RESUMO
INTRODUCTION: Generative artificial intelligence (GenAI) Large-language models such as ChatGPT have become increasingly popular in various fields. However, the impact of ChatGPT on dental research writing has yet to be quantified. This study aimed to assess ChatGPT's usage in dental research writing and discuss potential advantages and challenges. METHODS: Using a bibliometric design, we performed a keyword analysis of specific 'signaling words' indicative of ChatGPT use in the titles/abstracts of 299,695 dental research abstracts indexed PubMed 2018-2024. Statistical comparisons using normalized ratios per 10,000 dental publications compared changes in word frequency before and after the ChatGPT release on November 30, 2022. RESULTS: Before ChatGPT's release, the frequency of abstracts with signaling words was 47.1 per 10,000 papers. After the release, this increased to 224.2 per 10,000 papers, an increase of 177.2 per 10,000 papers (p = 0.014, 95 % CI 53.5-300.7). The word 'delve' showed the most significant usage increase (increased ratio=17.0). CONCLUSIONS: This study is among the first to systematically assess the use of GenAI, specifically ChatGPT, in dental research. We found evidence of the use and growth of ChatGPT in dental research publications. This trend indicates the widespread adoption of GenAI-assisted writing in scientific communication, consistent with other scientific fields. While GenAI can potentially increase productivity and inclusivity, it raises concerns such as bias, inaccuracy, and distortion of academic incentives. Therefore, our findings support the need for clear AI guidelines and standards for academic publishing to ensure responsible use and maintain scientific integrity.
RESUMO
BACKGROUND: Physics-informed neural networks (PINNs) have emerged as a powerful tool for solving inverse problems, especially in cases where no complete information about the system is known and scatter measurements are available. This is especially useful in hemodynamics since the boundary information is often difficult to model, and high-quality blood flow measurements are generally hard to obtain. METHODS: In this work, we use the PINNs methodology for estimating reduced-order model parameters and the full velocity field from scatter 2D noisy measurements in the aorta. Two different flow regimes, stationary and transient were studied. RESULTS: We show robust and relatively accurate parameter estimations when using the method with simulated data, while the velocity reconstruction accuracy shows dependence on the measurement quality and the flow pattern complexity. Comparison with a Kalman filter approach shows similar results when the number of parameters to be estimated is low to medium. For a higher number of parameters, only PINNs were capable of achieving good results. CONCLUSION: The method opens a door to deep-learning-driven methods in the simulations of complex coupled physical systems.
Assuntos
Modelos Cardiovasculares , Redes Neurais de Computação , Humanos , Velocidade do Fluxo Sanguíneo/fisiologia , Hemodinâmica/fisiologia , Aorta/fisiologia , Simulação por ComputadorRESUMO
The objective of this scoping review is to evaluate the potential of Magnetic Resonance Imaging (MRI) and to determine which of the available MRI techniques reported in the literature are the most promising for assessing treatment response in breast cancer patients following neoadjuvant radiotherapy (NRT). Ovid Medline, Embase, CINAHL, and Cochrane databases were searched to identify relevant studies published from inception until March 13, 2023. After primary selection, 2 reviewers evaluated each study using a standardized data extraction template, guided by set inclusion and exclusion criteria. A total of 5 eligible studies were selected. The positive and negative predictive values for MRI predicting pathological complete response across the studies were 67% to 88% and 76% to 85%, respectively. MRI's potential in assessing postradiotherapy tumor sizes was greater for volume measurements than uni-dimensional longest diameter measurements; however, overestimation in surgical tumor sizes was observed. Apparent diffusion coefficient (ADC) values and Time to Enhance (TTE) was seen to increase post-NRT, with a notable difference between responders and nonresponders at 6 months, indicating a potential role in assessing treatment response. In conclusion, this review highlights tumor volume measurements, ADC, and TTE as promising MRI metrics for assessing treatment response post-NRT in breast cancer. However, further research with larger cohorts is needed to confirm their utility. If MRI can accurately identify responders from nonresponders to NRT, it could enable a more personalized and tailored treatment approach, potentially minimizing radiation therapy related toxicity and enhancing cosmetic outcomes.
