Leveraging Large Language Models in the delivery of post-operative dental care: a comparison between an embedded GPT model and ChatGPT.

OBJECTIVE: This study underscores the transformative role of Artificial Intelligence (AI) in healthcare, particularly the promising applications of Large Language Models (LLMs) in the delivery of post-operative dental care. The aim is to evaluate the performance of an embedded GPT model and its comparison with ChatGPT-3.5 turbo. The assessment focuses on aspects like response accuracy, clarity, relevance, and up-to-date knowledge in addressing patient concerns and facilitating informed decision-making. MATERIAL AND METHODS: An embedded GPT model, employing GPT-3.5-16k, was crafted via GPT-trainer to answer postoperative questions in four dental specialties including Operative Dentistry & Endodontics, Periodontics, Oral & Maxillofacial Surgery, and Prosthodontics. The generated responses were validated by thirty-six dental experts, nine from each specialty, employing a Likert scale, providing comprehensive insights into the embedded GPT model's performance and its comparison with GPT3.5 turbo. For content validation, a quantitative Content Validity Index (CVI) was used. The CVI was calculated both at the item level (I-CVI) and scale level (S-CVI/Ave). To adjust I-CVI for chance agreement, a modified kappa statistic (K*) was computed. RESULTS: The overall content validity of responses generated via embedded GPT model and ChatGPT was 65.62% and 61.87% respectively. Moreover, the embedded GPT model revealed a superior performance surpassing ChatGPT with an accuracy of 62.5% and clarity of 72.5%. In contrast, the responses generated via ChatGPT achieved slightly lower scores, with an accuracy of 52.5% and clarity of 67.5%. However, both models performed equally well in terms of relevance and up-to-date knowledge. CONCLUSION: In conclusion, embedded GPT model showed better results as compared to ChatGPT in providing post-operative dental care emphasizing the benefits of embedding and prompt engineering, paving the way for future advancements in healthcare applications.

An Artificial Intelligence model for implant segmentation on periapical radiographs.

OBJECTIVE: To segment dental implants on PA radiographs using a Deep Learning (DL) algorithm. To compare the performance of the algorithm relative to ground truth determined by the human annotator. Methodology: Three hundred PA radiographs were retrieved from the radiographic database and consequently annotated to label implants as well as teeth on the LabelMe annotation software. The dataset was augmented to increase the number of images in the training data and a total of 1294 images were used to train, validate and test the DL algorithm. An untrained U-net was downloaded and trained on the annotated dataset to allow detection of implants using polygons on PA radiographs. RESULTS: A total of one hundred and thirty unseen images were run through the trained U-net to determine its ability to segment implants on PA radiographs. The performance metrics are as follows: accuracy of 93.8%, precision of 90%, recall of 83%, F-1 score of 86%, Intersection over Union of 86.4% and loss = 21%. CONCLUSIONS: The trained DL algorithm segmented implants on PA radiographs with high performance similar to that of the humans who labelled the images forming the ground truth.

A bibliometric analysis of scientific literature in digital dentistry from low- and lower-middle income countries.

OBJECTIVE: Bibliometric analysis and citation counts help to acknowledge influence of publications. The aim of this study was to conduct bibliometric and citation analysis of top-cited articles, from low- and lower-middle income countries, on use and application of digital technology in dentistry. METHODOLOGY: A search strategy based on "Digital Dentistry", "Low Income Countries", and "Lower-Middle Income Countries" was used in October 2023 using Scopus database to retrieve articles relevant to digital dentistry, with citation count of 10 or more. From 44 included articles, bibliometric information was analyzed on SPSS version 23. Network analysis based on co-citations, keywords, and number of citations was conducted on VOS software (version 1.6.20). RESULTS: Most relevant articles were published in 2021 (n = 8), with 52.3% original articles, out of which 40.9% were in vitro studies. India had the highest number of articles (n = 24), with most publications in The Journal of Indian Prosthodontic Society (n = 4), and in the domain of General Dentistry (n = 15, 34.1%). Co-authorship network analysis was not significant, but country-wise co-authorship analysis revealed India with the greatest link strength (4.0). Highest occurring keyword was 3D printing (link strength 5.0), and the citation analysis revealed Journal of Prosthetic Dentistry with the most number of published documents (3), having a citation count of 275. Bibliographic coupling for sources revealed Journal of Indian Prosthodontic Society to have the highest link strength of 15.33. CONCLUSION: This analysis uncovers interesting bibliometric and citation based information including key thematic trends, emphasizing crucial role of technologies like 3D printing, CAD/CAM, and CBCT in digital dentistry. The study underscores the imperative for increased original research efforts in low- and lower middle-income countries.

