To present a feasible workflow for artificial intelligence (AI)-assisted software engineering in dentistry as a technical innovation report. The use of this workflow is illustrated through three self-developed open-source dental applications. Four AI-assisted development approaches were employed: chat-based interfaces of large language models, command-line interface tools, integrated development environments with AI assistance, and agent-based architectures. The dental software applications were created by a single clinician without formal programming training. Three applications were created: (1) VirtualEndo Converter, a Blender add-on for automated CBCT derived STL conversion for augmented/virtual reality (AR/VR), (2) MeshComparisonTool, a 3D Slicer extension for quantitative 3D morphology comparisons, and (3) DentalEmergencyTrainer, an application for simulating dental trauma emergency calls. All the applications are publicly available under the MIT license on GitHub. This report demonstrates that AI-assisted software development can enable dental practitioners without formal programming training to create functional prototypes of applications for research, education, and potentially clinical support. However, the reproducibility of this approach remains to be established, as the three tools were developed by a single clinician, and their clinical deployment would require thorough validation, security auditing, and regulatory assessment. AI-assisted development can help dental practitioners prototype tools that address unmet needs in clinical workflows, research, and education, but clinical use requires cautious separation from validated medical software. Before deployment, such tools require defined intended use, safety evaluation, data-protection safeguards, maintenance plans, and regulatory assessment.
Frank Wilkinson qualified from Liverpool University in dentistry and medicine. He then gained important experience in the Royal Army Medical Corps, working alongside leaders in the evolving fields of maxillofacial and plastic surgery. Frank rose through the ranks of academia to become dental professor and dean of three dental schools, in Melbourne, Manchester and London. It was at the Eastman Dental Institute that he made his major mark, developing what had been a local children's dental clinic into a world-famous postgraduate dentistry hospital and school.
The aim of this retrospective study was to evaluate the long-term clinical outcomes of implant-supported prostheses and to analyze the type, frequency, and distribution of prosthetic complications in relation to prosthesis-related factors. Clinical records of patients who received implant-supported prosthetic treatment between 2017 and 2024 at the Department of Prosthodontics, Faculty of Dentistry, Gazi University, were retrospectively reviewed. A total of 269 patients with 778 implants were included. The follow-up period ranged from 14 to 228 months after functional loading, with a mean follow-up duration of 54.6 months. Fixed, hybrid, and implant-supported overdentures were evaluated. Prostheses were classified according to prosthesis type, restorative material, retention type, and impression level. Prosthetic complications were recorded. Statistical analyses included chi-square tests to evaluate complication distributions and Cox proportional hazards regression analysis to identify factors associated with time to first prosthetic complication. Statistical significance was set at P < 0.05. Implant-supported prostheses showed favorable long-term clinical outcomes; however, biological, mechanical, and technical complications were observed. The most common complications were screw loosening, prosthetic fractures and chipping of the veneering material. Complication rates differed significantly according to restorative material, prosthesis type, retention type, and impression level. Fixed partial dentures showed higher rates of chipping and screw-related complications compared with implant-supported single crowns. Hybrid prostheses, particularly acrylic-based prostheses, showed a higher incidence of acrylic tooth fractures. Cement-retained prostheses were more frequently associated with peri-implantitis, whereas screw-retained prostheses showed higher rates of mechanical complications. Cox regression analysis demonstrated that prosthesis type, restorative material, impression level, and retention type significantly influenced the risk of prosthetic complications (P < 0.05). Implant-supported prostheses exhibited high long-term survival in the present study; nevertheless, prosthetic complications were observed throughout the follow-up period. Prosthesis type, restorative material, retention type, and impression level were significantly associated with prosthetic complications. Appropriate treatment planning, material selection, regular follow-up, and patient compliance are essential to improve long-term clinical outcomes.
