Systematically assessing the quality of dental electronic health record data for an investigation into oral health care disparities.

OBJECTIVES: This work describes the process by which the quality of electronic health care data for a public health study was determined. The objectives were to adapt, develop, and implement data quality assessments (DQAs) based on the National Institutes of Health Pragmatic Trials Collaboratory (NIHPTC) data quality framework within the three domains of completeness, accuracy, and consistency, for an investigation into oral health care disparities of a preventive care program. METHODS: Electronic health record data for eligible children in a dental accountable care organization of 30 offices, in Oregon, were extracted iteratively from January 1, 2014, through March 31, 2022. Baseline eligibility criteria included: children ages 0-18 with a baseline examination, Oregon home address, and either Medicaid or commercial dental benefits at least once between 2014 and 2108. Using the NIHPTC framework as a guide, DQAs were conducted throughout data element identification, extraction, staging, profiling, review, and documentation. RESULTS: The data set included 91,487 subjects, 11 data tables comprising 75 data variables (columns), with a total of 6,861,525 data elements. Data completeness was 97.2%, the accuracy of EHR data elements in extracts was 100%, and consistency between offices was strong; 29 of 30 offices within 2 standard deviations of the mean (s = 94%). CONCLUSIONS: The NIHPTC framework proved to be a useful approach, to identify, document, and characterize the dataset. The concepts of completeness, accuracy, and consistency were adapted by the multidisciplinary research team and the overall quality of the data are demonstrated to be of high quality.

Evaluating and improving the usability of a mHealth platform to assess postoperative dental pain.

Objectives: The use of interactive mobile health (mHealth) applications to monitor patient-reported postoperative pain outcomes is an emerging area in dentistry that requires further exploration. This study aimed to evaluate and improve the usability of an existing mHealth application. Materials and methods: The usability of the application was assessed iteratively using a 3-phase approach, including a rapid cognitive walkthrough (Phase I), lab-based usability testing (Phase II), and in situ pilot testing (Phase III). The study team conducted Phase I, while providers and patients participated in Phase II and III. Results: The rapid cognitive walkthrough identified 23 potential issues that could negatively impact user experience, with the majority classified as system issues. The lab-based usability testing yielded 141 usability issues.; 43% encountered by patients and 57% by dentists. Usability problems encountered during pilot testing included undelivered messages due to mobile phone carrier and service-related issues, errors in patients' phone number data entry, and problems in provider training. Discussion: Through collaborative and iterative work with the vendor, usability issues were addressed before launching a trial to assess its efficacy. Conclusion: The usability of the mHealth application for postoperative dental pain was remarkably improved by the iterative analysis and interdisciplinary collaboration.

Assessing the performance of Bing Chat artificial intelligence: Dental exams, clinical guidelines, and patients' frequent questions.

OBJECTIVES: Bing Chat is a large language model artificial intelligence (AI) with online search and text generating capabilities. This study assessed its performance within the scope of dentistry in: (a) tackling exam questions for dental students, (ii) providing guidelines for dental practitioners, and (iii) answering patients' frequently asked questions. We discuss the potential of clinical tutoring, common patient communication and impact on academia. METHODS: With the aim of assessing AI's performance in dental exams, Bing Chat was presented with 532 multiple-choice questions and awarded scores based on its answers. In evaluating guidelines for clinicians, a further set of 15 questions, each with 2 follow-up questions on clinical protocols, was presented to the AI. The answers were assessed by 4 reviewers using electronic visual analog scale. In evaluating answers to patients' frequently asked questions, another list of 15 common questions was included in the session, with respective outputs assessed. RESULTS: Bing Chat correctly answered 383 out of 532 multiple-choice questions in dental exam part, achieving a score of 71.99 %. As for outlining clinical protocols for practitioners, the overall assessment score was 81.05 %. In answering patients' frequently asked questions, Bing Chat achieved an overall mean score of 83.8 %. The assessments demonstrated low inter-rater reliability. CONCLUSIONS: The overall performance of Bing Chat was above the regularly adopted passing scores, particularly in answering patient's frequently asked questions. The generated content may have biased sources. These results suggest the importance of raising clinicians' awareness of AI's benefits and risks, as well as timely adaptations of dental education curricula, and safeguarding its use in dentistry and healthcare in general. CLINICAL SIGNIFICANCE: Bing Chat AI performed above the passing threshold in three categories, and thus demonstrated potential for educational assistance, clinical tutoring, and answering patients' questions. We recommend popularizing its benefits and risks among students and clinicians, while maintaining awareness of possible false information.