Navigating challenges and opportunities: AI's contribution to Pakistan's sustainable development goals agenda - a narrative review.

Sustainable Developmental Goals (SDGs) were introduced by the United Nations to ensure the sustainable progress of mankind through various domains. Pakistan, a low-middle-income country, faces many challenges in achieving SDGs. Artificial Intelligence is a rapidly evolving technology presenting significant importance in achieving SDGs. Therefore, this narrative review aimed to evaluate the artificial intelligence technologies that have been utilized globally and nationally which can be implemented in Pakistan focusing on Goal 3 (Good Health and Well-being) of SDGs. AI has been utilized primarily in high-income countries aiming to improve healthcare, thereby progressing towards achieving different targets of Goal 3 of SDGs. Pakistan lacks such initiatives with modest to no improvement across different SDGs. Therefore, Pakistan can adapt initiatives undertaken by resourceful countries to achieve its own SDGs.

Multi-model deep learning approach for segmentation of teeth and periapical lesions on pantomographs.

INTRODUCTION: The fields of medicine and dentistry are beginning to integrate artificial intelligence (AI) in diagnostics. This may reduce subjectivity and improve the accuracy of diagnoses and treatment planning. Current evidence on pathosis detection on pantomographs (PGs) indicates the presence or absence of disease in the entire radiographic image, with little evidence of the relation of periapical pathosis to the causative tooth. OBJECTIVE: To develop a deep learning (DL) AI model for the segmentation of periapical pathosis and its relation to teeth on PGs. METHOD: 250 PGs were manually annotated by subject experts to lay down the ground truth for training AI algorithms on the segmentation of periapical pathosis. Two approaches were used for lesion detection: Multi-models 1 and 2, using U-net and Mask RCNN algorithms, respectively. The resulting segmented lesions generated on the testing data set were superimposed with results of teeth segmentation and numbering algorithms trained separately to relate lesions to causative teeth. Hence, both multi-model approaches related periapical pathosis to the causative teeth on PGs. RESULTS: The performance metrics of lesion segmentation carried out by U-net are as follows: Accuracy = 98.1%, precision = 84.5%, re-call = 80.3%, F-1 score = 82.2%, dice index = 75.2%, and Intersection over Union = 67.6%. Mask RCNN carried out lesion segmentation with an accuracy of 46.7%, precision of 80.6%, recall of 55%, and F-1 score of 63.1%. CONCLUSION: In this study, the multi-model approach successfully related periapical pathosis to the causative tooth on PGs. However, U-net outperformed Mask RCNN in the tasks performed, suggesting that U-net will remain the standard for medical image segmentation tasks. Further training of the models on other findings and an increased number of images will lead to the automation of the detection of common radiographic findings in the dental diagnostic workflow.

Comparative evaluation of ethanol versus citric acid in removal of oil-based calcium hydroxide from the apical third of root canal system.

Objectives: To evaluate the effectiveness of ethanol compared to citric acid in the removal of oil-based calcium hydroxide from the apical third of the root canal system using passive ultrasonic irrigation. METHODS: The in vitro study was conducted from September to October 2021 at the dental clinics of the Aga Khan University Hospital, Karachi, and comprised single-rooted teeth that were selected from institutional bank of extracted teeth. They were randomly divided into group A having 70% ethanol + passive ultrasonic irrigation, group B 10% citric acid + passive ultrasonic irrigation, group C positive controls and group D negative controls. The specimens were sectioned at 1mm and 3mm from the apex and examined under a dental operating microscope. A single examiner scored the specimens on two different occasions. Data was analysed using SPSS 25. RESULTS: Of the 90 teeth, there were 40(44.4%) in each of the 2 experimental groups and 5(5.5%) in each of the 2 control groups. At 3mm apical sections, ethanol was significantly more effective in the removal of oil-based calcium hydroxide (p=0.01). However, at 1mm from the apex, there was no significant difference between the experimental groups (p=0.064). Intragroup comparison showed that for groups A and B, residual medicament at 1mm sections was significantly greater than at 3mm sections (p<0.001, p=0.003). CONCLUSIONS: Neither irrigant showed complete removal at 1mm and 3mm from the apex. However, at 3mm apical sections, 70% ethanol was significantly more effective compared to 10% citric acid.