Cone Beam Computed Tomography (CBCT) provides detailed anatomical information for treatment planning in dentistry. However, manually identifying tooth and jawbone structures is time-consuming and can vary depending on the observer. The aim of this study is to analyze the performance of U-Net, DeepLab V3+, and YOLO V3 deep learning architectures for automatic segmentation of tooth and jawbone structures in CBCT images. This study utilized CBCT data from seven different patients. A total of 1,155 axial images were expertly labeled in terms of tooth and jawbone regions. The dataset was trained with U-Net, DeepLab V3+, and YOLO V3-based semantic segmentation models. The learning rate, number of epochs, and batch size parameters of the models were optimized using the GridSearch method. Dice Similarity Coefficient (DSC), Intersection over Union (IoU), precision, and recall performance evaluation metrics were used in the performance assessment. Successful results were obtained in tooth and jawbone segmentation in all deep learning models used in the study. Among the models used in the article, the most successful performance was obtained from the U-Net architecture. The U-Net model achieved DSC=0.9289, IoU=0.8671, precision=0.9213, and recall=0.9365 values with a learning rate of 0.001, 70 epochs, and 16 batch size parameters. The DeepLab V3+ deep learning algorithm also yielded similar results, while YOLO V3 showed lower performance compared to other models. Deep learning-based segmentation methods provide high accuracy in the automatic determination of tooth and jawbone structures in CBCT images. Among the deep learning models used in the study, U-Net was identified as the most successful model. The approach developed in this study has the potential to support treatment planning in dental applications and reduce the burden of manual segmentation. However, the method needs to be validated with larger, multi-center datasets before it can be put into clinical use.
Despite increasing interest in incorporating courses about the social determinants of health (SDOH) in dental curricula, there is limited knowledge about how ready students feel to address SDOH in their future practice. The purpose of this investigation was to describe dental students' perceptions of undergraduate learning about SDOH and to explore how prepared they felt to address SDOH in their future practice. Methods In this qualitative descriptive investigation, we invited third-year undergraduate dental students from the Faculty of Dental Medicine and Oral Health Sciences of McGill University, Montreal, to participate. Fifteen students took part in face-to-face, qualitative, semi-structured interviews, which were audio-recorded, transcribed verbatim and thematically analyzed. The participants reported having taken several courses related to social dentistry and explained that they were ready and willing to identify patients' SDOH. The participants recognized the importance of understanding people's oral health-related practices in the context of their social and cultural environments. However, even though they were willing to promote preventive practices tailored to patients' context of life and were open to participating in charitable community activities, they felt they lacked the necessary skills-and perhaps the motivation-to conduct upstream interventions such as advocacy. Current dental education attempts to sensitize future dentists about SDOH may not be enough to sustain effective practice changes. It would be appropriate for dental schools to collectively reflect on how to better train students on this subject.
Artificial intelligence models are increasingly used in healthcare education; however, their ability to handle both factual knowledge and analytical clinical reasoning in dentistry remains unclear. This study aimed to compare the accuracy of different AIs in answering knowledge-based and analytical multiple-choice questions in Oral and Maxillofacial Radiology (OMFR) and Oral and Maxillofacial Surgery (OMFS), and to evaluate performance differences according to cognitive task type. This cross-sectional comparative study analyzed 258 multiple-choice questions from the Turkish Dental Specialty Examination (DUS) conducted between 2012 and 2021 (202 knowledge-based, 56 analytical). Five AI models (ChatGPT-5.2 Go, ChatGPT-5.2 Plus, DeepSeek V3, Claude Sonnet 4.5, and Gemini 3 Flash) answered all questions under default settings in a single session. Accuracy rates were compared using Chi-square and Kruskal-Wallis tests with Bonferroni correction. Inter-model agreement and reliability were assessed using Cohen's kappa and the intraclass correlation coefficient (ICC) (α = 0.05). Significant differences among models were observed in knowledge-based questions (p = 0.048), analytical questions (p = 0.032), and overall accuracy (p = 0.006). Gemini achieved the highest accuracy in knowledge-based questions, while Claude demonstrated the lowest performance. Although a general difference was detected in analytical questions, pairwise comparisons did not show clear model superiority. Overall performance largely reflected success in knowledge-based tasks. Agreement analysis showed low kappa values (κ = 0.226-0.339) but moderate ICC levels (0.597-0.728). AIs demonstrate strong factual recall but remain limited in analytical clinical reasoning tasks. While these models may serve as supportive tools in dental education, their use as independent clinical decision-making systems is not yet reliable.