Longevity of dental restorations in Sjogren's disease patients using electronic dental and health record data.

BACKGROUND: Decreased salivary secretion is not only a risk factor for carious lesions in Sjögren's disease (SD) but also an indicator of deterioration of teeth with every restorative replacement. This study determined the longevity of direct dental restorations placed in patients with SD using matched electronic dental record (EDR) and electronic health record (EHR) data. METHODS: We conducted a retrospective cohort study using EDR and EHR data of Indiana University School of Dentistry patients who have a SD diagnosis in their EHR. Treatment history of patients during 15 years with SD (cases) and their matched controls with at least one direct dental restoration were retrieved from the EDR. Descriptive statistics summarized the study population characteristics. Cox regression models with random effects analyzed differences between cases and controls for time to direct restoration failure. Further the model explored the effect of covariates such as age, sex, race, dental insurance, medical insurance, medical diagnosis, medication use, preventive dental visits per year, and the number of tooth surfaces on time to restoration failure. RESULTS: At least one completed direct restoration was present for 102 cases and 42 controls resulting in a cohort of 144 patients' EDR and EHR data. The cases were distributed as 21 positives, 57 negatives, and 24 uncertain cases based on clinical findings. The average age was 56, about 93% were females, 54% were White, 74% had no dental insurance, 61% had public medical insurance, < 1 preventive dental visit per year, 94% used medications and 93% had a medical diagnosis that potentially causes dry mouth within the overall study cohort. About 529 direct dental restorations were present in cases with SD and 140 restorations in corresponding controls. Hazard ratios of 2.99 (1.48-6.03; p = 0.002) and 3.30 (1.49-7.31, p-value: 0.003) showed significantly decreased time to restoration failure among cases and positive for SD cases compared to controls, respectively. Except for the number of tooth surfaces, no other covariates had a significant influence on the survival time. CONCLUSION: Considering the rapid failure of dental restorations, appropriate post-treatment assessment, management, and evaluation should be implemented while planning restorative dental procedures among cases with SD. Since survival time is decreased with an increase in the number of surfaces, guidelines for restorative procedures should be formulated specifically for patients with SD.

Quantifying error introduced by iterative closest point image registration.

OBJECTIVES: The aim of this paper was to quantify the analysis error introduced by iterative closest point (ICP) image registration. We also investigated whether a subsequent subtraction process can reduce process error. METHODS: We tested metrology and two 3D inspection software using calibration standards at 0.39 µm, and 2.64 µm and mathematically perfect defects (softgauges) at 2 and 20 µm, on free form surfaces of increasing complexity and area, both with and without registration. Errors were calculated in percentage relative to the size of the defect being measured. Data were analysed in GraphPad Prism 9, normal and two-way ANOVA with post-hoc Tukey's was applied. Significance was inferred at p < 0.05. RESULTS: Using ICP registration introduced errors from 0 % to 15.63 % of the defect size depending on the surface complexity and size of the defect. Significant differences were observed in analysis measurements between metrology and 3D inspection software and within different 3D inspection software, however, one did not show clear superiority over another. Even in the absence of registration, defects at 0.39 µm, and 2.64 µm produced substantial measurement error (13.39-77.50 % of defect size) when using 3D inspection software. Adding an additional data subtraction process reduced registration error to negligible levels (<1 % independent of surface complexity or area). CONCLUSIONS: Commercial 3D inspection software introduces error during direct measurements below 3 µm. When using an ICP registration, errors over 15 % of the defect size can be introduced regardless of the accuracy of adjacent registration surfaces. Analysis output between software are not consistently repeatable or comparable and do not utilise ISO standards. Subtracting the datasets and analysing the residual difference reduced error to negligible levels. CLINICAL SIGNIFICANCE: This paper quantifies the significant errors and inconsistencies introduced during the registration process even when 3D datasets are true and precise. This may impact on research diagnostics and clinical performance. An additional data processing step of scan subtraction can reduce this error but increases computational complexity.

Automatic dental age calculation from panoramic radiographs using deep learning: a two-stage approach with object detection and image classification.

