Marker and model for diagnosing and / or predicting caries of preschool children and construction method of marker and model

Abstract

The invention relates to the field of early diagnosis of deciduous tooth caries of preschool children, in particular to a marker and a model for diagnosing and / or predicting caries of preschool children and a construction method thereof, the marker comprises age, urban and rural classification of residence, feeding modes of children within 6 months after birth, whether toothache experiences exist or not, and whether teeth are seen or not. Discrimination indexes of the constructed prediction model comprise AUC, sensitivity and specificity, the AUC, the sensitivity and the specificity are respectively 0.746, 0.765 and 0.624 in a model development set, 0.717, 0.680 and 0.678 in an internal verification set and 0.804, 0.807 and 0.660 in an external verification set, a calibration curve shows that the model prediction accuracy is good, and the model prediction accuracy is high. Clinical decision analysis and a clinical influence curve show that the prediction model conforms to clinical application conditions within a reasonable threshold range.

Description

Technical Field

[0001] The present invention relates to the technical field of disease diagnosis, and in particular to a marker, a model and a construction method thereof for diagnosing and / or predicting dental caries in preschool children. Background Art

[0002] Early childhood caries (ECC), which refers to the presence of one or more deciduous teeth in the mouth of children aged 6 years and below that are decayed, filled or missing due to caries, is the most common chronic disease among children worldwide. Deciduous tooth caries in preschool children have an important impact on the eruption and development of permanent teeth, oral health and overall health of children; deciduous tooth caries can also become a source of infection in the oral cavity. When deciduous tooth caries develop into endodontic periapical periodontitis, it can cause enamel hypoplasia of the permanent tooth germ underneath, such as the occurrence of Turner teeth. In addition, local alveolar bone destruction, abnormal root resorption, and residual root retention caused by periapical periodontitis of deciduous teeth can all cause abnormal eruption order and position of the successor permanent teeth; in addition, deciduous tooth caries will also bring a heavy economic burden to families and society.

[0003] However, caries of primary teeth is often ignored by parents because it progresses quickly and often has no symptoms. In the 5-year-old age group, 75.4% of caries lesions occur in 1 / 3 of children, indicating that ECC has the characteristics of skewed distribution, that is, some people will show a higher risk of caries due to various factors. The national oral health knowledge awareness rate of children's parents is 62.3%. Most parents only have some knowledge of oral diseases, which leads to the guardians referring young children to pediatric dentists only when they experience severe pain and late consequences of caries (such as abscesses and cellulitis). Therefore, it is urgent to develop a risk prediction tool for primary tooth caries to identify preschool children with a high risk of caries, allowing guardians to assess the risk of caries in preschool children without clinical examinations and professional backgrounds, so as to guide the timely dental referral of children at high risk of caries in community settings.

[0004] There are four main caries risk assessment systems widely used in clinical practice, namely the caries risk assessment tool proposed by AAPD, the caries risk assessment form proposed by ADA, the CAMBRA (Caries management by risk assessment) caries risk assessment management system, and the Cariogram system. However, the above caries risk assessment systems were all proposed by foreign professional academic institutions on the basis of evidence. Among them, CAMBRA, ADA and CAT all divide the assessment form into two parts: under 6 years old and over 6 years old. The CAMBRA items are relatively cumbersome, and when used for preschool children, it may overestimate the individual caries risk. CAT has been revised several times, but the application evaluation of its latest revised version has not been reported. Although the ADA system can be obtained through questionnaires and examinations, its predictive ability has yet to be verified. Cariogram uses software to quantitatively calculate the risk of caries, but its sensitivity in predicting caries in young children is low. The above caries prediction risk assessment systems are rarely used in my country. Of course, there are also reports on caries risk assessment systems in China. For example, the invention patent CN113284622A discloses a method, system and storage medium for assessing caries risk in young children. According to the target child data, a caries risk assessment model for young children based on the Logistic regression model is constructed and the caries risk probability of the target child is obtained, and then a caries risk report and written family care guidance for the target child are generated; the invention patent CN113035300A discloses a child caries risk assessment system, including a mobile phone APP, an assessment system applet, a WeChat cloud platform, a developer server and a WeChat interface service API. However, the etiology of caries is complex, and the use of multivariate caries risk assessment tools can more accurately predict the occurrence of caries. Using large data sets (such as national oral health epidemiological survey data or data collected in daily diagnosis and treatment routines) to establish a machine learning prediction model will provide an effective strategy for quantifying caries risk factors and predicting the probability of caries.