Remaining Root Dentin Thickness (RRDT) after root canal instrumentation is an important parameter to be assessed with newer file systems. This study aimed to evaluate the root dentin thickness following instrumentation of primary molar root canals using manual or pediatric rotary file systems. Fifty-one extracted human primary molars were prepared with access cavity preparation and working length determination. Preliminary Cone Beam Computed Tomography (CBCT) images were obtained with specimens mounted on vinyl polysiloxane templates. Root dentin thickness was measured at three predetermined levels. 51 mesiobuccal canals were randomly divided into 3 groups of 17 each. After instrumentation using the allocated file system i.e. Group 1 - Manual K files; Group 2 - Kedo SG Blue and Group 3 - Kedo S Square rotary files, CBCT images of the specimens were obtained. The remaining root dentin thickness was obtained using measurements of pre-and post-instrumentation CBCT. Data was analysed using paired t-test for intragroup comparisons. Intergroup comparison of RRDT was done using ANOVA. The Remaining Root Dentin Thickness (RRDT) after instrumentation did not differ significantly among the three study groups at the three predetermined levels (p > 0.05). Kedo S Square and Kedo SG Blue file systems showed comparable effectiveness to the manual K file system in the instrumentation of mesiobuccal canals of primary molars.
Clinicians aspire to predict the emergence of Bipolar Disorder (BD) in a timely manner. To accomplish this, markers reflecting mental states that can be gathered non-invasively and at large scale are needed. Here, we systematically evaluate evidence relating speech-based markers to mood states in BD. We searched Medline and Google Scholar for all published studies in English up to February 2026 on the use of speech markers in BD. We undertook thematic analysis on abstracts using topic modeling and a qualitative gap analysis to identify potential opportunities for future research. 43 out of 867 studies were included after screening. Topic analysis revealed an emerging focus on mapping mood states to automated speech features. Most studies focused on cross-sectional detection of bipolar mood states, or BD as a diagnosis, rather than the prediction of upcoming mood states. Speech features distinguished BD from schizophrenia, depression, and healthy controls. Manic states were characterized by quantifiable measures of pressured speech, derailment, grammatical errors, and word repetition; depressive states by an increased use of personal pronouns, reduced verbal fluency, and speech quantity. Overall, attempts to replicate observations were limited. Acoustic and lexical-semantic markers vary with manic, psychotic, or depressive states. At present, the evidence is insufficient for clinical utility in relapse prediction, response monitoring, or diagnosing mixed episodes or state changes in BD. We recommend that future research leverages the growing capabilities of natural language processing through longitudinal and cross-linguistic studies to strengthen the evidence base and advance the clinical utility of speech markers for BD.