BACKGROUND: Dental age is crucial for treatment planning in pediatric and orthodontic dentistry. Dental age calculation methods can be categorized into morphological, biochemical, and radiological methods. Radiological methods are commonly used because they are non-invasive and reproducible. When radiographs are available, dental age can be calculated by evaluating the developmental stage of permanent teeth and converting it into an estimated age using a table, or by measuring the length between some landmarks such as the tooth, root, or pulp, and substituting them into regression formulas. However, these methods heavily depend on manual time-consuming processes. In this study, we proposed a novel and completely automatic dental age calculation method using panoramic radiographs and deep learning techniques. METHODS: Overall, 8,023 panoramic radiographs were used as training data for Scaled-YOLOv4 to detect dental germs and mean average precision were evaluated. In total, 18,485 single-root and 16,313 multi-root dental germ images were used as training data for EfficientNetV2 M to classify the developmental stages of detected dental germs and Top-3 accuracy was evaluated since the adjacent stages of the dental germ looks similar and the many variations of the morphological structure can be observed between developmental stages. Scaled-YOLOv4 and EfficientNetV2 M were trained using cross-validation. We evaluated a single selection, a weighted average, and an expected value to convert the probability of developmental stage classification to dental age. One hundred and fifty-seven panoramic radiographs were used to compare automatic and manual human experts' dental age calculations. RESULTS: Dental germ detection was achieved with a mean average precision of 98.26% and dental germ classifiers for single and multi-root were achieved with a Top-3 accuracy of 98.46% and 98.36%, respectively. The mean absolute errors between the automatic and manual dental age calculations using single selection, weighted average, and expected value were 0.274, 0.261, and 0.396, respectively. The weighted average was better than the other methods and was accurate by less than one developmental stage error. CONCLUSION: Our study demonstrates the feasibility of automatic dental age calculation using panoramic radiographs and a two-stage deep learning approach with a clinically acceptable level of accuracy.

Survival analysis of posterior composite restorations in National Dental PBRN general dentistry practices.

OBJECTIVE: Quantify the survival of posterior composite restorations (PCR) placed during the study period in permanent teeth in United States (US) general dental community practices and factors predictive of that survival. METHODS: A retrospective cohort study was conducted utilizing de-identified electronic dental record (EDR) data of patients who received a PCR in 99 general dentistry practices in the National Dental Practice-Based Research Network (Network). The final analyzed data set included 700,885 PCRs from 200,988 patients. Descriptive statistics and Kaplan Meier (product limit) estimator were performed to estimate the survival rate (defined as the PCR not receiving any subsequent treatment) after the first PCR was observed in the EDR during the study time. The Cox proportional hazards model was done to account for patient- and tooth-specific covariates. RESULTS: The overall median survival time was 13.3 years. The annual failure rates were 4.5-5.8 % for years 1-5; 5.3-5.7 %, 4.9-5.5 %, and 3.3-5.2 % for years 6-10, 11-15, and 16-20, respectively. The failure descriptions recorded for < 7 % failures were mostly caries (54 %) and broken or fractured tooth/restorations (23 %). The following variables significantly predicted PCR survival: number of surfaces that comprised the PCR; having at least one interproximal surface; tooth type; type of prior treatment received on the tooth; Network region; patient age and sex. Based on the magnitude of the multivariable estimates, no single factor predominated. CONCLUSIONS: This study of Network practices geographically distributed across the US observed PCR survival rates and predictive factors comparable to studies done in academic settings and outside the US. CLINICAL SIGNIFICANCE: Specific baseline factors significantly predict the survival of PCRs done in US community dental practices.

