Temporomandibular disorders (TMD), a common condition in oral and maxillofacial surgery, significantly impairs patients' quality of life. Early prediction and appropriate treatment of TMD are therefore critically important. Research on TMD prediction models has evolved from traditional statistical methods to machine learning and subsequently to deep learning, each offering unique advantages and limitations. Traditional statistical methods can effectively identify independent risk factors influencing treatment outcomes but generally rely on substantial prior knowledge and assumptions. Machine learning techniques can process large-scale, high-dimensional data and autonomously learning patterns and regularities within datasets. However, they exhibit strong dependence on data quality and limited model generalization capabilities. Deep learning approaches excel at automatically extracting temporal patterns and trends from time-series data while effectively capturing complex nonlinear relationships, yet they require extensive training datasets and suffer from interpretability challenges due to their inherent black-box testing. This review comprehensively evaluates the implementation and performance of these computational approaches in TMD prediction, critically analyzes their respective strengths and constraints, and discusses promising future research directions.

Temporomandibular disorders (TMD), a common condition in oral and maxillofacial surgery, significantly impairs patients' quality of life. Early prediction and appropriate treatment of TMD are therefore critically important. Research on TMD prediction models has evolved from traditional statistical methods to machine learning and subsequently to deep learning, each offering unique advantages and limitations. Traditional statistical methods can effectively identify independent risk factors influencing treatment outcomes but generally rely on substantial prior knowledge and assumptions. Machine learning techniques can process large-scale, high-dimensional data and autonomously learning patterns and regularities within datasets. However, they exhibit strong dependence on data quality and limited model generalization capabilities. Deep learning approaches excel at automatically extracting temporal patterns and trends from time-series data while effectively capturing complex nonlinear relationships, yet they require extensive training datasets and suffer from interpretability challenges due to their inherent black-box testing. This review comprehensively evaluates the implementation and performance of these computational approaches in TMD prediction, critically analyzes their respective strengths and constraints, and discusses promising future research directions.

Objective:To explore the relationship between pathogens in the olfactory cleft area and olfactory disorders in patients with upper respiratory inflammation (URI) during the prevention and control of 2019 novel coronavirus disease (COVID-19).Methods:A total of 234 URI patients including acute upper respiratory infection, chronic rhinosinusitis (CRS), allergic rhinitis (AR) were continuously selected from September 2020 to March 2021 in Beijing Anzhen Hospital and 98 healthy adults were enrolled as controls. The secretions from the olfactory cleft of all subjects were collected with nasal swabs under nasal endoscopy. Multiple real-time fluorescent quantitative polymerase chain reaction detection method was used to detect nucleic acids of 33 types of respiratory pathogenic microorganism. Sniffin′ Sticks olfactory test was performed on all patients with URI. URI patients with olfactory dysfunction were followed up for 9 (8, 10) months ( M ( Q1, Q3)). SPSS 20.0 software was used for statistical analysis. Results:Among the 98 controls, 9 (9.18%) were positive for pathogenic microorganisms, including 1 (1.02%) rhinovirus, 1 (1.02%) parainfluenza virus type 3, 3 (3.06%) enterovirus, 1 (1.02%) staphylococcus aureus and 3 (3.06%) Moraxella catarrhalis. Among the 234 URI patients, 111 (47.44%) had olfactory disorders and 123 (52.56%) had normal sense of smell. In the olfactory disorder group (111 cases), 38 cases (34.23%) were positive for pathogenic microorganisms, and 4 cases (3.60%) were mixed infection, including 11 cases of rhinovirus (9.91%), 5 cases of coronavirus 229E (4.50%), 2 cases of coronavirus OC43/NL63 (1.80%), 3 cases of parainfluenza virus type 1 (2.70%), 2 cases of enterovirus (1.80%), 1 case of influenza B virus type BV (0.90%), 11 cases of Staphylococcus aureus (9.91%), 7 cases of Moraxella catarrhalis (6.31%), and 1 case of Klebsiella pneumoniae (0.90%). In the normal smell group (123 cases), 18 cases (14.63%) were positive for pathogenic microorganisms, and 1 case (0.81%) was mixed infection, including 3 cases of rhinovirus (2.44%), 4 cases of coronavirus 229E (3.25%), 1 case of Influenza virus type 3 (0.81%), 3 cases of enterovirus (2.44%), 3 cases of Staphylococcus aureus (2.44%), 4 cases of Moraxella catarrhalis (3.25%), and 1 case of Klebsiella pneumoniae (0.81%). Univariate analysis between the two groups found that there were significant differences in the detection rate of pathogenic microorganisms, rhinovirus and Staphylococcus aureus between the groups (all P<0.05). The detection rate of parainfluenza virus type 1, Staphylococcus aureus, and rhinovirus were different between the patients with olfactory disorder and normal olfactory function in the three subgroups of acute upper respiratory tract infection, CRS and AR, respectively (χ 2 value was 3.88, 4.53 and 4.73, respectively, all P<0.05). During the follow-up period, among the 111 patients with olfactory disorder, 71 (63.96%) patients′ olfactory function returned to normal, 32 (28.83%) patients′ olfactory function improved but not completely returned to normal, 8 (7.21%) patients′ olfactory function did not improve. Conclusions:During the prevention and control of COVID-19, rhinovirus or Staphylococcus aureus infection or colonization of URI patients is closely related to olfactory disorders. Parainfluenza virus type 1 infection can cause relatively persistent olfactory disorders in patients with acute upper respiratory tract infection. Staphylococcus aureus and rhinovirus colonization are related to the occurrence of olfactory dysfunction in CRS and AR patients respectively.