Cervical cancer is a leading cause of cancer mortality among Ghanaian women, yet screening uptake is under 5%. The Cervical Cancer Prevention and Training Centre (CCPTC) partnered with Rotary Clubs across the country to implement the first-ever nationally representative cervical precancer screening project and to demonstrate the feasibility of an integrated nationwide screening program. We conducted a cross-sectional analysis of 1,636 asymptomatic women aged 25 years and above screened at 29 government and private facilities across all 16 regions of Ghana (January-February 2025). Eligible women underwent concurrent hr-HPV genotyping (Sansure MA-6000 platform) and VIA by CCPTC-trained alumni, with immediate thermal ablation for eligible VIA-positive lesions (TZ type 1 or 2). Multivariable logistic regression (backward stepwise elimination, P < 0.25 retention threshold) identified factors associated with hr-HPV positivity and VIA positivity. Analyses were performed in Stata v17.0. Among 1,636 women, the overall hr-HPV prevalence was 26·6% (95% CI, 24·5-28·8) and the VIA 'positivity' was 4·0% (95% CI, 3·1-5·0). Predominant genotypes were HPV52 (5·3%), HPV58 (4·4%), and HPV51 (3·6%); HPV16 and HPV18 together accounted for <5% of infections. Independent factors associated with hr-HPV infection were HIV infection (aOR=5·77; 95% CI, 2·07-16·13, P = 0.001) and having a steady partner (aOR=2·02; 95% CI, 1·22-3·36, P = 0.006); being married/cohabiting (aOR=0·51; 95% CI, 0·38-0·69, P < 0.001) or widowed (aOR=0·43; 95% CI, 0·23-0·82, P = 0.011), and prior screening (aOR=0·67; 95% CI, 0·48-0·92, P = 0.014) were protective. VIA 'positivity' was independently associated with HIV infection (aOR 7.49, 95% CI 1.99-28.19, P = 0.003). Regional hr-HPV prevalence varied from 10·0% to 39·2%. Thirty-five percent of VIA-positive women received same-visit thermal ablation. This decentralized alumni-driven model integrating off-site HPV testing, task-shifted VIA, and immediate thermal ablation proved operationally feasible across Ghana's diverse health system and revealed a substantial hr-HPV burden. The approach offers a scalable blueprint for national cervical cancer control and informs Ghana's transition toward HPV-based screening.
The purpose of this study is to investigate the feasibility and diagnostic accuracy of ultrasonography in peri-implant defect identification and evaluation, and to compare ultrasonography (USG) with cone-beam computed tomography (CBCT) in defect identification and evaluation in an in vitro model. Seventy-two implants were placed in fresh porcine rib bone models with different types of artificially created defects (including buccal dehiscence, circumferential defect, circumferential defect with buccal dehiscence), and in control sites with no defect. Both USG and CBCT were used to assess the presence, type, and linear measurements (height and depth) of the defects. Diagnostic performance metrics, including the sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of USG and CBCT in characterizing different types of defects, were examined. Linear measurement results obtained by using USG and CBCT were compared with the direct measurement as the gold standard, with the computation of the intraclass correlation coefficient (ICC). USG demonstrated satisfactory diagnostic accuracy (> 95%) in defect type detection across all types. Its sensitivity in detecting circumferential defects (83.3%) and its positive predictive value for no-defect sites (87.0%) were lower. CBCT showed perfect diagnostic accuracy (100%) for defect detection. USG illustrated high agreement with direct measurements for defect depth (ICC = 0.934, p < 0.001) but poor agreement for defect height (ICC = 0.349, p < 0.001), underlining its limitations compared with CBCT. USG is a reliable, non-ionizing diagnostic tool for characterizing peri-implant defect types with performance comparable to CBCT. However, it underestimates defect height measurements in defects with infrabony components, where CBCT remains superior. This in vitro study established an experimental model and investigated the use of ultrasonography (USG) in detecting and measuring bone defects around dental implants as compared with cone‐beam computed tomography (CBCT) and direct visual assessment. Results showed that USG was generally reliable in peri‐implant defect type detection; however, it exhibited a mean measurement discrepancy of ∼0.9 mm, systematically underestimating the height of defects with an infrabony component. Clinicians should remain aware of these USG artifacts and potential bias when interpreting ultrasound data. CBCT demonstrated superior accuracy with a smaller margin of error (< 0.5 mm) that is less clinically significant. However, USG holds a significant advantage over CBCT in terms of being radiation‐free, making it an attractive tool for long‐term monitoring of dental implants. In the current clinical landscape, there are substantial challenges associated with the use of USG for assessing bone condition around dental implants. However, this study posits that with continued technological advancements, USG has the potential to become a non‐invasive and reliable tool for patient implant assessment. Therefore, we propose further development and research in this area.