Evaluating the precision and accuracy of digital dental models with a low-cost structured light device

ABSTRACT Objective: This study evaluated the accuracy and precision of digital models acquisition using a home-built, low-cost scanning system based on the structured light method. Methods: a plaster model (PM) was scanned using the experimental device (SL) and a dental desktop scanner (DS). The teeth dimensions of PM and SL models were measured in triplicate, with a caliper and digitally, respectively. The agreement of the measurements of each model was evaluated using the intraclass correlation coefficient, and the validity between the different measurement techniques was assessed using the Bland-Altman analysis. The accuracy and precision of the models were qualitatively investigated using the mesh superposition of the SL and DS models. Results: A high intraclass correlation coefficient was observed in all models (PM=0.964; SL1=0.998; SL2=0.995; SL3=0.998), and there was no statistical difference between the measurements of the SL models (p>0.05). PM and SL model measurements were found to be in good agreement, with only 3.57% of the observed differences between the same measurement being located outside 95% limits of agreement according to Bland and Altman (0.43 and -0.40 mm). In the superimpositions of SL-SL and SL-DS models, areas of discrepancy greater than 0.5 mm were observed mainly in interproximal, occlusal, and cervical sites. Conclusion: These results indicate that the home-built SL scanning system did not possess sufficient accuracy and precision for many clinical applications. However, the consistency in preserving the dental proportions suggests that the equipment can be used for planning, storage, and simple clinical purposes.

RESUMO Objetivo: Este estudo avaliou a acurácia e a precisão da aquisição de modelos digitais utilizando um sistema de digitalização caseiro e de baixo custo baseado no método de luz estruturada. Material e Métodos: Para isso, um modelo de gesso (MG) foi digitalizado utilizando o dispositivo experimental (LE) e um scanner dental de mesa (SD). As dimensões dos dentes dos modelos MG e LE foram medidas em triplicata com um paquímetro e digitalmente, respectivamente. A concordância das medidas de cada modelo foi avaliada usando o coeficiente de correlação intraclasse, e a validade entre as diferentes técnicas de medição foi avaliada usando a análise de Bland-Altman. A acurácia e a precisão dos modelos foram investigadas qualitativamente usando a sobreposição de malhas dos modelos LE e SD. Resultados: Um alto coeficiente de correlação intraclasse foi observado em todos os modelos (MG=0,964; LE1=0,998; LE2=0,995; LE3=0,998) e não houve diferença estatística entre as medições dos modelos LE (p>0,05). As medições dos modelos MG e LE mostraram boa concordância, com apenas 3,57% das diferenças observadas entre as mesmas medições localizadas fora dos limites de concordância de 95% de acordo com a análise de Bland-Altman (0,43 e -0,40 mm). Nas sobreposições dos modelos LE-LE e LE-SD, foram observadas áreas de discrepância maiores que 0,5 mm principalmente nos sítios interproximais, oclusais e cervicais. Conclusões: Esses resultados indicam que o sistema de digitalização por luz estruturada caseiro não possui precisão e acurácia suficientes para muitas aplicações clínicas. No entanto, a consistência na preservação das proporções dentárias sugere que o equipamento pode ser usado para planejamento, armazenamento e propósitos clínicos simples.

La educación a distancia en la Enseñanza Técnica Superior de Prótesis Estomatológica: experiencias de 2 cursos

Introducción: La asignatura Introducción a la Metodología de la Investigación, para la Enseñanza Técnica Superior de Prótesis Estomatológica, de la Facultad de Estomatología de La Habana, se incorporó a la educación a distancia, durante el curso 2021-2022. Objetivo: Describir la experiencia en la ejecución de la asignatura con el uso del aula virtual de la Institución. Material y Métodos: Se realizó una investigación de tipo descriptiva, retrospectiva, acerca de las experiencias en el proceso de enseñanza-aprendizaje, durante los cursos 2021-2022 y 2022-2023. Resultados: Se cumplieron los objetivos metodológicos de la asignatura, se debatieron los temas propuestos en los foros, se aclararon las dudas utilizando las herramientas interactivas de la plataforma y la atención al alumno fue personalizada. Conclusiones: Se logró impartir por primera vez con éxito, la asignatura, se digitalizaron los objetos de aprendizaje por temas, se diseñaron actividades para la evaluación del aprendizaje, y se obtuvieron calificaciones satisfactorias.

Introduction: The subject Introduction to Research Methodology, for the Higher Technical Education of Stomatological Prosthetics, of the Faculty of Stomatology of Havana, was incorporated into distance education, during the 2021-2022 academic year. Objective: Describe the experience in the execution of the subject with the use of the Institution's virtual classroom. Material and Methods: A descriptive, retrospective research was carried out on the experiences in the teaching-learning process, during the 2021-2022 and 2022-2023 academic years. Results: The methodological objectives of the subject were met, the topics proposed in the forums were debated, doubts were clarified using the platform's interactive tools, and student attention was personalized. Conclusions: The subject was successfully taught for the first time, the learning objects were digitized by topic, activities were designed for the evaluation of learning, and satisfactory grades were obtained.