Restoration of endodontically treated teeth: A cost-effectiveness analysis of an endocrown versus a complete crown.

STATEMENT OF PROBLEM: Endocrowns have been proposed as an alternative to post-and-core retained complete crowns for structurally compromised endodontically treated teeth. However, an analysis of their cost-effectiveness is lacking. PURPOSE: The purpose of this simulation study was to assess the cost-effectiveness of an endocrown versus a complete crown as a definitive restoration for structurally compromised endodontically treated teeth. MATERIAL AND METHODS: A Markov simulation model was constructed with endodontically treated permanent molar teeth using TreeAge Pro Healthcare (2023) as a starting point for an 18-year-old patient. Costs were extrapolated from the ADA dental survey based on the United States healthcare, and the probabilities of transition were derived from existing literature. The cost-effectiveness was determined by using Monte Carlo microsimulations. A sensitivity analysis was performed to validate the model internally, whereas an experienced health expert and an endodontist performed the face validation. RESULTS: The complete crown was associated with additional health benefits (1.36 and 0.9 more years over a period of 5 years and lifetime, respectively) but at an increased cost (an additional 1143 USD and 1535 USD over a period of 5 years and lifetime, respectively). Moreover, the endocrown was cost-effective at lower Willingness-To-Pay (WTP) values (92% acceptable at 250 USD for 5 years and 73% acceptable at 250 USD for the lifetime of an individual), whereas at increased WTP threshold values, the complete crown was a cost-effective restoration (98.6% acceptable at 1250 USD for 5 years and 99.5% acceptable at 8000 USD over an individual's lifetime). CONCLUSIONS: The endocrown was a cost-effective restorative option at lower WTP values. However, at an increased WTP threshold, the complete crown became a more cost-effective restoration.

Generative artificial intelligence: synthetic datasets in dentistry.

INTRODUCTION: Artificial Intelligence (AI) algorithms, particularly Deep Learning (DL) models are known to be data intensive. This has increased the demand for digital data in all domains of healthcare, including dentistry. The main hindrance in the progress of AI is access to diverse datasets which train DL models ensuring optimal performance, comparable to subject experts. However, administration of these traditionally acquired datasets is challenging due to privacy regulations and the extensive manual annotation required by subject experts. Biases such as ethical, socioeconomic and class imbalances are also incorporated during the curation of these datasets, limiting their overall generalizability. These challenges prevent their accrual at a larger scale for training DL models. METHODS: Generative AI techniques can be useful in the production of Synthetic Datasets (SDs) that can overcome issues affecting traditionally acquired datasets. Variational autoencoders, generative adversarial networks and diffusion models have been used to generate SDs. The following text is a review of these generative AI techniques and their operations. It discusses the chances of SDs and challenges with potential solutions which will improve the understanding of healthcare professionals working in AI research. CONCLUSION: Synthetic data customized to the need of researchers can be produced to train robust AI models. These models, having been trained on such a diverse dataset will be applicable for dissemination across countries. However, there is a need for the limitations associated with SDs to be better understood, and attempts made to overcome those concerns prior to their widespread use.

Irreversible pulpitis in mature permanent teeth: a cost-effectiveness analysis of pulpotomy versus root canal treatment.

INTRODUCTION: Evidence-based dentistry suggests pulpotomy as a potential alternative to root canal treatment in mature permanent teeth with irreversible pulpitis. However, the evidence surrounding the cost-valuation and cost-efficacy of this treatment modality is not yet established. In this context, we adopted an economic modeling approach to assess the cost-effectiveness of pulpotomy versus root canal treatment, as this could aid in effective clinical decision-making. METHODS: A Markov model was constructed following a mature permanent tooth with irreversible pulpitis in an 18-year-old patient over a lifetime using TreeAge Pro Healthcare 2022. Transition probabilities were estimated based on existing literature. Costs were estimated based on the United States healthcare following a private-payer perspective and parameter uncertainties were addressed using Monte-Carlo simulations. The model was validated internally by sensitivity analyses, and face validation was performed by an experienced endodontist and health economist. RESULTS: In the base case scenario, root canal treatment was associated with additional health benefit but at an increased cost (1.08 more years with an incremental cost of 311.20 USD) over a period of an individual's lifetime. The probabilistic sensitivity analysis revealed pulpotomy to be cost-effective at lower Willingness-To-Pay (WTP) values (99.9% acceptable at 50 USD) whereas increasing the values of WTP threshold root canal treatment was a cost-effective treatment (99.9% acceptable at 550 USD). CONCLUSION: Based on current evidence, pulpotomy was a cost-effective treatment option at lower WTP values for the management of irreversible pulpitis in mature permanent teeth. However, by increasing the WTP threshold, root canal treatment became a more cost-effective treatment option over a period of lifetime of an individual.