RESUMO
Recently, functional Near-Infrared Spectroscopy (fNIRS) was applied to obtain, non-invasively, the human perispinal Neuro-Vascular Response (NVR) under a non-noxious electrical stimulation of a peripheral nerve. This method allowed the measurements of changes in the concentration of oxyhemoglobin (O2Hb) and deoxyhemoglobin (HHb) from the perispinal vascular network. However, there is a lack of clarity about the potential differences in perispinal NVR recorded by the different fNIRS technologies currently available. In this work, the two main noninvasive fNIRS technologies were compared, i.e., LED and LASER-based. The recording of the human perispinal NVR induced by non-noxious electrical stimulation of a peripheral nerve was recorded simultaneously at C7 and T10 vertebral levels. The amplitude, rise time, and full width at half maximum duration of the perispinal NVRs were characterized in healthy volunteers and compared between both systems. The main difference was that the LED-based system shows about one order of magnitude higher values of amplitude than the LASER-based system. No statistical differences were found for rise time and for duration parameters (at thoracic level). The comparison of point-to-point wave patterns did not show significant differences between both systems. In conclusion, the perispinal NRV response obtained by different fNIRS technologies was reproducible, and only the amplitude showed differences, probably due to the power of the system which should be considered when assessing the human perispinal vascular network.
Assuntos
Lasers , Espectroscopia de Luz Próxima ao Infravermelho , Medula Espinal , Humanos , Espectroscopia de Luz Próxima ao Infravermelho/métodos , Masculino , Medula Espinal/irrigação sanguínea , Medula Espinal/diagnóstico por imagem , Medula Espinal/fisiologia , Adulto , Feminino , Adulto Jovem , Estimulação Elétrica , Hemoglobinas/análise , Hemoglobinas/metabolismoRESUMO
BACKGROUND: Toothpastes available in Europe contain a range of fluoride concentrations, with some meeting the recommended level for caries prevention (>1000 ppm fluoride (ppm)) and others containing low or no fluoride. This study evaluated toothpaste fluoride concentrations in Latvia and Lithuania to inform targeted public health strategies in regions with a high prevalence of dental caries. METHODS: This cross-sectional study was conducted from May 2019 to May 2020, using a validated questionnaire in Latvia and Lithuania. Nationally representative samples (1309 families and 5436 members) provided data through a mixed-mode survey (paper and online) on sociodemographic information, toothpaste type, brand, and type. Descriptive statistics and chi-square tests (p < 0.05) were used for analysis. RESULTS: Fifteen percent of families used non-fluoridated toothpaste and 12% used <1000 part per million (ppm) fluoride. In Latvia, 56.8% of preschoolers and 28.7% of schoolchildren used <1000 ppm or non-fluoride toothpaste, whereas in Lithuania, 47.2% of preschoolers and 29.1% of schoolchildren used <1000 ppm or non-fluoride toothpaste; 63% of adolescents and 73% of adults used toothpaste with optimal fluoride content (≥1000 ppm). Of the 228 registered toothpaste types, 62% contained more than 1000 ppm, which is optimal for caries prevention; 29% of Latvian and 24% of Lithuanian families used at least one non-fluoridated toothpaste. CONCLUSION: This study revealed significant gaps in the use of fluoride toothpaste among families in Latvia and Lithuania, especially among children. To effectively prevent dental caries, targeted interventions, and education must promote optimal fluoride toothpaste use, particularly among vulnerable populations.
RESUMO
Introduction: Obesity and its metabolic complications can impact the heart's structure and function in childhood, although demonstrating this impact has been challenging. New echocardiographic parameters such as left atrial strain (LAε) and left ventricular strain (LVε), as well as myocardial work (MW), could reveal subclinical alterations in cardiac function. Objective: The aim is to evaluate the feasibility of these parameters in adolescents with severe obesity and explore their associations with body fat, metabolic comorbidities, and physical capacity. Methods: This is a cross-sectional study in adolescents with obesity who underwent echocardiography with analysis of LAε, LVε and MW using speckle tracking. Feasibility and association with anthropometry, body fat percentage, comorbidities and cardiopulmonary test were analyzed. Results: Twenty adolescents were recruited, 13 (65%) were males, median age 14.2 (interquartile range: 12.9-14.9) years old. The median Z-score for BMI (zBMI) was +3.03 (2.87-3.14), 14 (70%) had severe obesity (zBMI ≥+3), 12 (60%) body fat ≥95th percentile, 9 (45%) high blood pressure (HBP) and 8 (40%) metabolic syndrome. The analysis of the echocardiographic parameters was feasible in 95% (LAε) and 100% (LVε and MW). LVε was lower in adolescents with vs. without metabolic syndrome: 17.8% (17.5-19.3%) vs. 19.3% (18.3-20.3%), Pâ =â 0.046; and with vs. without HBP 17.8% (17.5-18.6%) vs. 19.7% (18.4-20.3%), Pâ =â 0.02. Those with body fat ≥95th percentile had lower LAε and MW parameters, without association with cardiopulmonary test. Conclusion: Echocardiographic evaluation of LAε, LVε and MW is feasible in adolescents with severe obesity. A higher proportion of body fat and the presence of comorbidities are associated with alterations in these new echocardiographic functional parameters suggesting myocardial impact of higher metabolic compromise.