Periodontitis and Metabolic Syndrome: Statistical and Machine Learning Analytics of a Nationwide Study.

This study aimed to analyze the associations between periodontitis and metabolic syndrome (MetS) components and related conditions while controlling for sociodemographics, health behaviors, and caries levels among young and middle-aged adults. We analyzed data from the Dental, Oral, and Medical Epidemiological (DOME) record-based cross-sectional study that combines comprehensive sociodemographic, medical, and dental databases of a nationally representative sample of military personnel. The research consisted of 57,496 records of patients, and the prevalence of periodontitis was 9.79% (5630/57,496). The following parameters retained a significant positive association with subsequent periodontitis multivariate analysis (from the highest to the lowest OR (odds ratio)): brushing teeth (OR = 2.985 (2.739-3.257)), obstructive sleep apnea (OSA) (OR = 2.188 (1.545-3.105)), cariogenic diet consumption (OR = 1.652 (1.536-1.776)), non-alcoholic fatty liver disease (NAFLD) (OR = 1.483 (1.171-1.879)), smoking (OR = 1.176 (1.047-1.322)), and age (OR = 1.040 (1.035-1.046)). The following parameters retained a significant negative association (protective effect) with periodontitis in the multivariate analysis (from the highest to the lowest OR): the mean number of decayed teeth (OR = 0.980 (0.970-0.991)); North America as the birth country compared to native Israelis (OR = 0.775 (0.608-0.988)); urban non-Jewish (OR = 0.442 (0.280-0.698)); and urban Jewish (OR = 0.395 (0.251-0.620)) compared to the rural locality of residence. Feature importance analysis using the eXtreme Gradient Boosting (XGBoost) machine learning algorithm with periodontitis as the target variable ranked obesity, OSA, and NAFLD as the most important systemic conditions in the model. We identified a profile of the "patient vulnerable to periodontitis" characterized by older age, rural residency, smoking, brushing teeth, cariogenic diet, comorbidities of obesity, OSA and NAFLD, and fewer untreated decayed teeth. North American-born individuals had a lower prevalence of periodontitis than native Israelis. This study emphasizes the holistic view of the MetS cluster and explores less-investigated MetS-related conditions in the context of periodontitis. A comprehensive assessment of disease risk factors is crucial to target high-risk populations for periodontitis and MetS.

A Classroom Practice and Assessment Resource for Dental Clinical Treatment Planning.

Introduction: Dental students' opportunities to practice clinical treatment planning prior to the clinical experience are often limited. This resource permits students to practice treatment planning in the classroom and allows the instructor to efficiently provide feedback to a large number of students. Methods: I developed a computer-based treatment plan worksheet (Microsoft Excel, Google Sheets, or Qualtrics) consisting of four sequence steps and a list of Common Dental Terminology (CDT) codes. I implemented this educational activity in the Case Management and Treatment Planning course (second year, spring quarter). The course included 10 weekly sessions of 1 hour each. I gave students a practice case at the beginning of the course. They submitted comprehensive sequenced treatment plans as assignments using the Qualtrics treatment plan worksheet 2 days after the interdisciplinary treatment planning didactic session. In the subsequent debrief session, I discussed the case with the entire class and focused on themes observed from the students' performance on the assignments. Results: I graded all students' submissions (n = 87) using CDT codes both with and without consideration of sequence steps and observed variations in students' performance. Many students (n = 36, 41%) opted to give feedback through an anonymous survey, with 88%-96% of them agreeing or strongly agreeing that the educational activity allowed them to practice, demonstrate their knowledge, and improve their skills in treatment planning. Discussion: This computer-based worksheet lets students practice clinical treatment planning in the classroom and allows the instructor to effectively assess a large number of students' assignments.

Digital surveillance: The interest in mouthwash-related information.

