NTSM: a non-salient target segmentation model for oral mucosal diseases.

BACKGROUND: Oral mucosal diseases are similar to the surrounding normal tissues, i.e., their many non-salient features, which poses a challenge for accurate segmentation lesions. Additionally, high-precision large models generate too many parameters, which puts pressure on storage and makes it difficult to deploy on portable devices. METHODS: To address these issues, we design a non-salient target segmentation model (NTSM) to improve segmentation performance while reducing the number of parameters. The NTSM includes a difference association (DA) module and multiple feature hierarchy pyramid attention (FHPA) modules. The DA module enhances feature differences at different levels to learn local context information and extend the segmentation mask to potentially similar areas. It also learns logical semantic relationship information through different receptive fields to determine the actual lesions and further elevates the segmentation performance of non-salient lesions. The FHPA module extracts pathological information from different views by performing the hadamard product attention (HPA) operation on input features, which reduces the number of parameters. RESULTS: The experimental results on the oral mucosal diseases (OMD) dataset and international skin imaging collaboration (ISIC) dataset demonstrate that our model outperforms existing state-of-the-art methods. Compared with the nnU-Net backbone, our model has 43.20% fewer parameters while still achieving a 3.14% increase in the Dice score. CONCLUSIONS: Our model has high segmentation accuracy on non-salient areas of oral mucosal diseases and can effectively reduce resource consumption.

Potential salivary and serum biomarkers for burning mouth syndrome and their relationship with anxiety/depression.

Background/purpose: The pathophysiology of burning mouth syndrome (BMS), although considered a multifactorial etiology including psychological factors, is still not well understood. Hence, this study aimed to investigate the potential usage of salivary and serum biomarkers, including brain-derived neurotrophic factor (BDNF), interferon-gamma (IFN-γ), interleukin-1beta (IL-1ß), interleukin-6 (IL-6), interleukin-8 (IL-8), and tumor necrosis factor-alpha (TNF-α), in diagnosing BMS and their correlations with anxiety/depression. Materials and methods: 45 BMS patients and 14 healthy volunteers were enrolled. The patients were divided into BMS with anxiety/depression group and BMS without anxiety/depression group according to the scores of the Zung Self-rating Anxiety Scale (SAS) and Zung Self-rating Depression Scale (SDS). Additionally, concentrations of BDNF, IFN-γ, IL-1ß, IL-6, IL-8, and TNF-α in saliva and those in serum among the patients and healthy volunteers were assessed by multiplex assay using Luminex 200TM system and Enzyme-linked immunosorbent assay (ELISA), respectively. Results: Among all the serum biomarkers, only BDNF showed a statistically significant decrease in the patients than the healthy volunteers (P < 0.05). Regarding saliva biomarkers, BDNF, IL-1ß, and IL-8 all exhibited a statistically significant increase in all the BMS patients versus the healthy volunteers (P < 0.05) but only BDNF was significantly different between patients with anxiety/depression and healthy individuals when considering anxiety/depression. Among BMS patients with anxiety/depression, saliva TNF-α had positive associations with other biomarkers including BDNF, IFN-γ, IL-1ß, IL-6, and IL-8 (P < 0.05). Conclusion: The increased concentration of saliva BDNF holds strong potential for diagnosing BMS and the elevated level of saliva TNF-α is crucial in identifying BMS patients with anxiety/depression.

Evaluating photodynamic therapy for oral precancerous lesions: Highlighting outcome measure of malignant transformation.