Incidence of chronic obstructive pulmonary disease and asthma diagnosis were lower during and after the Coronavirus disease 2019 pandemic in Alberta, Canada. However, it is unknown whether incidences were actually lower or if the pandemic created circumstances where patients did not seek care. As such, the objective of the current study was to explore the impact of COVID-19 on patient and clinician experiences of healthcare access and delivery. The study was conducted between October 2023 and July 2024. We used interpretive description, a qualitative approach with the end-goal of informing clinical decisions. Analysis was informed by Braun and Clarke's six phases of reflexive thematic analysis. We completed thirteen interviews. Two key themes were generated: (1) The pandemic impacted care-seeking behaviours; and (2) A time and place for virtual and in-person care. Clinicians discussed how access to entry points to the health system were impacted by the pandemic and highlighted how strategies to manage health and stressors impacted symptoms and subsequent care-seeking behaviours. Participants highlighted the positives of virtual and in-person care with the consensus that both are valuable. Future use of virtual care modalities should include a visual element at minimum and prioritize the therapeutic relationship.
Gender disparities in research trial leadership have been reported in nearly every medical specialty. The true extent of women's involvement as principal investigators (PIs) in oral and maxillofacial surgery (OMS) research trials remains unknown. The current study aims to address this knowledge gap through analyzing relevant trials registered on ClinicalTrials.gov in the 21st century. A validated search strategy, comprising procedure names and field-specific terms, was utilized to identify OMS research trials. The analysis was limited to trials where at least one PI was an oral and maxillofacial surgeon or an OMS department-affiliated researcher. Trial characteristics, including study type, phase, funding status, and investigator details were extracted from records. PI gender was determined using Genderize.io. Overall, men occupied 73.7% of PI positions for OMS research trials, and women 26.3%. Linear regression analysis revealed a significant yearly increase in proportion of women PIs over time. Geographically, women PI representation was highest in African (predominantly Egypt) and Asian institutions, and lowest in European and North American institutions. Relative to other phases, women PIs were significantly more likely to lead Phase 3 trials and less likely to lead Phase 2 trials compared to men PIs. While government funding patterns were similar between genders, women PIs were significantly less likely to lead industry-funded trials. Despite recent improvements, women remain underrepresented in OMS research trial leadership, especially in North American institutions. Targeted interventions and cultural shift are essential to fostering gender-based parity in this domain.
Immigration Removal Centres (IRCs) are used to detain people for immigration control by the UK government. This scoping review aims to examine the experiences of detainees within UK IRCs, specifically how conditions within them, including the regime, affect their mental wellbeing. The Joanna Briggs Institute guideline for scoping reviews was followed. Six bibliographic databases and additional grey literature sources were searched for quantitative and qualitative evidence. Descriptive analyses and quality assessments were conducted. Fifteen research studies and nine pieces of grey literature were included, comprising a total of 1353 participants and 11 IRCs. The majority of data was qualitative in methodology and published after 2015. Main findings from articles were charted according to Maslow's (1943) Hierarchy of Needs, revealing persistent failings across all dimensions (physiological, safety, belongingness, esteem, and self-actualization). The regime within the IRCs as well as the psychosocial environment led to emotional distress and feelings of disempowerment, dehumanization, and criminality. This review highlights the negative impact of IRCs within the UK on the mental wellbeing of detainees and the need for urgent policy reform. Changes addressing temporal uncertainty of detention and use of community-based settings are proposed for UK governmental review.