[0005] In view of the above problems, the present invention provides a diagnostic model for predicting dental caries in preschool children. Based on the big data of the national oral health epidemiological survey, the optimal prediction factors of the model are determined through LASSO regression analysis, univariate and multivariate Logistic regression analysis, and then a preschool children's deciduous tooth caries prediction model based on the Logistic regression model is constructed to predict the probability of caries in the deciduous teeth of the test children. The model is further clinically applied in the form of a web calculator using HTML+CSS+JavaScript technology. Summary of the invention

[0006] The first object of the present invention is to provide a marker for diagnosing and / or predicting dental caries in preschool children, characterized in that the marker includes one or more combinations of age, place of residence, feeding method within 6 months after the child is born, whether the child has experienced toothache, and whether the child has seen a dentist.

[0007] Preferably, the markers include age, place of residence, feeding method within 6 months after the child is born, whether there is toothache experience and whether the child has seen a dentist.

[0008] Preferably, the markers consist of age, place of residence, feeding method within 6 months after the child is born, whether there is toothache experience and whether the child has seen a dentist.

[0009] The second object of the present invention is to provide the use of the marker for diagnosing and / or predicting dental caries in preschool children in the preparation of a diagnostic and / or predictive reagent product, database or prediction model for dental caries in preschool children.

[0010] The third object of the present invention is to provide a prediction model for diagnosing and / or predicting dental caries in preschool children, wherein the prediction model for predicting dental caries in preschool children is constructed using the markers.

[0011] The fourth object of the present invention is to provide a method for constructing a prediction model for diagnosing and / or predicting dental caries in preschool children, comprising the following steps:

[0012] (1) Obtaining the data set: Collecting oral epidemiological data of preschool children and dividing the obtained data set into model development set, internal validation set, and external validation set;

[0013] (2) Variable screening: The baseline data obtained in step (1) were divided into a dental caries group and a healthy group. The LASSO regression algorithm was used to eliminate collinearity in the baseline data in the model development set and to screen for risk factors (the s1 coefficient was not 0);

[0014] (3) Model construction: The risk factors obtained in step (2) were included in univariate and multivariate logistic regression analysis to determine the model predictors, and risk factors with P < 0.05 were selected for modeling;

[0015] (4) Model validation: The prediction model obtained in step (3) was used to test the model’s discrimination (ROC curve), calibration (calibration curve), and clinical net benefit (DCA curve, CIC curve) in the model development set, internal validation set, and external validation set;

[0016] (5) Construction of prediction model nomogram: Diagnose and / or predict dental caries in preschool children through the predictive factors described in step (3), each predictive factor has a prediction reference value based on the risk ratio of the statistical results, the prediction reference value shows the weight of each parameter, and the total score will distinguish between healthy individuals and children with dental caries.

[0017] Preferably, the regression model formula in step (3) is:

[0018] Among them, Y represents the output value, P represents the probability of dental caries, a represents the constant term, β represents the regression coefficient, and X is the predictor.

[0019] A fifth object of the present invention is to provide a system for diagnosing and / or predicting dental caries in preschool children, which is applied to the construction method, comprising:

[0020] (1) a data collection module, at least used for data collection and obtaining a sample data set;

[0021] (2) a data processing module, at least used to extract valid samples from the sample data set that can be used to build an evaluation model;

[0022] (3) a model building module, at least for randomly dividing the complete data set of the valid samples into a training set and a validation set, fitting the training set using LASSO regression and Logistics regression methods, and recording the optimal model parameters;

[0023] (4) a probability calculation module, at least for calculating the probability of dental caries according to the prediction model formula;

[0024] (5) A result display module, at least used to display the prediction results.

