Background and purpose: In comparison to conventional medical imaging diagnostic modalities, the aim of this overview article is to analyze the accuracy of the application of Artificial Intelligence (AI) techniques in the identification and diagnosis of malignant tumors in adult patients.Data sources: The acronym PIRDs was used and a comprehensive literature search was conducted on PubMed, Cochrane, Scopus, Web of Science, LILACS, Embase, Scielo, EBSCOhost, and grey literature through Proquest, Google Scholar, and JSTOR for systematic reviews of AI as a diagnostic model and/or detection tool for any cancer type in adult patients, compared to the traditional diagnostic radiographic imaging model. There were no limits on publishing status, publication time, or language. For study selection and risk of bias evaluation, pairs of reviewers worked separately.Results: In total, 382 records were retrieved in the databases, 364 after removing duplicates, 32 satisfied the full-text reading criterion, and 09 papers were considered for qualitative synthesis. Although there was heterogeneity in terms of methodological aspects, patient differences, and techniques used, the studies found that several AI approaches are promising in terms of specificity, sensitivity, and diagnostic accuracy in the detection and diagnosis of malignant tumors. When compared to other machine learning algorithms, the Super Vector Machine method performed better in cancer detection and diagnosis. Computer-assisted detection (CAD) has shown promising in terms of aiding cancer detection, when compared to the traditional method of diagnosis.Conclusions: The detection and diagnosis of malignant tumors with the help of AI seems to be feasible and accurate with the use of different technologies, such as CAD systems, deep and machine learning algorithms and radiomic analysis when compared with the traditional model, although these technologies are not capable of to replace the professional radiologist in the analysis of medical images. Although there are limitations regarding the generalization for all types of cancer, these AI tools might aid professionals, serving as an auxiliary and teaching tool, especially for less trained professionals. Therefore, further longitudinal studies with a longer follow-up duration are required for a better understanding of the clinical application of these artificial intelligence systems.Trial registration: Systematic review registration.Prospero registration number:CRD42022307403.

The aim of this systematic review with meta-analysis was to analyze the effects of intra-pregnancy nonsurgical periodontal therapy on periodontal inflammatory biomarkers and adverse pregnancy outcomes. On June 5, 2017, we searched PubMed, Cochrane, SCOPUS, Web of Science, LILACS, ProQuest, Open Grey, and Google Scholar databases. Randomized clinical trials in which pregnant women with chronic periodontitis underwent nonsurgical periodontal therapy, compared with an untreated group, tested for inflammatory biomarkers, and followed till delivery were included. Primary outcomes were preterm birth, low birth weight, and preeclampsia. Meta-analysis was performed with 5.3.5 version of Review Manager software. We found 565 references in the databases, 326 after duplicates removal, 28 met criteria for full text reading, and 4 met eligibility criteria for quantitative and qualitative synthesis. Intra-pregnancy nonsurgical periodontal therapy improved periodontal clinical parameters (periodontal pocket depth, clinical attachment level, and bleeding on probing) and reduced biomarker level from gingival crevicular fluid (GCF), and some from blood serum; however, it did not influence biomarker level from umbilical cord blood. Meta-analysis showed tendency for reduction of the risk of preterm birth before 37 weeks for treated group (risk ratio (RR) = 0.54, 95% CI 0.38–0.77; p = 0.0007; inconsistency indexes (I2) 32%) but did not show any difference for low birth weight occurrence (RR = 0.78, 95%CI 0.50–1.21; p = 0.27; I2 41%). No included study considered preeclampsia as a gestational outcome. These results demonstrated that the intra-pregnancy nonsurgical periodontal therapy decreased periodontal inflammatory biomarker levels from gingival crevicular fluid and some from serum blood, with no influence on inflammatory biomarker level from cord blood, and it did not consistently reduce adverse gestational adverse outcome occurrence. PROSPERO CRD42015027750

: There is emerging evidence that several oral diseases and conditions can be associated with DM. Dental caries, hyposalivation, fungal diseases and endodontics lesions may represent potential oral complications that can be aggravated by chronic hyperglycemia. Individuals with DM have a low perception of oral diseases which can lead to clinically important oral and systemic complications. This review aims to provide data on the most common oral signs and symptoms related to DM and to explore the mechanisms that might explain associations between DM and oral diseases in order to clarify the risks posed by poor oral health in DM. Since the linkage between oral diseases and DM is part of multifactorial aspects related to chronic hyperglycemic status and several common conditions affecting the whole body, both require rigorous self-control from patients and attention from medical and dental professionals.