The main outcome measure assessed in previous studies on photodynamic therapy (PDT) for oral precancerous lesions (OPL) is clinical response based on the alteration in lesion size after treatment. However, the primary and secondary outcome measures of the interventions for OPL should be malignant transformation and recurrence. Thus, the objective of this short communication is to summarize the evidence on PDT in preventing the recurrence and malignant transformation of OPL. There were 16 eligible studies which addressed the issue of OPL patients who received PDT with recurrence outcome, and the pooled recurrence rate (95% confidence interval) was analyzed to be 20.1% (16.2-24.6%). Notably, only 1 study reported that 7.5% of malignant transformation rate for OPL received PDT. These should be interpreted with caution due to low-level evidence, such as differences in study design, clinical and pathological features of patients enrolled, limited sample size, short follow-up time. Given few evaluated the effect of PDT on malignant transformation, we highlight that this primary outcome measure of OPL needs to be investigated in further well-designed longitudinal studies with adequate follow-up periods.

Oral mucosal disease recognition based on dynamic self-attention and feature discriminant loss.

OBJECTIVES: To develop a dynamic self-attention and feature discrimination loss function (DSDF) model for identifying oral mucosal diseases presented to solve the problems of data imbalance, complex image background, and high similarity and difference of visual characteristics among different types of lesion areas. METHODS: In DSDF, dynamic self-attention network can fully mine the context information between adjacent areas, improve the visual representation of the network, and promote the network model to learn and locate the image area of interest. Then, the feature discrimination loss function is used to constrain the diversity of channel characteristics, so as to enhance the feature discrimination ability of local similar areas. RESULTS: The experimental results show that the recognition accuracy of the proposed method for oral mucosal disease is the highest at 91.16%, and is about 6% ahead of other advanced methods. In addition, DSDF has recall of 90.87% and F1 of 90.60%. CONCLUSIONS: Convolutional neural networks can effectively capture the visual features of the oral mucosal disease lesions, and the distinguished visual features of different oral lesions can be extracted better using dynamic self-attention and feature discrimination loss function, which is conducive to the auxiliary diagnosis of oral mucosal diseases.

Palatal perforation caused by Alternaria alternata infection in an immunocompetent adolescent.

Opportunistic oral mucosal fungal infection caused by Alternaria alternata is extremely rare. Herein, we present a rare palatal perforation as a result of oral infection caused by A. alternata in an immunocompetent adolescent. An 18-year-old boy, who had previously been healthy, was admitted to our institution with persistent pain in the palate for the past 12 months. Upon impression of palatal bone resorption based on computed tomography imaging and chronic granulomatous inflammation based on biopsy (hematoxylin-eosin staining), the patient was examined for commonly relevant causes such as potential tumor and Mycobacterium tuberculosis infection. All test results were inconclusive. After a thorough diagnostic investigation, an unusual fungal infection, A. alternata infection, was confirmed by next-generation sequencing and biopsy (periodic acid-Schiff staining and immunofluorescence staining). The patient underwent surgical debridement and was subjected to voriconazole treatment postoperatively for over a period of 5 months. Thus, these findings highlight the importance of considering A. alternata as a potential pathogenic factor in an etiological palatal perforation.

Association of polymorphisms in Th1/Th2-related cytokines (IFN-γ, TGFß1, IL-1ß, IL-2, IL-4, IL-18) with oral lichen planus: A pooled analysis of case-control studies.

Background/purpose: Increasing evidence suggests that single-nucleotide polymorphisms (SNPs) in Th1/Th2-related cytokine genes correlated with oral lichen planus (OLP) susceptibility. However, these results were inconsistent and inconclusive. Hence, the aim of this study is to draw a more precise estimation of the genetic associations between SNPs in 6 cytokines (IFN-γ, IL-18, TGFß1, IL-1ß, IL-2, IL-4) and OLP. Materials and methods: A systematic literature search was conducted to identify all eligible case-control studies on the association between SNPs in 6 cytokines and OLP susceptibility. Odds ratios (ORs) and 95% confidence intervals (CIs) from each study were pooled to estimate the strength of the association. Results: A significant association of IFN-γ (874A/T) polymorphism with OLP was found (OR, 1.49; 95%CI, 1.22-1.81; P < 0.001) based on 6 eligible studies. A significant association of IL-18 (137G/C) polymorphism with OLP was found (OR, 1.64; 95%CI, 1.24-2.18; P < 0.001) based on 3 studies. A marginally significant association of TGFß1 (509C/T) polymorphism in allele model with OLP was found (OR, 1.31; 95%CI, 1.01-1.71; P = 0.05) based on 4 studies. Nevertheless, lack of significant association of IL-1ß (3954C/T), IL-2 (330T/G), IL-4 (590C/T), and IL-18 (607C/A) polymorphisms with OLP was found (P > 0.05) based on 3 studies, respectively. Conclusion: This is the first meta-analysis to investigate the associations of 6 cytokines polymorphisms with OLP, suggesting that SNPs in IFN-γ, IL-18, and TGFß1 may act as genetic factors for OLP risk. Further well-designed studies with larger sample size and multiple ethnicities are needed to validate these associations.