[0025] A fifth object of the present invention is to provide a computer program product stored on a computer-readable medium, comprising a computer-readable program, which, when executed on an electronic device, provides a user input interface for applying the system for treating dental caries in preschool children.

[0026] The sixth object of the present invention is to provide a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to apply the system for treating dental caries in preschool children.

[0027] The beneficial effects of the present invention are as follows: (1) The present invention studies the baseline data of preschool children, and screens out a marker combination consisting of age, place of residence, feeding method within 6 months after birth, whether the child has experienced toothache, and whether the child has seen a dentist through a machine learning model; (2) The marker combination obtained by screening is used to construct a prediction model for caries in preschool children, and the AUC (95% CI) of the prediction model in the model development set, internal validation set, and external validation set are 0.746 (0.714-0.777), 0.717 (0.666-0.768), and 0.804 (0.765-0.844), respectively. The calibration curve shows that the model prediction accuracy is good, and the clinical decision analysis and clinical impact curve show that the prediction model meets the clinical application situation within a reasonable threshold range. The prediction model described in the present invention has easy-to-understand model prediction factors and low professional and technical barriers, which provides an effective tool for the management of deciduous tooth caries in preschool children in the community. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Case screening and model building flow chart

[0029] Figure 2 LASSO regression analysis to screen variables

[0030] Figure 3 Predictive performance of the prediction model in the model development set Note: (A) ROC curve; (B) calibration curve; (C) clinical decision analysis; (D) clinical impact curve;

[0031] Figure 4 Predictive performance of the prediction model in the internal validation set Note: (A) ROC curve; (B) calibration curve; (C) clinical decision analysis; (D) clinical impact curve;

[0032] Figure 5 Predictive performance of the prediction model in the external validation set Note: (A) ROC curve; (B) calibration curve; (C) clinical decision analysis; (D) clinical impact curve;

[0033] Figure 6 Subgroup analysis of the predictive ability of the prediction model in urban and rural children Note: (A) ROC curve of the prediction model in urban children; (B) calibration curve of the prediction model in urban children; (C) ROC curve of the prediction model in rural children; (D) calibration curve of the prediction model in rural children;

[0034] Figure 7 Nomogram

[0035] Figure 8 Web Calculator DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 should fall within the scope of protection of the present invention.

[0037] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0038] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0039] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0040] Embodiment: A diagnostic model for predicting dental caries in preschool children

[0041] 1. Inclusion and exclusion criteria for preschool children

[0042] Inclusion criteria: preschool children aged 3-5 years in Gansu Province (residence time ≥6 months), with informed consent from the children's guardians.

[0043] Exclusion criteria: children who could not cooperate during the oral examination, children with missing clinical data, and children with missing questionnaire data.

[0044] 2. Caries diagnosis criteria

[0045] No caries: A healthy crown without a filling or caries-related sign is recorded as a caries-free tooth. Pre-cavity and similar early caries are not recorded as caries because the diagnosis is unreliable. The following conditions are not diagnosed as crown caries: (1) chalky spots; (2) discolored or rough spots on the crown, with no tissue softening felt when probing with the CPI probe; (3) stained pits and fissures on the enamel surface, but no visible subenamel undermining, and no softening of the cavity bottom or groove wall detected by the CPI probe; (4) hard, dark, concave defects on the enamel caused by moderate to severe fluorosis; (5) abrasion of the enamel surface; (6) wedge-shaped defects without caries.

[0046] Crown caries: The crown of a tooth with obvious caries, obvious subenamel destruction, and clear lesions that can be probed and softened on the bottom or wall of the cavity is recorded as crown caries; temporary fillings and pit and fissure sealants on the tooth with caries are counted as crown caries. CPI probes should be used to confirm the caries judged by the clinic on the occlusal surface and buccal and lingual surface. If there is any doubt, it should not be recorded as crown caries.

[0047] Filled with caries: A tooth with one or more permanent fillings made due to caries on the crown and accompanied by one or more caries lesions is recorded as filled with caries, without distinguishing between primary caries and secondary caries (i.e. regardless of whether the caries lesions are related to the filling).