MicroRNAs (miRNAs), small non-coding RNA molecules, have been considered as pivotal regulators of gene expression, influencing many biological functions, including nerve cell development, synaptic plasticity, and inflammatory responses. Their biogenesis involves a multi-step process, beginning with transcription by RNA polymerase II, followed by processing through Drosha and Dicer enzymes, culminating in the formation of mature miRNAs that integrate into the RNA-induced silencing complex [1] to control target messenger RNA (mRNA) stability and translation. In the context of epilepsy, a neurological disorder characterized by recurrent seizures, miRNAs have a role in the modulation of neuronal excitability and network synchronization. Dysregulation of specific miRNAs can disrupt the delicate balance between excitatory and inhibitory neurotransmission, contributing to the pathogenesis of epilepsy. Moreover, miRNAs influence neuroplasticity by regulating genes involved in synaptic remodeling and neuronal connectivity, processes that are often aberrant in epileptic brains. Inflammatory pathways are also modulated by miRNAs, with certain miRNAs acting as key regulators of cytokine expression and immune cell activation, thereby influencing neuroinflammation associated with epileptogenesis. Clinically, altered miRNA expression profiles have been shown in the blood and cerebrospinal fluid of epilepsy patients, suggesting their potential as non-invasive biomarkers for diagnosis and prognosis. Furthermore, therapeutic strategies targeting miRNAs are being explored, aiming to restore normal gene expression patterns and mitigate seizure activity. This review delves into the intricate roles of miRNAs in epilepsy, encompassing their biogenesis, involvement in disease pathophysiology, impact on neuroplasticity and inflammation, and their emerging clinical applications.
Widening participation in medicine is a key societal priority. To improve representation of non-traditional applicants, UK medical schools use contextual admissions, although definitions of under-represented groups vary across institutions. This study examined the educational and training trajectories of one such group-mature non-graduates. We aimed to determine whether their progression was comparable to that of school-leavers and graduate entrants, whether they progressed through medical school non-inferiorly, and whether they were equally likely to secure postgraduate training posts. Data on UK medical students and resident doctors (2007-2015) were extracted from the UK Medical Education Database (UKMED). Undergraduate performance (Educational Performance Measure (EPM), Prescribing Safety Assessment (PSA), Foundation Programme Situational Judgement Test (FP-SJT) and postgraduate performance (Membership of the Royal College of Physicians, MRCP and MRCGP Applied Knowledge Test, MRCGP AKT), Annual Review of Competency Progression (ARCP), and obtaining an offer for a Level 1 training post) were compared across the three groups using multivariate statistical analyses, including Kruskal-Wallis tests where appropriate. Across all measures of undergraduate academic attainment (EPM, PSA and FP-SJT) and postgraduate examination performance (MRCP Part 1, Part 2 and PACES), mature non-graduates performed less well than school-leavers or graduates. They were more frequently released from training (ARCP Outcome 4) and received higher numbers of Developmental ARCP outcomes. Mature non-graduates were also less likely to receive an offer for competitive specialty training on their first application; however, the likelihood of them applying exclusively to General Practice did not differ from that of school-leavers or graduates. Although performance differences persist for a subset of widening-participation candidates named here as mature non-graduates, they still progress through training, obtain Certificates of Completion of Training, and contribute to the medical workforce. Further research is needed to examine how their life experience and non-traditional educational trajectories influence their long-term practice.
