagnostic result of the child to be tested.

[0038] Understandably, a model level is determined based on age, gender, height, and weight. Then, based on the model level, a target OSA diagnostic model corresponding to the chosen model level is selected from multiple pediatric OSA diagnostic models. Based on facial features and imaging data, OSA diagnosis is performed using this target model. Since the target model matches the child being tested, the accuracy of the diagnosis is improved. Furthermore, the initial diagnosis is corrected based on the clinical data characteristics of the child being tested, further enhancing the accuracy. Simultaneously, because the pediatric OSA diagnosis process only uses the child's basic information, facial images, and imaging data, without involving polysomnography, and because these basic information, facial images, and imaging data are readily available, the applicability and convenience of pediatric OSA diagnosis are improved.

[0039] In some embodiments of this application, S102 can be implemented by S201-S204, as follows: S201. Extract features from the facial image using a facial recognition model to determine the position coordinates of the eyes, nose, lips, and jaw.

[0040] S202. Based on the position coordinates of the eyes, nose, lips and jaw respectively, calculate the first distance between the eyes, the second distance between the eyes and the nose, the third distance between the upper lip and the lower lip, the fourth distance between the upper lip and the nose, and the fifth distance between the lower lip and the jaw.

[0041] S203. Analyze the first distance, second distance, third distance, fourth distance and fifth distance respectively, and determine multiple analysis results.

[0042] S204. Perform data fusion on multiple analysis results to determine the facial features of the child to be tested.

[0043] For example, for a child to be tested, a facial recognition model is used to extract features from the child's facial image to determine the first coordinates corresponding to the eyes, the second coordinates corresponding to the nose, the third coordinates corresponding to the lips, and the fourth coordinates corresponding to the jaw. The first coordinates include the first left coordinate corresponding to the left eye and the first right coordinate corresponding to the right eye; the third coordinates include the third upper coordinate corresponding to the upper lip and the third lower coordinate corresponding to the lower lip. Based on the first left and first right coordinates, the first distance between the eyes is calculated. The midpoint coordinate is obtained by taking the median of the first left and first right coordinates. Using the midpoint coordinate and the second coordinate, the second distance between the eyes and the nose is calculated. Based on the third upper and third lower coordinates, the third distance is calculated. Based on the third upper and second coordinates, the fourth distance between the upper lip and the nose is calculated. Based on the third lower and fourth coordinates, the fifth distance between the lower lip and the jaw is calculated.

[0044] The first, second, third, fourth, and fifth distances are compared and analyzed with their respective preset range values ​​to obtain the analysis results for each distance. The analysis results for each of the five distances are then fused to obtain the facial features of the child being tested. These facial features include the analysis results for each of the five distances, as well as the combined features obtained by fusing the five analysis results.

[0045] Understandably, in the process of extracting features from facial images to determine the facial features of the child to be tested, the facial organ features are analyzed separately, and the results of the separate analyses are fused together, making the determined facial features more representative, thereby improving the accuracy of diagnosing OSA in children.

[0046] In some embodiments of this application, such as Figure 2 As shown, S106-S109 are executed before S103, as follows: S106. Obtain historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data of multiple subjects; among them, the age distribution of multiple subjects ranges from 0 days to 18 years.

[0047] S107. Based on historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data, perform univariate, minimum absolute contraction, and selection operator regression analyses to determine independent predictors of childhood OSA.

[0048] S108. Based on independent predictors, perform multivariate logistic regression to construct a basic OSA model for children.

[0049] S109. Divide the subjects into multiple age groups according to their ages; and determine the sub-independent predictive factors for each of the six age groups based on their historical basic information, historical OSA symptoms, historical facial features, historical clinical signs and historical imaging data. The multiple age groups include six age groups: 0-28 days, 1-12 months, 1-3 years, 4-6 years, 7-12 years and 13-18 years.

[0050] S1010. By using sub-independent predictive factors, the parameters of the basic model of childhood OSA are adjusted to determine multiple diagnostic models for childhood OSA.

[0051] Understandably, by constructing high-precision, low-cost OSA diagnostic models for different age groups, non-invasive and efficient screening has replaced complex and expensive PSG testing, providing an objective and reliable solution for early screening and hierarchical diagnosis and treatment of OSA in children. It is especially suitable for large-scale application in areas with scarce medical resources, improving the applicability and convenience of OSA diagnosis in children.