[0048] Filled and caries-free: A tooth crown with one or more permanent fillings made due to caries and without caries in any part is recorded as filled and caries-free; a tooth that has a crown restored due to caries is also recorded as filled and caries-free.

[0049] Loss due to caries: Primary teeth extracted due to caries. This score is only used for the loss of primary teeth that cannot be explained by normal replacement teeth.

[0050] 3. Description of predictors and outcome variables

[0051] The data sets required to build the model are shown in Table 1, including:

[0052] (1) Sociodemographic characteristics: gender, age, place of residence, household registration type, and ethnicity;

[0053] (2) Guardian status: guardian’s highest education level, guardian’s knowledge of oral health, guardian’s oral health attitude, guardian’s evaluation of the child’s overall health status, and guardian’s evaluation of the child’s oral health status;

[0054] (3) Early life habits: feeding method within 6 months after birth, age when children start brushing teeth, frequency of eating sweet snacks, frequency of consuming sweetened drinks, frequency of eating sweetened foods, and whether to eat desserts before going to bed;

[0055] (4) Oral hygiene habits: whether the child brushes teeth, how often the child brushes teeth, whether the guardian helps the child brush teeth, and whether the child uses toothpaste;

[0056] (5) Dental treatment experience: whether the patient has experienced toothache and whether the patient has seen a dentist.

[0057] Table 1 Baseline characteristics of all patients in the model development set and internal validation set

[0058]

[0059]

[0060]

[0061]

[0062]

[0063] 4. Variable screening (LASSO regression)

[0064] The least absolute shrinkage and selection operator (LASSO) regression was used to identify potential risk factors, and the coefficients were continuously compressed by adding a penalty function to deal with the multicollinearity of the predictor variables. Figure 2 The LASSO coefficient curves of 23 clinical characteristics are shown. The optimal penalty coefficient λ was determined using 10-fold cross validation and the minimum value criterion, and a dashed vertical line was drawn at the λ value. The s1 coefficients of the 23 clinical characteristics are shown in Table 2. Variables with s1 coefficients not equal to 0 were selected for subsequent analysis, namely: age, place of residence, guardian's highest education, guardian's evaluation of the child's oral health status, feeding method within 6 months after birth, frequency of sweetened beverage intake, whether there was toothache experience, and whether the child had seen a dentist. The regression was implemented using the “glmnet” package in R software (version 4.3.1).

[0065] Table 2 Variable screening for the prediction model of primary tooth caries (LASSO regression)

[0066]

[0067]

[0068] 5. Model construction and verification

[0069] Univariate and multivariate logistic regression analysis was used to determine the predictive factors of primary tooth caries, and the odds ratio (OR) and 95% confidence interval (95% CI) were calculated to estimate the degree of association between risk factors and primary tooth caries prevalence. Combined with clinical experience, factors selected by LASSO regression (s1 coefficient was not 0) and univariate logistic analysis (P<0.05) were included in multivariate logistic regression to establish a prediction model.

[0070] The test results were considered statistically significant only when P < 0.05.

[0071] As shown in Table 3 , after controlling for confounding factors, older age, living in rural areas, exclusive / mainly breastfeeding within 6 months after birth, toothache experience, and visiting a dentist were risk factors for caries in deciduous teeth of preschool children ( P < 0.05).

[0072] Then a logistic regression equation was constructed based on the above five predictors, each of which had a predicted reference value based on the risk ratio of the statistical results, which showed the weight of each parameter, and the total score would distinguish between healthy individuals and children with caries. The regression model formula is:

[0073] Among them, Y represents the output value, P represents the probability of dental caries, a represents the constant term, β represents the regression coefficient, and X is the predictor.

[0074] Table 3 Univariate and multivariate logistic regression analysis of risk factors for caries in primary teeth

[0075]

[0076]

[0077] As shown in Table 4 , predictive factors were included in the prediction model, and data from 645 preschoolers were collected as an external validation set. The dental caries diagnosis criteria and inclusion and exclusion criteria were the same as those of the model development set and internal validation set.