Insulin resistance (IR) is a common complication in diabetic dogs, impairing glycemic control and causing metabolic dysfunction. To assess the utility and determine cutoff values of C-peptide, homeostasis model assessment of IR by C-peptide (HOMA-IR by C-peptide), HOMA-IR by C-Peptide version 2 (HOMA-IRCP2), and homeostasis model assessment 2 of IR (HOMA2-IR) as diagnostic tools for detecting IR and identifying factors related to IR in dogs with diabetes. Seventy-six dogs were enrolled from a referral center and classified into 4 groups: healthy dogs (Healthy group, n = 17), pre-diabetes mellitus (PreDM, n = 11), insulin-dependent diabetes mellitus (IDDM, n = 20), and insulin-resistant diabetes mellitus (DMIR, n = 28). A cross-sectional prospective study. Blood samples were collected from dogs from which food was withheld and analyzed for glucose, C-peptide, and other metabolic indicators. HOMA-IR by C-peptide, HOMA-IRCP2, and HOMA2-IR were determined and compared across groups. The DMIR group exhibited the highest C-peptide and HOMA-IR by C-peptide, HOMA-IRCP2, and HOMA2-IR. Receiver Operating Characteristic curve analysis demonstrated that C-peptide, HOMA-IR by C-peptide, HOMA-IRCP2, and HOMA2-IR identified IR, with area under curve of 0.843 (95% CI, 0.756-0.930), 0.897 (95% CI, 0.819-0.975), 0.893 (95% CI, 0.812-0.973), and 0.789 (95% CI, 0.686-0.893), respectively (P < 0.001). A cutoff level of ≥0.7 ng/mL for C-peptide and ≥120 for HOMA-IR by C-peptide demonstrated the highest sensitivity (92.9%), whereas ≥1.65 for HOMA-IRCP2 showed the highest specificity (97.9%). C-peptide, HOMA-IR by C-peptide, HOMA-IRCP2, and HOMA2-IR are promising, non-invasive indicators of IR. Early identification of IR using these markers might improve clinical outcomes.
Artificial intelligence (AI) is increasingly integrated into healthcare education and clinical practice. Understanding health sciences students' knowledge, attitudes, and practices (KAP) toward AI is important for informing curriculum development, particularly in resource-limited educational settings. To assess knowledge, attitudes, and practices toward artificial intelligence among health sciences students at a Palestinian university. A cross-sectional study was conducted during the 2024-2025 academic year among 666 undergraduate students from nursing, medicine and health sciences, and dentistry programs. Data were collected using a structured questionnaire assessing AI knowledge (7 items), attitudes (10 items), and practices (7 items). Descriptive statistics were calculated. Independent-samples t-tests and one-way analysis of variance (ANOVA) were used to examine group differences. Effect sizes were reported using Cohen's d and eta squared (η²). Statistical significance was set at p < 0.05. The overall AI knowledge accuracy rate was 42.9% (mean 3.00 ± 1.61), indicating limited foundational understanding, particularly regarding machine learning and deep learning concepts. Attitudes toward AI were generally positive (mean 3.60 ± 0.44), with high endorsement of ethical awareness. AI practice levels were moderate (mean 3.32 ± 0.70), with frequent use for assignments and research activities. Formal AI training was associated with higher knowledge scores (t(664)=7.45, p < 0.001, d = 0.69) and slightly more positive attitudes (t(664)=2.59, p = 0.010, d = 0.23), but not with practice frequency (p = 0.807). Differences across colleges and academic years demonstrated small effect sizes (η² ≤ 0.03). Health sciences students demonstrated positive attitudes and regular academic use of AI tools despite limited foundational knowledge. These findings suggest the potential benefit of structured AI literacy integration within health sciences curricula.
This 12-month, double-blind, parallel, randomized controlled trial compared the material fracture/retention rates of occluso-proximal direct composite resin restorations associated with polyethylene fiber (CR+PF; Ribbond®) with conventional CR restorations in primary molars. A total of 60 children received occluso-proximal restorations using either CR+PF or CR. Children should be 5 to 10 years were selected for the study. The restorations were evaluated at baseline and after 3, 6, and 12 months of clinical service according to the FDI World Dental Federation criteria. Additional data were collected regarding participant characteristics, restored teeth, and clinical working time. Statistical analyses included Kaplan-Meier survival and the Chi-square test. Cox regression analysis was used and independent t-test (α=0.05). After 12 months, 53 restorations were evaluated. Seven restorations failed (four for CR+PF group). The retention rates (95% confidence interval [CI]) were 85.7% (81.5%-99.9%) for CR+PF group and 88.0% (81.1%-96.0%) for the CR group (p > 0.05). The hazard ratio (95% CI) was 0.52 (0.27 to 1.01), also showing no significant difference between groups. All other FDI parameters indicated that the restorations were clinically acceptable. Cox regression analysis showed that age had a significant effect on clinical success (p < 0.05). The mean clinical working time was 30.7 (28.0%-33.3%) min in the CR group and 25.2 (21.4%-29.0%) min in the CR+PF group (p < 0.05). The survival rate of the occluso-proximal restorations with or without PF was not statistically different, although the use of PF was associated with a significantly longer clinical working time. This study, the first clinical trial to investigate this topic, demonstrates that the incorporation of polyethylene fibers does not improve the clinical performance of composite resin restorations. Therefore, their use is not justified in clinical practice, allowing simplification of the restorative protocol without compromising outcomes.