Research and application of artificial intelligence in dentistry from lower-middle income countries - a scoping review.

Artificial intelligence (AI) has been integrated into dentistry for improvement of current dental practice. While many studies have explored the utilization of AI in various fields, the potential of AI in dentistry, particularly in low-middle income countries (LMICs) remains understudied. This scoping review aimed to study the existing literature on the applications of artificial intelligence in dentistry in low-middle income countries. A comprehensive search strategy was applied utilizing three major databases: PubMed, Scopus, and EBSCO Dentistry & Oral Sciences Source. The search strategy included keywords related to AI, Dentistry, and LMICs. The initial search yielded a total of 1587, out of which 25 articles were included in this review. Our findings demonstrated that limited studies have been carried out in LMICs in terms of AI and dentistry. Most of the studies were related to Orthodontics. In addition gaps in literature were noted such as cost utility and patient experience were not mentioned in the included studies.

Analysis of impact of consumption of proton pump inhibitor on osseo integration of dental implants to the alveolar bone: A systematic critical appraisal.

Proton pump inhibitors are one of the most frequently prescribed medicines primarily for reducing the production of gastric acid. Every medicine has some adverse effects associated with it, including effects on the bone tissues. Dental implant is one of the most preferred options for teeth replacement. The current literature review was planned to evaluate the association between intake of proton pump inhibitors and its impact on the bone around the dental implant. Literature review entailed search on Google Scholar, Web of Science and PubMed databases using a range of search terms. Chronic intake of proton pump inhibitors has been associated with decrease in the density of bone, which eventually leads to increased risk of dental implant failure. However, since limited studies have been carried out, further research is required, especially clinical trials, to evaluate the relationship between the intake of proton pump inhibitors and the failure of dental implants.

Cochrane systematic reviews in dentistry: an Altmetric and network analysis.

Introduction Cochrane systematic reviews (CSRs) play an important role in evidence-based decision-making. Therefore, the present study aimed to determine the social impact of CSRs in dentistry and the inclusivity and diversity of researchers contributing to one of the largest databases in health care research.Methodology The Altmetric and bibliometric data for CSRs in dentistry were obtained through Altmetric Explorer and the Dimensions database and were analysed to determine the trends. Furthermore, the correlation between the number of citations and the Altmetric Attention Score (AAS) was identified using Spearman's correlation co-efficient.Results Mendeley was found to be the most active Altmetric resource, followed by Twitter. The tweets were more popular among the members of the public (65.5%) and had a diverse geographic spread. The co-authorship network analysis revealed an overall dense network of researchers. In the co-citation network analysis, the Journal of Community Dentistry had the greatest influence. Moreover, a weaker correlation was noticed between the citation counts and AAS (rs=0.325; p <0.01).Conclusion CSRs had a modest social impact in terms of AAS; however, the social network of contributing researchers was diverse and the researchers affiliated with the University of Manchester, UK were found to have the strongest link.

The Dilemma Of Management Of Cystic Lesions; An Uncertain Way Forward: A Case Report.

Periapical diseases ranges from mild granulomatous lesions to large cystic ones, with the treatments corresponding to their respective pre-operative diagnoses. However, the determination of cause of periapical radiolucency is impossible on pre-operative clinical and radiographic examinations. We present a case highlighting the difficulties encountered in treating a periapical cyst using the current evidence in literature. It demonstrates the uncertainty involved in treating such lesions, owing to the impossible nature of determining the histopathological nature of the cyst, i.e., being either true cysts or pocket cysts. This case includes orthograde re-treatment; decompression of the cystic lesion, followed by peri-apical surgery of two teeth over a course of three years; and the uncertain outcomes encountered after each phase of the treatment.

Implementation of transfer learning for the segmentation of human mesenchymal stem cells-A validation study.