[0078] Table 4 Baseline characteristics of all patients in the external validation set

[0079]

[0080] The receiver operating characteristic (ROC) curve was drawn using the "proc" package in R software and the area under the ROC curve (AUC) was calculated to evaluate the discrimination of the model, see 3A, 4A, and 5A; the calibration curve was drawn using the "rms" and "ggplot2" packages in R software, and the calibration of the model was evaluated by the calibration curve of the predicted probability and the actual probability, see 3B, 4B, and 5B; the clinical decision analysis curve and clinical impact curve were drawn using the "rms", "rmda", and "ggplot2" packages in R software to evaluate the net clinical benefit of the model. Figure 3 C, 4C, and 5C show that the blue curve in the clinical decision analysis curve represents the prediction performance of the model, the gray curve represents the assumption that all children have caries, and the black curve represents the assumption that all children do not have caries. Within a reasonable threshold probability, the prediction model can obtain a higher benefit; Figure 3As shown in D, 4D, and 5D, the red curve in the clinical impact curve represents the number of people classified as positive by the prediction model at each threshold probability, and the blue curve represents the number of true positives at each threshold probability. When the threshold probability is within a reasonable range, the prediction accuracy determined by the prediction model is highly matched with the actual population suffering from caries, confirming that the prediction model exhibits excellent clinical efficacy.

[0081] As shown in Table 5, the AUC (95% CI) of the model predictors were: age: 0.580 (0.545-0.615), place of residence: 0.619 (0.588-0.651), feeding method within 6 months after birth: 0.567 (0.536-0.599), whether there was toothache experience: 0.662 (0.633-0.691), whether to see a dentist: 0.549 (0.528-0.570), P < 0.001.

[0082] As shown in Table 5 and Figure 3 As shown in Figures 4 and 5, the AUC (95% CI) of the prediction model in the model development set, internal validation set, and external validation set were 0.746 (0.714-0.777), 0.717 (0.666-0.768), and 0.804 (0.765-0.844), respectively. The calibration curve showed that the model had good prediction accuracy. The clinical decision analysis and clinical impact curve showed that the prediction model was in line with clinical application within a reasonable threshold range.

[0083] Table 5 Predictors and prediction parameters of different cohorts

[0084]

[0085] 6. Model Subgroup Analysis

[0086] After the model was established and validated, further subgroup analysis was performed based on the urban-rural classification of children’s residence to evaluate the validity of the prediction model in urban and rural areas.

[0087] As shown in Table 6 and Figure 6 As shown, the AUC (95% CI) of the prediction model in urban and rural areas were 0.729 (0.648-0.809) and 0.728 (0.669-0.787), respectively. The calibration curve showed that the model had good prediction accuracy, proving the effectiveness of the prediction model in urban and rural areas.

[0088] Table 6 Subgroup analysis

[0089]

[0090] 7. Model Visualization

[0091] The nomogram of the prediction model was drawn using the "regploy" package of R software. The prediction line was drawn upward in the nomogram to obtain the score corresponding to each prediction factor. The scores of each prediction factor were added together to obtain the total score. The prediction line was drawn downward to obtain the risk of caries. Figure 7 As shown in the figure, for a 3-year-old preschool child who lives in an urban area, is exclusively breastfed or mainly breastfed within 6 months of birth, has no history of toothache or dental visits, the total score of all predictors is 1.32, and the corresponding probability of caries is 0.486.

[0092] To ensure the public use of the prediction model, we created a web calculator for the Nomogram using HTML+CSS+JavaScript technology. The risk level is as follows: low: 0%-33%, medium: 34%-66%, high: 67%-100%. Figure 8 As shown, a 3-year-old preschool child who lives in an urban area, is exclusively breastfed or mainly breastfed within 6 months of birth, has no history of toothache or dental visits, has a total score of -0.0575, and the corresponding probability of caries is 48.56%, which is classified as medium risk.