This study provides insight into the in-vivo three-dimensional rotational angiographic (3D-RA) evaluation of the vascular anatomy surrounding the temporomandibular joint (TMJ), emphasizing clinically relevant variations for surgical planning. Superficial temporal artery (STA), transverse facial artery, maxillary artery (MA), and middle meningeal artery (MMA) were retrospectively analysed in 67 patients (100 hemicrania) using 3D-RA to delineate their courses, bifurcation patterns, and spatial relationships with TMJ landmarks in relation to age and sex. Findings indicated that the STA followed a laterocondylar course in 51% of cases. Bifurcation was present in 99%, and a straight arterial course occurred in 71%. The STA-condyle distance increased with age (P = 0.001), suggesting potential surgical relevance in older patients. The MA was mostly superficial to the lateral pterygoid muscle, and, when located deeply, the MA-sigmoid notch distance significantly increased (P = 0.001), potentially affecting surgical procedures which are performed near the sigmoid notch. Sex-related differences were also observed: the TFA-glenoid fossa distance was larger in males (P = 0.001) and correlated with age (P = 0.016), while the MMA-articular eminence distance was greater in males (P = 0.001). These findings refine current understanding of TMJ vascular anatomy and provide detailed, in-vivo data to enhance preoperative assessment and minimize vascular injury risk during TMJ surgeries.
To assess the efficacy and safety of intravenous remimazolam besylate for sedation during dental procedures in patients with cognitive disabilities, a group frequently managed under general anesthesia due to poor cooperation and increased perioperative risks. In this single-center prospective single-arm observational study, 43 adult patients with cognitive or motor disabilities (ASA II-III) received outpatient dental care under intravenous remimazolam. Sedation was titrated to moderate levels according to EMA and IACSD guidelines. Outcomes included procedural success, Ramsay Sedation Scale, Post-Anesthetic Discharge Scoring System (PADSS), Modified Aldrete Score (MAS), vital signs, drug dosage, recovery, adverse events, and caregiver feedback. All treatments were successfully completed without anesthesiologist intervention or flumazenil reversal. Mean onset of sedation was 3.8 ± 2.3 min, recovery time 47 ± 20.4 min, and time to discharge 72.5 ± 22.5 min. The mean total remimazolam dose was 14 ± 5.9 mg. Most patients reached Ramsay 4, while 9.7% remained ≤ 3. Vital parameters remained stable with no episodes of hypoxemia or airway compromise. At discharge, PADSS averaged 8.9 ± 1.0 and MAS 9.4 ± 0.6, confirming recovery despite motor or neurological limitations affecting PADSS scoring. Caregiver interviews indicated that 42% of patients slept during the day, 85% slept normally at night, 68% appeared calmer, and only 7% experienced minor adverse effects such as nausea or brief agitation. Remimazolam provided safe and effective intravenous sedation with rapid onset and reliable recovery. Combining PADSS with MAS improved discharge assessment in this patient population. Remimazolam deserves further investigation as a potential practical alternative to general anesthesia for dental procedures in patients with cognitive disabilities, with the potential to enable safe care, procedural success, and favorable recovery profiles. International Standard Randomised Controlled Trial Number (ISRCTN) registry, ISRCTN39322806. Retrospectively registered on 10 November 2025.