RESUMO
INTRODUCTION: Interest is growing in the potential of artificial intelligence (AI) chatbots and large language models like OpenAI's ChatGPT and Google's Gemini, particularly in dental education. To explore dental educators' perceptions of AI chatbots and large language models, specifically their potential benefits and challenges for dental education. MATERIALS AND METHODS: A global cross-sectional survey was conducted in May-June 2023 using a 31-item online-questionnaire to assess dental educators' perceptions of AI chatbots like ChatGPT and their influence on dental education. Dental educators, representing diverse backgrounds, were asked about their use of AI, its perceived impact, barriers to using chatbots, and the future role of AI in this field. RESULTS: 428 dental educators (survey views = 1516; response rate = 28%) with a median [25/75th percentiles] age of 45 [37, 56] and 16 [8, 25] years of experience participated, with the majority from the Americas (54%), followed by Europe (26%) and Asia (10%). Thirty-one percent of respondents already use AI tools, with 64% recognising their potential in dental education. Perception of AI's potential impact on dental education varied by region, with Africa (4[4-5]), Asia (4[4-5]), and the Americas (4[3-5]) perceiving more potential than Europe (3[3-4]). Educators stated that AI chatbots could enhance knowledge acquisition (74.3%), research (68.5%), and clinical decision-making (63.6%) but expressed concern about AI's potential to reduce human interaction (53.9%). Dental educators' chief concerns centred around the absence of clear guidelines and training for using AI chatbots. CONCLUSION: A positive yet cautious view towards AI chatbot integration in dental curricula is prevalent, underscoring the need for clear implementation guidelines.
RESUMO
Knowledge gaps exist regarding optimal silver diammine fluoride (SDF) regimens and the efficacy of new products for arresting dental caries in young children. We evaluated the effectiveness of 38%-SDF (SDI-RivaStar), Tiefenfluorid (TF) comparing with Placebo (P), all in conjunction with behavioural modification (BM), in preventing major complications (endodontic/extractions/pain)-a patient-centred outcome-due to early childhood caries over 12 months in children under 71-months. A six-arm, patient/parent-blinded, superiority, placebo-controlled randomised control trial at the university clinic in Riga, Latvia, from 1/9/20-31/8/22 (Protocol registration ISRCTN17005348). The trial tested six protocols, using three compounds (P/SDF/TF) under two regimes: annual and biannual (P1/P2/TF1/TF2/SDF1/SDF2) for major complications. Secondary outcomes included minor complications and parental satisfaction. All groups received BM. 373/427 randomised children (87.3%) completed the study. SDF2 had a significantly lower rate and risk of major (21.5%, OR = 0.28, 95%CI [0.11, 0.72], p < 0.05) and minor complications (OR = 0.16 (95%CI [0.05, 0.50], p = 0.002). Overall satisfaction was 96% (p > 0.05). SDF biannual application with BM effectively prevented major complications of early childhood caries and was well accepted by children and their parents. Trial registration number: ISRCTN17005348, principal investigator: Ilze Maldupa, registration date: 30/06/2021.Clinical trial registration number: ISRCTN registry: ISRCTN17005348, https://doi.org/10.1186/ISRCTN17005348 , registration date: 30/06/2021.
