Better Reporting of Studies on Artificial Intelligence: CONSORT-AI and Beyond

Schwendicke, Falk; Krois, Joachim

doi:10.1177/0022034521998337

Cited by 22 publications

(12 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From a surveillance point of view, the abundance of new strategies and tools seems to guide us towards a possible combination of technological strategies that will be very powerful in the preventive diagnosis of periodontitis. Combined with self-report questionnaires that are applicable in any setting (i.e., at any sort of health service or even at home), we will be able to create tree health and risk assessments along with 1.0 and 2.0 software tools [ 31 , 32 ]. We foresee that a set of interactions with patients may be carried out prior to any clinical diagnosis, based only on self-report and complementary diagnostic tests (radiographs, blood and/or salivary analyses), by using artificial intelligence empowered strategies [ 33 ].…”

Section: Discussionmentioning

confidence: 99%

Self-Reported Measures of Periodontitis in a Portuguese Population: A Validation Study

Machado

Lyra²,

Santos³

et al. 2022

JPM

View full text Add to dashboard Cite

Self-reported questionnaires have been developed and validated in multiple populations as useful tools to estimate the prevalence of periodontitis in epidemiological settings. This study aimed to explore the accuracy of self-reporting for predicting the prevalence of periodontitis in a Portuguese population. The questionnaires were given to patients at a university clinic. Thirteen self-reported questions on periodontal health were gathered in a patient-reported questionnaire. Then, self-reporting responses were validated using full-mouth periodontal examination as a comparison. Multivariable logistic regression was used to analyze sensitivity, specificity, accuracy, precision, and area under the curve-receiver operator characteristic (AUC-ROC). Self-reported answers from 103 participants (58 females and 45 males) were included. Self-reported gum health, loose teeth, tooth appearance, and use of dental floss were associated with different definitions of severe periodontitis. The self-reported questions on “having gum disease,” combined with “having gum treatment” and “having lost bone” were the items with higher performance for the 2018 case definition and the 2012 case definition, as well as for each respective severity staging. Categorization of tooth loss was only valuable for the prediction of periodontitis cases according to the 2012 case definition and its severe stage. Multiple self-reporting set-ups showed elevated performance levels for predicting periodontitis in Portuguese patients. These results may pave the way for future epidemiological surveillance programs using self-reporting approaches.

show abstract

Section: Discussionmentioning

confidence: 99%

Self-Reported Measures of Periodontitis in a Portuguese Population: A Validation Study

Machado

Lyra²,

Santos³

et al. 2022

JPM

View full text Add to dashboard Cite

show abstract

“…To achieve these goals, researchers should also increase the overall quality of their research protocols, providing a comprehensive report on the methods implemented and outlining the risk of bias. New efforts to ensure the development, transparency, replicability, and ethics of AI's research environment should be embodied in new guidelines for evaluating and presenting this type of research [Wiens and Shenoy, 2018;Schwendicke and Krois, 2021]. Training will be required for the use of resources and automated systems to carry out these tasks incorporating ML, which should be guided by multidisciplinary teams, where practitioners, radiologists, epidemiologists, statisticians, and public policymakers, among others, must collaborate.…”

Section: Implications For Research and Clinical Practicementioning

confidence: 99%

Machine Learning in the Diagnosis and Prognostic Prediction of Dental Caries: A Systematic Review

et al. 2022

View full text Add to dashboard Cite

We performed a systematic review to evaluate the success of machine learning algorithms in the diagnosis and prognostic prediction of dental caries. The review protocol was a priori registered in the PROSPERO, CRD42020183447. The search involved electronic bibliographic databases: PubMed/Medline, Scopus, EMBASE, and Web of Science, and grey literature until December 2020. We excluded review articles, case series, case reports, editorials, letters, comments, educational methodologies or assessments of robotic devices and articles with less than 10 participants or specimens. Two independent reviewers selected the studies and performed the assessment of the methodological quality based on standardized scales. We summarize data on the machine learning algorithms used, software, performance outcomes such as accuracy/precision sensitivity/recall, specificity, area under the receiver operating characteristic curve (AUC), and positive/negative predictive values related to dental caries. Meta-analyses were not performed due to methodological differences. Our review included 15 studies (10 diagnostic studies and 5 prognostic prediction studies). Cross-sectional design studies were predominant (12). The most frequently used statistical measure of performance reported in diagnostic studies was AUC value, which ranged from 0.745 to 0.987. For most diagnostic studies, data from contingency tables were not available. Reported sensitivities were higher in low risk of bias prognostic prediction studies (median [IQR] of 0.996 [0.971-1.000] versus unclear/high risk of bias studies 0.189 [0-0.340]; p-value 0.025). While there were no significant differences in the specificity between these subgroups. We concluded that the use of these technologies for the diagnosis and prognostic prediction of dental caries, although promising, is at an early stage. The general applicability of the evidence was limited given that most models were developed outside the real clinical setting with a prevalence of unclear/high risk of bias. Researchers must increase the overall quality of their research protocols by providing a comprehensive report on the methods implemented.

show abstract

“…By taking reference to the natural dentition, the prostheses in questions should be able to retain the original functionality and to have no interference to the jaw movements. It is hypothesized that using a 3D Generative Adversarial Network (GAN) [ 20 , 21 ], AI can learn the morphology and positional relationship between teeth in an individual subject and automate the design of biomimetic single-tooth dental prostheses from the features of the remaining dentition after sufficient trainings.…”

Section: Introductionmentioning

confidence: 99%

Artificial intelligence-designed single molar dental prostheses: A protocol of prospective experimental study

et al. 2022

View full text Add to dashboard Cite

Background Dental prostheses, which aim to replace missing teeth and to restore patients’ appearance and oral functions, should be biomimetic and thus adopt the occlusal morphology and three-dimensional (3D) position of healthy natural teeth. Since the teeth of an individual subject are controlled by the same set of genes (genotype) and are exposed to mostly identical oral environment (phenotype), the occlusal morphology and 3D position of teeth of an individual patient are inter-related. It is hypothesized that artificial intelligence (AI) can automate the design of single-tooth dental prostheses after learning the features of the remaining dentition. Materials and methods This article describes the protocol of a prospective experimental study, which aims to train and to validate the AI system for design of single molar dental prostheses. Maxillary and mandibular dentate teeth models will be collected and digitized from at least 250 volunteers. The (original) digitized maxillary teeth models will be duplicated and processed by removal of right maxillary first molars (FDI tooth 16). Teeth models will be randomly divided into training and validation sets. At least 200 training sets of the original and the processed digitalized teeth models will be input into 3D Generative Adversarial Network (GAN) for training. Among the validation sets, tooth 16 will be generated by AI on 50 processed models and the morphology and 3D position of AI-generated tooth will be compared to that of the natural tooth in the original maxillary teeth model. The use of different GAN algorithms and the need of antagonist mandibular teeth model will be investigated. Results will be reported following the CONSORT-AI.

show abstract

Better Reporting of Studies on Artificial Intelligence: CONSORT-AI and Beyond

Cited by 22 publications

References 27 publications

Self-Reported Measures of Periodontitis in a Portuguese Population: A Validation Study

Self-Reported Measures of Periodontitis in a Portuguese Population: A Validation Study

Machine Learning in the Diagnosis and Prognostic Prediction of Dental Caries: A Systematic Review

Artificial intelligence-designed single molar dental prostheses: A protocol of prospective experimental study

Contact Info

Product

Resources

About