BACKGROUND: The comprehension of the interests of Internet users regarding their health-related searches may reveal the community's demands about oral health. The study aimed to characterize the interests of Google users related to mouthwash in Australia, Brazil, Chile, Japan, Mexico, Russia, the United Kingdom, the United States, Saudi Arabia and South Africa applying the Google Trends. METHODS: This longitudinal retrospective study analysed the mouthwash-related interest of Google users from January 2004 to December 2020. The monthly variation of relative search volume (RSV) and the main queries related were determined using Google Trends. Autoregressive integrated moving average (ARIMA) forecasting models were built to establish the predictive RSV values for mouthwash for additional 12 months. Auto-correlation plots and a general additive model (GAM) were used to diagnose trends and seasonality in RSV curves. In addition, the influence of social isolation related to the outbreak of COVID-19 were analysed. RESULTS: The RSVs curves showed a considerable increase in searches related to mouthwash to AUS, BRA, JAP, MEX, GBR and USA (RSV > 25), while the growth was slight to CHI, KSA, RSA and RUS (RSV < 25) over the years, without influence of monthly seasonality. All countries showed a significant increase in mouthwash interest after the outbreak of COVID-19, except for KSA and RUS. The mouthwash-related searches were associated to specific brands or chemical compositions, treatments, whitening agents, homemade mouthwash and indications for the 'best mouthwash'. CONCLUSIONS: In general, there was an increasing interest of Google users in mouthwash-related topics between 2004 and 2020. In addition, in most countries, there was an expansion in searches during the social isolation of the COVID-19 pandemics.

Influence of Practice Characteristics on the Adoption of Electronic Dental Records in Jeddah, Saudi Arabia.

OBJECTIVES: The adoption of electronic dental records (EDRs) is less extensively studied than electronic medical records (EMRs) in Saudi Arabia. Therefore, a multivariate analysis was conducted to calculate the adoption of EDRs and determine the practice characteristics that influence adoption. METHODS: An online survey was conducted with 220 dental practices in Jeddah from August to December 2021. The questionnaire contained 10 items that measured the adoption of EDRs and identified the region, district, practice characteristics, and practice size. A regression analysis was used to ascertain the relationships between EDR adoption and the predictor variables. RESULTS: About 93% of the dental practices, we surveyed in Jeddah had adopted EDRs. Public dental practices and large practices were associated with higher rates of adoption (respectively, 97.0%, p = 0.016; 97.8%, p = 0.009). The logistic regression model showed statistically significant results regarding practice characteristics, practice size, and the acceptance of insurance patients. EDR adoption was 89% less likely for private dental practices, 99% less likely for smaller dental practices (≥2 dentists), and 98% less likely in dental practices that did not treat patients with insurance. CONCLUSIONS: Our study sample showed a high rate of EDR adoption. Among the participants, public practices, large practices, and practices that treat patients with insurance were the most positively inclined toward EDR adoption.

Dentists' perceptions and usability testing of the implant disease risk assessment IDRA, a tool for preventing peri­implant disease: A qualitative study.

INTRODUCTION: we aimed to explore dentists' perceptions toward the implementation of a dental informatics risk assessment tool which estimates the risk for a patient to develop peri­implantitis. MATERIALS AND METHODS: the Implant Disease Risk Assessment Tool (IDRA) was presented to a convenience sample of seven dentists working in a university clinic, whom were asked to use IDRA with the information of three clinical cases whilst thinking aloud and then fill the System Usability Scale (SUS). A semi-structured interview technique was used with audio record to allow free expression of participants' perceptions related to the IDRA. The interviews information was categorized and analyzed by the authors. RESULTS: to our knowledge, this is the first study conducted to develop a qualitative usability test of IDRA, evaluating the effectiveness, efficiency, and users' satisfaction. There were more variations in responses the greater the degree of complexity of the clinical case. Generally, the participants classified the tool as good, getting usability values of 77,2 (SD 19,8) and learnability 73,2 (SD 24,5). CONCLUSION: four additional factors should be considered to improve IDRA tool: 1) considering the relation between contour angle and peri-implant tissue height; 2) automatic periodontal classification in the IDRA tool after completing the periodontogram in the clinical software; 3) presentation of a flowchart to assist therapeutic decisions alongside the final score defined by the IDRA tool; 4) integrating of precision tests such as Implantsafe® DR… (dentognostics gmbh, Jena) and Oralyzer®(dentognostics gmbh, Jena). CLINICAL SIGNIFICANCE: etiology and pathogenesis of peri­implant diseases is multifactorial. These tools must follow a natural integration to be easily applied in a clinical setting. It is important to study their usability from the clinicians' point of view, evaluating the effectiveness, efficiency, and users' satisfaction.