[0052] Based on the above embodiments of the pediatric OSA diagnostic method based on facial image and clinical data features, this application also provides a pediatric OSA diagnostic system based on facial image and clinical data features, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of a pediatric OSA diagnostic system based on facial image and clinical data features, provided in an embodiment of this application. The pediatric OSA diagnostic system 3 based on facial image and clinical data features includes: an acquisition unit 301, a determination unit 302, and a diagnostic unit 303; wherein, The acquisition unit 301 is used to acquire basic information, facial images, and imaging detection data of the child to be tested; wherein, the basic information includes the age, gender, height, and weight of the child to be tested; and the facial features of the child to be tested are obtained by extracting features from the facial images using a pre-determined facial recognition model. The determining unit 302 is used to determine the model level based on the age, gender, height, and weight; and select a target child OSA diagnostic model corresponding to the model level from multiple child OSA diagnostic models according to the model level. The diagnostic unit 303 is used to perform OSA diagnosis on the child based on the facial features and the imaging detection data using the child OSA diagnostic model, and determine the initial diagnostic result of the child to be tested; and to perform a secondary diagnosis using the pre-acquired clinical data features of the child to be tested and the initial diagnostic result, to obtain the final diagnostic result of the child to be tested.

[0053] Based on the above embodiments of the pediatric OSA diagnostic method based on facial image and clinical data features, this application also provides a pediatric OSA diagnostic device based on facial image and clinical data features, such as... Figure 4 As shown, Figure 4 This is a schematic diagram of a pediatric OSA diagnostic device based on facial image and clinical data features, provided in an embodiment of this application. The pediatric OSA diagnostic device 4 includes a processor 401 and a memory 402. The memory 402 stores a computer program; the processor 401 retrieves and runs the computer program from the memory to execute the pediatric OSA diagnostic method based on facial image and clinical data features as described in the above embodiment.

[0054] In the embodiments of this application, the processor 401 described above can be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor. It is understood that for different devices, the electronic device used to implement the above processor function can also be other types, and the embodiments of this application do not specifically limit it.

[0055] This application provides a computer-readable storage medium storing a computer program for implementing, when executed by a processor, a pediatric OSA diagnostic method based on facial image and clinical data features as described in any of the above embodiments.

[0056] For example, the program instructions corresponding to a pediatric OSA diagnosis method based on facial image and clinical data features in this embodiment can be stored on a storage medium such as an optical disc, hard disk, or USB flash drive. When the program instructions corresponding to the pediatric OSA diagnosis method based on facial image and clinical data features in the storage medium are read or executed by an electronic device, the pediatric OSA diagnosis method based on facial image and clinical data features as described in any of the above embodiments can be realized.

[0057] Furthermore, the functional modules in the embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional module.

[0058] If the integrated unit is implemented as a software functional module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the method of this embodiment. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0059] It should be understood that the phrases "one embodiment," "an embodiment," or "some embodiments" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to an embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment," "in one embodiment," or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments; their similarities or commonalities can be referred to mutually, and for the sake of brevity, they will not be repeated here.

[0060] The modules described above as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules. They may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0061] In addition, each functional module in the various embodiments of this application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated modules can be implemented in hardware or in the form of hardware plus software functional units.

[0062] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0063] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0064] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0065] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0066] The above description is merely an embodiment of this application, but the protection scope of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A diagnostic method for pediatric OSA based on facial image and clinical data features, characterized in that, The method includes: Acquire basic information, facial images, and imaging data of the child to be tested; wherein, the basic information includes the child's age, gender, height, and weight; The facial features of the child to be detected are obtained by extracting features from the facial image using a pre-determined facial recognition model. Based on the age, gender, height, and weight, the model level is determined; and according to the model level, a target child OSA diagnostic model corresponding to the model level is selected from multiple child OSA diagnostic models. Based on the facial features and the imaging data, OSA diagnosis of the child is performed using the target child OSA diagnostic model to determine the initial diagnostic result of the child to be tested; By using the pre-acquired clinical data characteristics of the child to be tested and the initial diagnostic results, a secondary diagnosis is performed to obtain the final diagnostic result of the child to be tested.

2. The method according to claim 1, characterized in that, The step of extracting features from the facial image using a pre-determined facial recognition model to obtain the facial features of the child to be detected includes: The facial recognition model is used to extract features from the facial image to determine the position coordinates of the eyes, nose, lips and jaw. Based on the position coordinates corresponding to the eyes, nose, lips, and jaw, calculate the first distance between the eyes, the second distance between the eyes and the nose, the third distance between the upper and lower lips, the fourth distance between the upper lip and the nose, and the fifth distance between the lower lip and the jaw. The first distance, the second distance, the third distance, the fourth distance, and the fifth distance are analyzed respectively to determine multiple analysis results; The multiple analysis results are fused to determine the facial features of the child to be tested.

3. The method according to claim 1, characterized in that, The process of determining the model level based on the age, gender, height, and weight includes: Based on the age and gender, determine the standard height and weight range for the child to be tested; The height and weight are compared with the standard height and weight range to obtain the comparison results; If the comparison results indicate that the height and weight are normal, then the model level is determined based on the age; If the comparison result indicates that the height and / or weight is abnormal, then the weight values ​​of the age, height and weight are determined respectively; and the age, height and weight are weighted according to the weight values ​​to obtain the corrected age, and the model level is determined based on the corrected age.