Assuntos
Cariostáticos , Cárie Dentária , Pré-Escolar , Humanos , Cariostáticos/uso terapêutico , Cárie Dentária/prevenção & controle , Cárie Dentária/tratamento farmacológico , Suscetibilidade à Cárie Dentária , Tratamento com Flúor , Fluoretos Tópicos/uso terapêutico , Pais , Satisfação do Paciente , Satisfação Pessoal , Compostos de Amônio Quaternário/uso terapêutico , Compostos de Prata/uso terapêutico , Resultado do Tratamento , LactenteRESUMO
OBJECTIVE: Authors reported multiple definitions of e-oral health and related terms, and used several definitions interchangeably, like mhealth, teledentistry, teleoral medicine and telehealth. The International Association of Dental Research e-Oral Health Network (e-OHN) aimed to establish a consensus on terminology related to digital technologies used in oral healthcare. METHOD: The Crowdsourcing Delphi method used in this study comprised of four main stages. In the first stage, the task force created a list of terms and definitions around digital health technologies based on the literature and established a panel of experts. Inclusion criteria for the panellists were: to be actively involved in either research and/or working in e-oral health fields; and willing to participate in the consensus process. In the second stage, an email-based consultation was organized with the panel of experts to confirm an initial set of terms. In the third stage, consisted of: a) an online meeting where the list of terms was presented and refined; and b) a presentation at the 2022-IADR annual meeting. The fourth stage consisted of two rounds of feedback to solicit experts' opinion about the terminology and group discussion to reach consensus. A Delphi-questionnaire was sent online to all experts to independently assess a) the appropriateness of the terms, and b) the accompanying definitions, and vote on whether they agreed with them. In a second round, each expert received an individualised questionnaire, which presented the expert's own responses from the first round and the panellists' overall response (% agreement/disagreement) to each term. It was decided that 70% or higher agreement among experts on the terms and definitions would represent consensus. RESULTS: The study led to the identification of an initial set of 43 terms. The list of initial terms was refined to a core set of 37 terms. Initially, 34 experts took part in the consensus process about terms and definitions. From them, 27 experts completed the first rounds of consultations, and 15 the final round of consultations. All terms and definitions were confirmed via online voting (i.e., achieving above the agreed 70% threshold), which indicate their agreed recommendation for use in e-oral health research, dental public health, and clinical practice. CONCLUSION: This is the first study in oral health organised to achieve consensus in e-oral health terminology. This terminology is presented as a resource for interested parties. These terms were also conceptualised to suit with the new healthcare ecosystem and the place of e-oral health within it. The universal use of this terminology to label interventions in future research will increase the homogeneity of future studies including systematic reviews.
Assuntos
Ecossistema , Saúde Bucal , Humanos , ConsensoRESUMO
Evidence shows corticomotor plasticity diminishes with age. Nevertheless, whether strength-training, a proven intervention that induces corticomotor plasticity in younger adults, also takes effect in older adults, remains untested. This study examined the effect of a single-session of strength-exercise on corticomotor plasticity in older and younger adults. Thirteen older adults (72.3 ± 6.5 years) and eleven younger adults (29.9 ± 6.9 years), novice to strength-exercise, participated. Strength-exercise involved four sets of 6-8 repetitions of a dumbbell biceps curl at 70-75% of their one-repetition maximum (1-RM). Muscle strength, cortical, corticomotor and spinal excitability, before and up to 60-minutes after the strength-exercise session were assessed. We observed significant changes over time (p < 0.05) and an interaction between time and age group (p < 0.05) indicating a decrease in corticomotor excitability (18% p < 0.05) for older adults at 30- and 60-minutes post strength-exercise and an increase (26% and 40%, all p < 0.05) in younger adults at the same time points. Voluntary activation (VA) declined in older adults immediately post and 60-minutes post strength-exercise (36% and 25%, all p < 0.05). Exercise had no effect on the cortical silent period (cSP) in older adults however, in young adults cSP durations were shorter at both 30- and 60- minute time points (17% 30-minute post and 9% 60-minute post, p < 0.05). There were no differences in short-interval cortical inhibition (SICI) or intracortical facilitation (ICF) between groups. Although the corticomotor responses to strength-exercise were different within groups, overall, the neural responses seem to be independent of age.