Designing audit and feedback dashboards for dentists to monitor their opioid prescribing.

BACKGROUND AND OBJECTIVE: Prescription drug abuse is a major factor leading to drug overdose deaths in the US and dentists are one of the leading prescribers of opioid pain medication. Knowing that Audit & Feedback (A&F) dashboards are an effective tool and are used as quality improvement interventions, we aimed to develop such dashboards personalized for dental providers which could allow them to monitor their own opioid prescribing performance. METHODS: In this paper we report on the process for designing the A&F dashboards for dentists which were developed by using an iterative human-centered design process. The results obtained from each iteration were used to enrich the information needs analyses, provide function testing, and guide the design decisions of the next iteration. RESULTS: Engaging dentists in the development and refinement of the dashboards while using the think-aloud protocol for user-testing, provided rapid feedback and identified areas that were confusing and needed either a redesign or additional explanatory content. The final version of dashboards consisted of displaying necessary information through easy to interpret visualizations and interactive features. These included providing access to current national and organizational prescribing guidelines, displaying changes in individual prescribing behavior over time, comparing individual prescribing rate to peer group rate and target rate, displaying procedure specific prescribing, integrating patient reported post-operative dental pain experience and providing navigation and interpretation tips for users. The dashboards were easy to learn and understand for the dentists and were deemed as worth using often in dental practice. CONCLUSION: Our research was able to demonstrate the creation of useful and usable A&F dashboards using data from electronic dental records and patient surveys, for dentists to effectively monitor their opioid prescribing behavior. Efficacy of the dashboards will be tested in future work.

Conformer-Based Dental AI Patient Clinical Diagnosis Simulation Using Korean Synthetic Data Generator for Multiple Standardized Patient Scenarios.

The goal of clinical practice education is to develop the ability to apply theoretical knowledge in a clinical setting and to foster growth as a professional healthcare provider. One effective method of achieving this is through the utilization of Standardized Patients (SP) in education, which familiarizes students with real patient interviews and allows educators to assess their clinical performance skills. However, SP education faces challenges such as the cost of hiring actors and the shortage of professional educators to train them. In this paper, we aim to alleviate these issues by utilizing deep learning models to replace the actors. We employ the Conformer model for the implementation of the AI patient, and we develop a Korean SP scenario data generator to collect data for training responses to diagnostic questions. Our Korean SP scenario data generator is devised to generate SP scenarios based on the provided patient information, using pre-prepared questions and answers. In the AI patient training process, two types of data are employed: common data and personalized data. The common data are employed to develop natural general conversation skills, while personalized data, from the SP scenario, are utilized to learn specific clinical information relevant to a patient's role. Based on these data, to evaluate the learning efficiency of the Conformer structure, a comparison was conducted with the Transformer using the BLEU score and WER as evaluation metrics. Experimental results showed that the Conformer-based model demonstrated a 3.92% and 6.74% improvement in BLEU and WER performance compared to the Transformer-based model, respectively. The dental AI patient for SP simulation presented in this paper has the potential to be applied to other medical and nursing fields, provided that additional data collection processes are conducted.

Accurate gingival segmentation from 3D images with artificial intelligence: an animal pilot study.