INTRODUCTION: Stem cell therapy has been gaining interest in the regeneration rather than repair of lost human tissues. However, the manual analysis of stem cells prior to implantation is a cumbersome task that can be automated to improve the efficiency and accuracy of this process. OBJECTIVE: To develop a Deep Learning (DL) algorithm for segmentation of human mesenchymal stem cells (MSCs) on micrographic images and to validate its performance relative to the ground truth laid down via annotation. METHODOLOGY: Pre-trained DeepLab algorithms were trained on annotated images of human MSCs obtained from the open-source EVICAN dataset. This dataset comprises of partially annotated images; a limitation that is overcome by blurring backgrounds of these images which consequently blurs the unannotated cells. Two algorithms were trained on the two different kinds of images from this dataset; with blurred and normal backgrounds, respectively. Algorithm 1 was trained on 139 images with blurred backgrounds and algorithm 2 was trained on 37 images from the same dataset with normal backgrounds to replicate real-life scenarios. RESULTS: The performance metrics of algorithm 1 included accuracy of 99.22%, dice co-efficient of 99.66% and Intersection over Union (IoU) score of 0.84. Algorithm 2 was 96.34% accurate with dice co-efficient and IoU scores of 98.39% and 0.48, respectively. CONCLUSION: Both algorithms showed adequate performance in the segmentation of human MSCs with performance metrics close to the ground truth. However, algorithm 2 has better clinical applicability, even with smaller dataset and relatively lower performance metrics.

Writing Letters to the Editors as a Method to Teach and Support Learning.

Null.

Oculo-dento-digital dysplasia: a systematic analysis of published dental literature.

INTRODUCTION: Oculo-dento-digital dysplasia (ODDD, OMIM# 164200) is a rare genetic disorder caused by mutation in Gap junction alpha gene that encodes connexin 43 (Cx43) protein. In this paper, the case of a 16-year-old boy is reported who presented with the complaint of toothache. Examination revealed unusual facial features, i.e., long narrow nose, hypertelorism, prominent epicanthal folds along with syndactyly and camptodactyly. We have also compiled available dental literature on ODDD that will help clinicians in early diagnosis and management of this condition. MATERIALS AND METHODS: A literature search was performed in PubMed NLM, EBSCO Dentistry & Oral Sciences Source, and EBSCO CINAHL Plus. RESULTS: A total of 309 articles were identified in the literature search. Only 17 articles were included based on the predetermined inclusion and exclusion criteria in the review synthesis. The included articles were case reports (n = 15), a case report and review (n = 1), and an original article (n = 1). Enamel hypoplasia, hypomineralization, microdontia, pulp stones, curved roots, and taurodontism were common dental findings in ODDD. CONCLUSIONS: After establishing definitive diagnosis, a multidisciplinary team should work in cohesion to improve the quality of life of patients. Immediate treatment should be focused on the correction of current oral condition and symptomatic treatment. In the long term, attention should be diverted to prevent tooth wear and maintaining the occlusal vertical dimension to establish adequate function.

An artificial intelligence model for instance segmentation and tooth numbering on orthopantomograms.

AIM: To develop a deep learning (DL) artificial intelligence (AI) model for instance segmentation and tooth numbering on orthopantomograms (OPGs). MATERIALS AND METHODS: Forty OPGs were manually annotated to lay down the ground truth for training two convolutional neural networks (CNNs): U-net and Faster RCNN. These algorithms were concurrently trained and validated on a dataset of 1280 teeth (40 OPGs) each. The U-net algorithm was trained on OPGs specifically annotated with polygons to label all 32 teeth via instance segmentation, allowing each tooth to be denoted as a separate entity from the surrounding structures. Simultaneously, teeth were also numbered according to the Fédération Dentaire Internationale (FDI) numbering system, using bounding boxes to train Faster RCNN. Consequently, both trained CNNs were combined to develop an AI model capable of segmenting and numbering all teeth on an OPG. RESULTS: The performance of the U-net algorithm was determined using various performance metrics including precision = 88.8%, accuracy = 88.2%, recall = 87.3%, F-1 score = 88%, dice index = 92.3%, and Intersection over Union (IoU) = 86.3%. The performance metrics of the Faster RCNN algorithm were determined using overlap accuracy = 30.2 bounding boxes (out of a possible of 32 boxes) and classifier accuracy of labels = 93.8%. CONCLUSIONS: The instance segmentation and tooth numbering results of our trained AI model were close to the ground truth, indicating a promising future for their incorporation into clinical dental practice. The ability of an AI model to automatically identify teeth on OPGs will aid dentists with diagnosis and treatment planning, thus increasing efficiency.