A potential reciprocal emergence of regulatory T cells in oral lichen planus and pemphigus vulgaris: A meta-analysis.

Oral lichen planus (OLP) and pemphigus vulgaris (PV) are 2 common inflammatory mucocutaneous diseases of immune-based etiology. Evidence indicates regulatory T (Treg) cells maybe play a role in the pathogenesis of OLP and PV, which are caused by aberrant immune responses. In this report, 7 eligible case-control studies containing 517 OLP patients and 261 healthy controls (HC) were identified. The level of Tregs was significantly higher in OLP patients than HC (mean difference: 1.79; 95%CI: 0.99, 2.60). On the other hand, 7 eligible case-control studies containing 169 PV patients and 121 HC were identified. Conversely, the level of Tregs was significantly lower in PV patients than HC (mean difference: -2.49; 95%CI: -3.90, -1.08). Collectively, this analysis for the first time reported reciprocal emergence of Tregs in OLP and PV using meta-analysis, and provided an interesting insight into a previously undescribed the linkage of the 2 mucocutaneous diseases.

Association of salivary miRNAs with onset and progression of oral potentially malignant disorders: Searching for noninvasive biomarkers.

Background/purpose: There is an urgent need for noninvasive biomarkers to diagnose oral potentially malignant disorders (OPMD). A wide range of over 20 miRNAs in saliva of OPMD patients have been investigated in different studies. Yet, which of the ones provide a better power of discrimination for the diagnosis of OPMD onset and progression are uncertain. Materials and methods: A total of 17 eligible studies including 426 cases of OPMD and 486 control subjects (352 normal mucosa and 134 oral squamous cell carcinoma) were summarized. Results: The bubble chart analysis showed that the most power salivary miRNA associated with OPMD onset was miR-21, followed by miR-31 and miR-142; the better power miRNAs associated with recurrence and malignant progression of OPMD were miR-31, miR-21, and miR-184. Conclusion: Salivary miRNAs, especially miR-21 and miR-31, were associated with onset and progression of OPMD, and could then serve as noninvasive biomarkers for screening OPMD and detecting malignant changes.

Dysregulation and implications of lncRNAs and miRNAs in oral tongue squamous cell carcinoma: In reply with emphasis on the role of ceRNAs.

Increasing evidence indicates that long noncoding RNAs (lncRNAs) as competing endogenous RNAs (ceRNAs) competitively sequestering microRNAs (miRNAs) participate in biological processes of oral tongue squamous cell carcinoma (TSCC). In this Letter, the ceRNA regulatory networks consisting of lncRNA/miRNA/mRNA axes in TSCC were summarized. Dysregulated profiles containing 33 lncRNAs and 31 miRNAs were identified by cancer-associated phenotypes verification. Almost all the lncRNAs could exert the oncogenic roles to sponge miRNAs and regulate targeting mRNA expression, thereby modulating cell proliferation, cell cycle, apoptosis, invasion, migration, metastasis, epithelial-mesenchymal transition, as well as chemoresistance. Significantly, the implications of functional ceRNAs deactivated in tumor cells contribute to the exploitation of novel diagnostic and therapeutic strategies for TSCC.