Assuntos
Potencial Evocado Motor , Córtex Motor , Força Muscular , Treinamento Resistido , Humanos , Masculino , Feminino , Idoso , Adulto , Treinamento Resistido/métodos , Força Muscular/fisiologia , Córtex Motor/fisiologia , Potencial Evocado Motor/fisiologia , Músculo Esquelético/fisiologia , Estimulação Magnética Transcraniana , Plasticidade Neuronal/fisiologia , Envelhecimento/fisiologia , Fatores Etários , Adulto Jovem , EletromiografiaRESUMO
BACKGROUND/AIM: This study aimed to develop a protocol that combines cone-beam computed tomography (CBCT), software, and 3D printing to design 3D replicas for tooth autotransplantation. The goal was to evaluate the impact of this approach on the extraoral time of the donor teeth and the total surgical time, thereby enhancing surgical efficiency and outcomes. MATERIALS AND METHODS: A non-randomized trial (protocol 10.1186/ISRCTN13563091) was conducted at Riga Stradins University, enrolling 46 patients (13-22 years old) who required molar extraction and possessed a non-erupted third molar. The patients were sequentially assigned to a 3D replica group (24 patients) or a control group (22 patients). The primary outcome measured was the extra-alveolar time of the donor tooth, and the secondary outcome was the total duration of surgery. Both were assessed using a sample size capable of detecting a 10-min difference. A generalized linear model adjusted for various factors was used to test for significant time differences (p < .05) between the groups. RESULTS: Forty-six patients were included in this analysis. The effect of using 3D replicas was not statistically significant and was associated with a decrease in the extraoral time of the donor tooth in seconds (ß = -9.35, 95% CI [-40.86, 22.16]). For the total surgical time in minutes, the use of 3D replicas had a statistically significant impact, reducing the operation duration in minutes (ß = -13.42, 95% CI [-24.50, -2.34]). No early complications were observed in either group, with all teeth present at 3-4 weeks post-surgery. CONCLUSIONS: The integration of 3D printing technology can enhance the efficiency of autotransplantation surgeries, primarily by reducing surgical time.
Assuntos
Dente Serotino , Dente Molar , Adolescente , Adulto , Humanos , Adulto Jovem , Tomografia Computadorizada de Feixe Cônico/métodos , Dente Serotino/cirurgia , Dente Serotino/transplante , Impressão Tridimensional , Software , Transplante Autólogo/métodosRESUMO
Medical image synthesis represents a critical area of research in clinical decision-making, aiming to overcome the challenges associated with acquiring multiple image modalities for an accurate clinical workflow. This approach proves beneficial in estimating an image of a desired modality from a given source modality among the most common medical imaging contrasts, such as Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and Positron Emission Tomography (PET). However, translating between two image modalities presents difficulties due to the complex and non-linear domain mappings. Deep learning-based generative modelling has exhibited superior performance in synthetic image contrast applications compared to conventional image synthesis methods. This survey comprehensively reviews deep learning-based medical imaging translation from 2018 to 2023 on pseudo-CT, synthetic MR, and synthetic PET. We provide an overview of synthetic contrasts in medical imaging and the most frequently employed deep learning networks for medical image synthesis. Additionally, we conduct a detailed analysis of each synthesis method, focusing on their diverse model designs based on input domains and network architectures. We also analyse novel network architectures, ranging from conventional CNNs to the recent Transformer and Diffusion models. This analysis includes comparing loss functions, available datasets and anatomical regions, and image quality assessments and performance in other downstream tasks. Finally, we discuss the challenges and identify solutions within the literature, suggesting possible future directions. We hope that the insights offered in this survey paper will serve as a valuable roadmap for researchers in the field of medical image synthesis.
Assuntos
Aprendizado Profundo , Humanos , Processamento de Imagem Assistida por Computador/métodos , Tomografia Computadorizada por Raios X , Tomografia por Emissão de Pósitrons , Imageamento por Ressonância MagnéticaRESUMO
Background: Child neglect is a public health, human rights, and social problem, with potentially devastating and costly consequences. The aim of this study was to: (1) summarize the oral health profile of children across the globe; (2) provide a brief overview of legal instruments that can offer children protection from dental neglect; and (3) discuss the effectiveness of these legal instruments. Methods: We summarized and highlighted the caries profile and status of implementation of legislation on child dental neglect for 26 countries representing the World Health Organization regions: five countries in Africa (Nigeria, South Africa, Sudan, Tanzania, Zimbabwe), eight in the Americas (Argentina, Brazil, Canada, Chile, Mexico, Peru, Unites States of America, Uruguay), six in the Eastern Mediterranean (Egypt, Iran, Libya, Jordan, Qatar, Saudi Arabia), four in Europe (Italy, Latvia, Serbia, United Kingdom), two in South-East Asia (India and Indonesia) and one country (China) with its special administrative region (Hong Kong) in the Western Pacific. Results: Twenty-five of the 26 countries have legal instruments to address child neglect. Only two (8.0%) of these 25 countries had specific legal instruments on child dental neglect. Although child neglect laws can be interpreted to establish a case of child dental neglect, the latter may be difficult to establish in countries where governments have not addressed barriers that limit children's access to oral healthcare. Where there are specific legal instruments to address child dental neglect, a supportive social ecosystem has also been built to facilitate children's access to oral healthcare. A supportive legal environment, however, does not seem to confer extra protection against risks for untreated dental caries. Conclusions: The institution of specific country-level legislation on child dental neglect may not significantly reduce the national prevalence of untreated caries in children. It, however, increases the prospect for building a social ecosystem that may reduce the risk of untreated caries at the individual level. Social ecosystems to mitigate child dental neglect can be built when there is specific legislation against child dental neglect. It may be more effective to combine public health and human rights-based approaches, inclusive of an efficient criminal justice system to deal with child dental neglect.