[0093] In summary, the present invention studies the baseline data of preschool children, and screens out a marker combination consisting of age, place of residence, feeding method within 6 months after birth, whether the child has experienced toothache, and whether the child has seen a dentist through a machine learning model; the marker combination obtained by screening is used to construct a prediction model for caries in preschool children, and the AUC (95% CI) of the prediction model in the model development set, internal validation set, and external validation set are 0.746 (0.714-0.777), 0.717 (0.666-0.768), and 0.804 (0.765-0.844), respectively. The calibration curve shows that the model prediction accuracy is good, and the clinical decision analysis and clinical impact curve show that the prediction model meets the clinical application situation within a reasonable threshold range. The prediction model described in the present invention has easy-to-understand model prediction factors and low professional and technical barriers, which provides an effective tool for the management of caries of preschool children's deciduous teeth in the community.

Claims

1. A marker for diagnosing and / or predicting dental caries in preschool children, characterized in that: The markers include one or more combinations of age, place of residence, feeding method within 6 months after the child is born, whether the child has experienced toothache, and whether the child has visited a dentist.

2. The marker for diagnosing and / or predicting dental caries in preschool children according to claim 1, characterized in that: The markers included age, place of residence, feeding method of the child in the first 6 months of life, whether the child had experienced toothache and whether the child had visited a dentist.

3. The marker for diagnosing and / or predicting dental caries in preschool children according to claim 2, characterized in that: The markers consist of age, place of residence, feeding method of the child within 6 months of birth, whether the child has experienced toothache and whether the child has visited a dentist.

4. Use of the marker for diagnosing and / or predicting dental caries in preschool children as described in any one of claims 1 to 3 in the preparation of a reagent product, database or prediction model for diagnosing and / or predicting dental caries in preschool children.

5. A prediction model for diagnosing and / or predicting dental caries in preschool children, characterized in that: The model for predicting dental caries in preschool children is constructed using the markers described in any one of claims 1-3.

6. The method for constructing a prediction model for diagnosing and / or predicting dental caries in preschool children according to claim 5, characterized in that: The steps include: (1) Obtaining the data set: Collecting oral epidemiological data of preschool children and dividing the obtained data set into model development set, internal validation set, and external validation set; (2) Variable screening: The baseline data obtained in step (1) were divided into a dental caries group and a healthy group. The LASSO regression algorithm was used to eliminate collinearity in the baseline data in the model development set and to screen for risk factors (the s1 coefficient was not 0); (3) Model construction: The risk factors obtained in step (2) were included in univariate and multivariate logistic regression analysis to determine the model predictors, and risk factors with P < 0.05 were selected for modeling; (4) Model validation: The prediction model obtained in step (3) was used to test the model’s discrimination (ROC curve), calibration (calibration curve), and clinical net benefit (DCA curve, CIC curve) in the model development set, internal validation set, and external validation set; (5) Construction of prediction model nomogram: Diagnose and / or predict dental caries in preschool children through the predictive factors described in step (3), each predictive factor has a prediction reference value based on the risk ratio of the statistical results, the prediction reference value shows the weight of each parameter, and the total score will distinguish between healthy individuals and children with dental caries.

7. The construction method according to claim 6, characterized in that: The regression model formula described in step (3) is: Among them, Y represents the output value, P represents the probability of dental caries, a represents the constant term, β represents the regression coefficient, and X is the predictor.

8. A system for diagnosing and / or predicting dental caries in preschool children, characterized in that: The method for constructing the same as claimed in claim 6 comprises: (1) a data collection module, at least used for data collection and obtaining a sample data set; (2) a data processing module, at least used to extract valid samples from the sample data set that can be used to build an evaluation model; (3) a model building module, at least for randomly dividing the complete data set of the valid samples into a training set and a validation set, fitting the training set using LASSO regression and Logistics regression methods, and recording the optimal model parameters; (4) a probability calculation module, at least for calculating the probability of dental caries according to the prediction model formula; (5) A result display module, at least used to display the prediction results.

9. A computer program product stored on a computer-readable medium, comprising a computer-readable program, which, when executed on an electronic device, provides a user input interface to apply the system for treating dental caries in preschool children as claimed in claim 8.

10. A computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to apply the system for treating dental caries in preschool children as claimed in claim 8.

Images