BACKGROUND: Gingival phenotype plays an important role in dental diagnosis and treatment planning. Traditionally, determining the gingival phenotype is done by manual probing of the gingival soft tissues, an invasive and time-consuming procedure. This study aims to evaluate the feasibility and accuracy of an alternatively novel, non-invasive technology based on the precise 3-dimension (3D) soft tissue reconstruction from intraoral scanning and cone beam computed tomography (CBCT) to predict the gingival biotype. METHODS: As a proof-of-concept, Yorkshire pig mandibles were scanned, and the CBCT data were fed into a deep-learning model to reconstruct the teeth and surrounding bone structure in 3D. By overlaying the CBCT scan with the intraoral scans, an accurate superposition was created and used for virtual measurements of the soft tissue thickness. Meanwhile, gingival thicknesses were also measured by a periodontal probe and digital caliper on the buccal and lingual sides at 3 mm apical to the gingival margin of the posterior teeth and compared with the virtual assessment at the same location. The data obtained from virtual and clinical measurements were compared by Wilcoxon matched-pairs signed-rank analysis, while their correlation was determined by Pearson's r value. The Mann-Whitney U test was used for intergroup comparisons of the amount of difference. RESULTS: Among 108 investigated locations, the clinical and virtual measurements are strongly positively correlated (r = 0.9656, P < 0.0001), and only clinically insignificant differences (0.066 ± 0.223 mm) were observed between the two assessments. There is no difference in the agreement between the virtual and clinical measurements on sexually matured samples (0.087 ± 0.240 mm) and pre-pubertal samples (0.033 ± 0.195 mm). Noticeably, there is a greater agreement between the virtual and clinical measurements at the buccal sites (0.019 ± 0.233 mm) than at the lingual sites (0.116 ± 0.215 mm). CONCLUSION: In summary, the artificial intelligence-based virtual measurement proposed in this work provides an innovative technique potentially for accurately measuring soft tissue thickness using clinical routine 3D imaging systems, which will aid clinicians in generating a more comprehensive diagnosis with less invasive procedures and, in turn, optimize the treatment plans with more predictable outcomes.

Plasticity of Dental Cell Types in Development, Regeneration, and Evolution.

Recent years have improved our understanding of the plasticity of cell types behind inducing, building, and maintaining different types of teeth. The latest efforts were aided by progress in single-cell transcriptomics, which helped to define not only cell states with mathematical precision but also transitions between them. This includes new aspects of dental epithelial and mesenchymal stem cell niches and beyond. These recent efforts revealed continuous and fluid trajectories connecting cell states during dental development and exposed the natural plasticity of tooth-building progenitors. Such "developmental" plasticity seems to be employed for organizing stem cell niches in adult continuously growing teeth. Furthermore, transitions between mature cell types elicited by trauma might represent a replay of embryonic continuous cell states. Alternatively, they could constitute transitions that evolved de novo, not known from the developmental paradigm. In this review, we discuss and exemplify how dental cell types exhibit plasticity during dynamic processes such as development, self-renewal, repair, and dental replacement. Hypothetically, minor plasticity of cell phenotypes and greater plasticity of transitions between cell subtypes might provide a better response to lifetime challenges, such as damage or dental loss. This plasticity might be additionally harnessed by the evolutionary process during the elaboration of dental cell subtypes in different animal lineages. In turn, the diversification of cell subtypes building teeth brings a diversity of their shape, structural properties, and functions.

Developing Automated Computer Algorithms to Track Periodontal Disease Change from Longitudinal Electronic Dental Records.

OBJECTIVE: To develop two automated computer algorithms to extract information from clinical notes, and to generate three cohorts of patients (disease improvement, disease progression, and no disease change) to track periodontal disease (PD) change over time using longitudinal electronic dental records (EDR). METHODS: We conducted a retrospective study of 28,908 patients who received a comprehensive oral evaluation between 1 January 2009, and 31 December 2014, at Indiana University School of Dentistry (IUSD) clinics. We utilized various Python libraries, such as Pandas, TensorFlow, and PyTorch, and a natural language tool kit to develop and test computer algorithms. We tested the performance through a manual review process by generating a confusion matrix. We calculated precision, recall, sensitivity, specificity, and accuracy to evaluate the performances of the algorithms. Finally, we evaluated the density of longitudinal EDR data for the following follow-up times: (1) None; (2) Up to 5 years; (3) > 5 and ≤ 10 years; and (4) >10 and ≤ 15 years. RESULTS: Thirty-four percent (n = 9954) of the study cohort had up to five years of follow-up visits, with an average of 2.78 visits with periodontal charting information. For clinician-documented diagnoses from clinical notes, 42% of patients (n = 5562) had at least two PD diagnoses to determine their disease change. In this cohort, with clinician-documented diagnoses, 72% percent of patients (n = 3919) did not have a disease status change between their first and last visits, 669 (13%) patients' disease status progressed, and 589 (11%) patients' disease improved. CONCLUSIONS: This study demonstrated the feasibility of utilizing longitudinal EDR data to track disease changes over 15 years during the observation study period. We provided detailed steps and computer algorithms to clean and preprocess the EDR data and generated three cohorts of patients. This information can now be utilized for studying clinical courses using artificial intelligence and machine learning methods.