4. The method according to claim 1, characterized in that, The model levels include Level 1, Level 2, Level 3, Level 4, Level 5, and Level 6; wherein Level 1 corresponds to 0-28 days, Level 2 corresponds to 1-12 months, Level 3 corresponds to 1-3 years, Level 4 corresponds to 4-6 years, Level 5 corresponds to 7-12 years, and Level 6 corresponds to 13-18 years. The step of selecting a pediatric OSA diagnostic model corresponding to the model level from multiple pediatric OSA diagnostic models includes: Based on one of the first level, the second level, the third level, the fourth level, the fifth level, and the sixth level, a target child OSA diagnostic model corresponding to the model level is selected from multiple child OSA diagnostic models; wherein, the first level, the second level, the third level, the fourth level, the fifth level, and the sixth level each correspond to a child OSA diagnostic model.

5. The method according to claim 1, characterized in that, The imaging data includes at least one of the following: nasopharyngeal lateral radiograph data, CT scan data, and MRI scan data. The process of diagnosing OSA in children based on the facial features and imaging data using the target child OSA diagnostic model, and determining the initial diagnostic result for the child to be tested, includes: Based on at least one of the nasopharyngeal lateral radiograph data, the CT examination data, and the MRI examination data, characteristic parameters of pediatric OSA are extracted; wherein, the characteristic parameters include adenoid parameters, airway parameters, and maxillofacial structure parameters; The facial features, adenoid parameters, airway parameters, and maxillofacial structure parameters are input into the target child OSA diagnostic model to perform child OSA diagnosis and determine the initial diagnostic result of the child to be tested.

6. The method according to claim 1, characterized in that, The clinical data features include snoring frequency, mouth breathing information, tonsil grading, and degree of adenoid hypertrophy. The process of performing a secondary diagnosis based on the pre-acquired clinical data characteristics of the child to be tested and the initial diagnostic results to obtain the final diagnostic result of the child to be tested includes: Determine the first assessment value corresponding to the initial diagnostic result; Based on the snoring frequency, the mouth breathing information, the tonsil grading, and the degree of adenoid hypertrophy, a secondary diagnosis of OSA in children is performed to obtain a second assessment value. The final evaluation value is obtained by superimposing the first evaluation value and the second evaluation value. The final assessment value is compared with the preset assessment value to determine the final diagnosis result of the child to be tested.

7. The method according to claim 1, characterized in that, Before determining the model level based on the age, gender, height, and weight; and before selecting a target pediatric OSA diagnostic model corresponding to the model level from multiple pediatric OSA diagnostic models according to the model level, the method further includes: The study obtained historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data of multiple subjects; the age distribution of these subjects ranged from 0 days to 18 years. Based on the historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data, univariate, minimum absolute contraction, and selection operator regression analyses were performed to determine independent predictors of childhood OSA. Based on the independent predictive factors, a multivariate logistic regression was performed to construct a basic OSA model for children. The subjects were divided into multiple age groups based on their ages. Sub-independent predictive factors were determined for each of the six age groups based on their historical basic information, historical OSA symptoms, historical facial features, historical clinical signs, and historical imaging data. The multiple age groups included six age groups: 0-28 days, 1-12 months, 1-3 years, 4-6 years, 7-12 years, and 13-18 years. By adjusting the parameters of the basic pediatric OSA model using the aforementioned independent predictive factors, the multiple pediatric OSA diagnostic models are determined.

8. A diagnostic system for pediatric OSA based on facial image and clinical data features, characterized in that, include: Acquisition unit, determination unit, and diagnosis unit; among which, The acquisition unit is used to acquire basic information, facial images, and imaging detection data of the child to be tested; wherein, the basic information includes the age, gender, height, and weight of the child to be tested; and the facial features of the child to be tested are obtained by extracting features from the facial images using a pre-determined facial recognition model. The determining unit is configured to determine the model level based on the age, gender, height, and weight; and select a target child OSA diagnostic model corresponding to the model level from multiple child OSA diagnostic models according to the model level. The diagnostic unit is used to perform OSA diagnosis on the child based on the facial features and the imaging data using the pediatric OSA diagnostic model, and determine the initial diagnostic result of the child to be tested; and to perform a secondary diagnosis using the pre-acquired clinical data features of the child to be tested and the initial diagnostic result, to obtain the final diagnostic result of the child to be tested.

9. A diagnostic device for pediatric OSA based on facial image and clinical data features, characterized in that, include: Processor and memory, of which, The memory is used to store computer programs; The processor is configured to call and run the computer program from the memory to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores executable instructions for causing a processor to execute, thereby implementing the method of any one of claims 1 to 7.

Images