RESUMO
The diagnosis of heart failure usually includes a global functional assessment, such as ejection fraction measured by magnetic resonance imaging. However, these metrics have low discriminate power to distinguish different cardiomyopathies, which may not affect the global function of the heart. Quantifying local deformations in the form of cardiac strain can provide helpful information, but it remains a challenge. In this work, we introduce WarpPINN, a physics-informed neural network to perform image registration to obtain local metrics of heart deformation. We apply this method to cine magnetic resonance images to estimate the motion during the cardiac cycle. We inform our neural network of the near-incompressibility of cardiac tissue by penalizing the Jacobian of the deformation field. The loss function has two components: an intensity-based similarity term between the reference and the warped template images, and a regularizer that represents the hyperelastic behavior of the tissue. The architecture of the neural network allows us to easily compute the strain via automatic differentiation to assess cardiac activity. We use Fourier feature mappings to overcome the spectral bias of neural networks, allowing us to capture discontinuities in the strain field. The algorithm is tested on synthetic examples and on a cine SSFP MRI benchmark of 15 healthy volunteers, where it is trained to learn the deformation mapping of each case. We outperform current methodologies in landmark tracking and provide physiological strain estimations in the radial and circumferential directions. WarpPINN provides precise measurements of local cardiac deformations that can be used for a better diagnosis of heart failure and can be used for general image registration tasks. Source code is available at https://github.com/fsahli/WarpPINN.
Assuntos
Insuficiência Cardíaca , Imageamento por Ressonância Magnética , Humanos , Redes Neurais de Computação , Coração/diagnóstico por imagem , FísicaRESUMO
BACKGROUND: We aimed to assess the adherence to five transparency practices (data availability, code availability, protocol registration and conflicts of interest (COI), and funding disclosures) from open access Coronavirus disease 2019 (COVID-19) related articles. METHODS: We searched and exported all open access COVID-19-related articles from PubMed-indexed journals in the Europe PubMed Central database published from January 2020 to June 9, 2022. With a validated and automated tool, we detected transparent practices of three paper types: research articles, randomized controlled trials (RCTs), and reviews. Basic journal- and article-related information were retrieved from the database. We used R for the descriptive analyses. RESULTS: The total number of articles was 258,678, of which we were able to retrieve full texts of 186,157 (72%) articles from the database Over half of the papers (55.7%, n = 103,732) were research articles, 10.9% (n = 20,229) were review articles, and less than one percent (n = 1,202) were RCTs. Approximately nine-tenths of articles (in all three paper types) had a statement to disclose COI. Funding disclosure (83.9%, confidence interval (CI): 81.7-85.8 95%) and protocol registration (53.5%, 95% CI: 50.7-56.3) were more frequent in RCTs than in reviews or research articles. Reviews shared data (2.5%, 95% CI: 2.3-2.8) and code (0.4%, 95% CI: 0.4-0.5) less frequently than RCTs or research articles. Articles published in 2022 had the highest adherence to all five transparency practices. Most of the reviews (62%) and research articles (58%) adhered to two transparency practices, whereas almost half of the RCTs (47%) adhered to three practices. There were journal- and publisher-related differences in all five practices, and articles that did not adhere to transparency practices were more likely published in lowest impact journals and were less likely cited. CONCLUSION: While most articles were freely available and had a COI disclosure, adherence to other transparent practices was far from acceptable. A much stronger commitment to open science practices, particularly to protocol registration, data and code sharing, is needed from all stakeholders.
Assuntos
COVID-19 , Humanos , COVID-19/epidemiologia , Publicações , Conflito de Interesses , Revelação , Europa (Continente)RESUMO
Aim: The COVID-19 pandemic has accelerated teledentistry research with great interest reflected in the increasing number of publications. In many countries, teledentistry programs were established although not much is known about the extent of incorporating teledentistry into practice and healthcare systems. This study aimed to report on policies and strategies related to teledentistry practice as well as barriers and facilitators for this implementation in 19 countries. Methods: Data were presented per country about information and communication technology (ICT) infrastructure, income level, policies for health information system (HIS), eHealth and telemedicine. Researchers were selected based on their previous publications in teledentistry and were invited to report on the situation in their respective countries including Bosnia and Herzegovina, Canada, Chile, China, Egypt, Finland, France, Hong Kong SAR, Iran, Italy, Libya, Mexico, New Zealand, Nigeria, Qatar, Saudi Arabia, South Africa, United Kingdom, Zimbabwe. Results: Ten (52.6%) countries were high income, 11 (57.9%) had eHealth policies, 7 (36.8%) had HIS policies and 5 (26.3%) had telehealth policies. Six (31.6%) countries had policies or strategies for teledentistry and no teledentistry programs were reported in two countries. Teledentistry programs were incorporated into the healthcare systems at national (n = 5), intermediate (provincial) (n = 4) and local (n = 8) levels. These programs were established in three countries, piloted in 5 countries and informal in 9 countries. Conclusion: Despite the growth in teledentistry research during the COVID-19 pandemic, the use of teledentistry in daily clinical practice is still limited in most countries. Few countries have instituted teledentistry programs at national level. Laws, funding schemes and training are needed to support the incorporation of teledentistry into healthcare systems to institutionalize the practice of teledentistry. Mapping teledentistry practices in other countries and extending services to under-covered populations increases the benefit of teledentistry.
RESUMO
PURPOSE: The phase of a MRI signal is used to encode the velocity of blood flow. Phase unwrapping artifacts may appear when aiming to improve the velocity-to-noise ratio (VNR) of the measured velocity field. This study aims to compare various unwrapping algorithms on ground-truth synthetic data generated using computational fluid dynamics (CFD) simulations. METHODS: We compare four different phase unwrapping algorithms on two different synthetic datasets of four-dimensional flow MRI and 26 datasets of 2D PC-MRI acquisitions including the ascending and descending aorta. The synthetic datasets are constructed using CFD simulations of an aorta with a coarctation, with different levels of spatiotemporal resolutions and noise. The error of the unwrapped images was assessed by comparison against the ground truth velocity field in the synthetic data and dual-VENC reconstructions in the in vivo data. RESULTS: Using the unwrapping algorithms, we were able to remove aliased voxels in the data almost entirely, reducing the L2-error compared to the ground truth by 50%-80%. Results indicated that the best choice of algorithm depend on the spatiotemporal resolution and noise level of the dataset. Temporal unwrapping is most successful with a high temporal and low spatial resolution ( δ t = 30 $$ \delta t=30 $$ ms, h = 2 . 5 $$ h=2.5 $$ mm), reducing the L2-error by 70%-85%, while Laplacian unwrapping performs better with a lower temporal or better spatial resolution ( δ t = 60 $$ \delta t=60 $$ ms, h = 1 . 5 $$ h=1.5 $$ mm), especially for signal-to-noise ratio (SNR) 12 as opposed to SNR 15, with an error reduction of 55%-85% compared to the 50%-75% achieved by the Temporal method. The differences in performance between the methods are statistically significant. CONCLUSIONS: The temporal method and spatiotemporal Laplacian method provide the best results, with the spatiotemporal Laplacian being more robust. However, single- V enc $$ {V}_{\mathrm{enc}} $$ methods only situationally and not generally reach the performance of dual- V enc $$ {V}_{\mathrm{enc}} $$ unwrapping methods.
Assuntos
Coartação Aórtica , Imageamento por Ressonância Magnética , Humanos , Imageamento por Ressonância Magnética/métodos , Aorta/diagnóstico por imagem , Razão Sinal-Ruído , Algoritmos , Coartação Aórtica/diagnóstico por imagem , Reprodutibilidade dos Testes , Velocidade do Fluxo Sanguíneo , Imageamento Tridimensional/métodosRESUMO
OBJECTIVE: Federated Learning (FL) enables collaborative training of artificial intelligence (AI) models from multiple data sources without directly sharing data. Due to the large amount of sensitive data in dentistry, FL may be particularly relevant for oral and dental research and applications. This study, for the first time, employed FL for a dental task, automated tooth segmentation on panoramic radiographs. METHODS: We employed a dataset of 4,177 panoramic radiographs collected from nine different centers (n = 143 to n = 1881 per center) across the globe and used FL to train a machine learning model for tooth segmentation. FL performance was compared against Local Learning (LL), i.e., training models on isolated data from each center (assuming data sharing not to be an option). Further, the performance gap to Central Learning (CL), i.e., training on centrally pooled data (based on data sharing agreements) was quantified. Generalizability of models was evaluated on a pooled test dataset from all centers. RESULTS: For 8 out of 9 centers, FL outperformed LL with statistical significance (p<0.05); only the center providing the largest amount of data FL did not have such an advantage. For generalizability, FL outperformed LL across all centers. CL surpassed both FL and LL for performance and generalizability. CONCLUSION: If data pooling (for CL) is not feasible, FL is shown to be a useful alternative to train performant and, more importantly, generalizable deep learning models in dentistry, where data protection barriers are high. CLINICAL SIGNIFICANCE: This study proves the validity and utility of FL in the field of dentistry, which encourages researchers to adopt this method to improve the generalizability of dental AI models and ease their transition to the clinical environment.
Assuntos
Inteligência Artificial , Aprendizado Profundo , Humanos , Radiografia Panorâmica , PesquisadoresRESUMO
OBJECTIVE: To accurately estimate liver PDFF from chemical shift-encoded (CSE) MRI using a deep learning (DL)-based Multi-Decoder Water-Fat separation Network (MDWF-Net), that operates over complex-valued CSE-MR images with only 3 echoes. METHODS: The proposed MDWF-Net and a U-Net model were independently trained using the first 3 echoes of MRI data from 134 subjects, acquired with conventional 6-echoes abdomen protocol at 1.5 T. Resulting models were then evaluated using unseen CSE-MR images obtained from 14 subjects that were acquired with a 3-echoes CSE-MR pulse sequence with a shorter duration compared to the standard protocol. Resulting PDFF maps were qualitatively assessed by two radiologists, and quantitatively assessed at two corresponding liver ROIs, using Bland Altman and regression analysis for mean values, and ANOVA testing for standard deviation (STD) (significance level: .05). A 6-echo graph cut was considered ground truth. RESULTS: Assessment of radiologists demonstrated that, unlike U-Net, MDWF-Net had a similar quality to the ground truth, despite it considered half of the information. Regarding PDFF mean values at ROIs, MDWF-Net showed a better agreement with ground truth (regression slope = 0.94, R2 = 0.97) than U-Net (regression slope = 0.86, R2 = 0.93). Moreover, ANOVA post hoc analysis of STDs showed a statistical difference between graph cuts and U-Net (p < .05), unlike MDWF-Net (p = .53). CONCLUSION: MDWF-Net showed a liver PDFF accuracy comparable to the reference graph cut method, using only 3 echoes and thus allowing a reduction in the acquisition times. CLINICAL RELEVANCE STATEMENT: We have prospectively validated that the use of a multi-decoder convolutional neural network to estimate liver proton density fat fraction allows a significant reduction in MR scan time by reducing the number of echoes required by 50%. KEY POINTS: ⢠Novel water-fat separation neural network allows for liver PDFF estimation by using multi-echo MR images with a reduced number of echoes. ⢠Prospective single-center validation demonstrated that echo reduction leads to a significant shortening of the scan time, compared to standard 6-echo acquisition. ⢠Qualitative and quantitative performance of the proposed method showed no significant differences in PDFF estimation with respect to the reference technique.
Assuntos
Fígado , Água , Humanos , Estudos Prospectivos , Fígado/diagnóstico por imagem , Imageamento por Ressonância Magnética/métodos , Abdome , Redes Neurais de Computação , Reprodutibilidade dos TestesRESUMO
Aortic stenosis is a condition which is fatal if left untreated. Novel quantitative imaging techniques which better characterise transvalvular pressure drops are being developed but require refinement and validation. A customisable and cost-effective workbench valve phantom circuit capable of replicating valve mechanics and pathology was created. The reproducibility and relationship of differing haemodynamic metrics were assessed from ground truth pressure data alongside imaging compatibility. The phantom met the requirements to capture ground truth pressure data alongside ultrasound and magnetic resonance image compatibility. The reproducibility was successfully tested. The robustness of three different pressure drop metrics was assessed: whilst the peak and net pressure drops provide a robust assessment of the stenotic burden in our phantom, the peak-to-peak pressure drop is a metric that is confounded by non-valvular factors such as wave reflection. The peak-to-peak pressure drop is a metric that should be reconsidered in clinical practice. The left panel shows manufacture of low cost, functional valves. The central section demonstrates circuit layout, representative MRI and US images alongside gross valve morphologies. The right panel shows the different pressure drop metrics that were assessed for reproducibility.