Objective: Functional near infrared spectroscopy (fNIRS) was used to assess brain oxygenated hemoglobin (Oxy Hb) activation in college students with different sleep quality under the verbal fluency task (VFT), so as to better provide a theoretical basis for the neural mechanism for sleep quality improvement of college students. Methods: A simple random sampling method was used to investigate 96 college students from one university during 2020 and 2021. According to the results of the Pittsburgh Sleep Quality Index(PSQI), participants were divided into 3 groups: good sleep quality group( n =45), moderate group( n =33), and poor group( n =18). The 53 channel near infrared spectroscopy to collect cerebral blood oxygen signals under the VFT task. Association between oxygenated hemoglobin with sleep quality was analyzed. Results: About 18.75% of college students reported sleep quality problems, including long sleep latency (0.97±0.97) and poor subjective sleep quality (0.96±0.72). There was a significant negative correlation between PSQI score and average oxygenated hemoglobin (Avg HbO) index of dorsolateral prefrontal lobe ( r =-0.23, P =0.03). The Avg HbO index differed significantly between good and poor sleep quality groups on dorsolateral prefrontal lobe( P =0.05). Conclusion This study verified that there is a positive correlation between sleep quality and cognitive ability among college students. The fNIRS technique could accurately collect blood oxygen signals from dorsolateral prefrontal lobe during cognitive tasks, which proves to be an effective tool for identifying sleep quality of college students.

AIM: To study the polymorphism distribution of methylenetetrahydrofolate reductase (MTHFR) and methionine synthase reductase (MTRR) genes and their influence on serum homocysteine (Hcy) concentration. METHODS: A total of 148 patients diagnosed with ischemic stroke from November 2020 to February 2021 in Yijishan Hospital of Wanan Medical College were selected for the study, and patients were typed for MTHFR 677C/T and MTRR 66A/G genes using fluorescent staining in situ hybridization technique. Serum Hcy concentrations were measured in 21 patients using a circulating enzyme assay. The distribution of MTHFR 677C/T and MTRR 66A/G gene polymorphisms were analyzed, and the differences in serum Hcy concentrations between patients with different genotypes were compared. RESULTS: The mutation rates of MTHFR 677C/T and MTRR 66A/G genes were 42.57% and 26.01%, respectively, and no significant differences in gene distribution frequencies were observed between men and women (P>0.05). The mean Hcy serum concentration was (16.04±4.34) μmol/L in 21 patients, including 8 patients (38.10%) with <15 μmol/L and 13 patients (61.90%) with ≥15 μmol/L. The Hcy serum concentrations in patients with different genotypes of MTHFR were TT (18.91±5.34) μmol/L, CT (14.38±1.84) μmol/L and CC (13.58±2.86) μmol/L, respectively, and were statistically different (P<0.001). Serum Hcy concentrations in patients with different genotypes of MTRR were not statistically different (P>0.05). CONCLUSION: MTHFR gene polymorphisms can affect serum Hcy concentrations. The MTHFR genotyping can be considered for individualized folic acid supplement. This conclusion should be further verified by expanding the clinical sample size.

Paget's disease of bone (PDB) is a common metabolic bone disease that is characterized by aberrant focal bone remodeling, which is caused by excessive osteoclastic bone resorption followed by disorganized osteoblastic bone formation. Genetic factors are a critical determinant of PDB pathogenesis, and several susceptibility genes and loci have been reported, including SQSTM1, TNFSF11A, TNFRSF11B, VCP, OPTN, CSF1 and DCSTAMP. Herein, we report a case of Chinese familial PDB without mutations in known genes and identify a novel c.163G>C (p.Val55Leu) mutation in FKBP5 (encodes FK506-binding protein 51, FKBP51) associated with PDB using whole-exome sequencing. Mutant FKBP51 enhanced the Akt phosphorylation and kinase activity in cells. A study of osteoclast function using FKBP51V55L KI transgenic mice proved that osteoclast precursors from FKBP51V55L mice were hyperresponsive to RANKL, and osteoclasts derived from FKBP51V55L mice displayed more intensive bone resorbing activity than did FKBP51WT controls. The osteoclast-specific molecules tartrate-resistant acid phosphatase, osteoclast-associated receptor and transcription factor NFATC1 were increased in bone marrow-derived monocyte/macrophage cells (BMMs) from FKBP51V55L mice during osteoclast differentiation. However, c-fos expression showed no significant difference in the wild-type and mutant groups. Akt phosphorylation in FKBP51V55L BMMs was elevated in response to RANKL. In contrast, IκB degradation, ERK phosphorylation and LC3II expression showed no difference in wild-type and mutant BMMs. Micro-CT analysis revealed an intensive trabecular bone resorption pattern in FKBP51V55L mice, and suspicious osteolytic bone lesions were noted in three-dimensional reconstruction of distal femurs from mutant mice. These results demonstrate that the mutant FKBP51V55L promotes osteoclastogenesis and function, which could subsequently participate in PDB development.
Acid Phosphatase ; Animals ; Asian Continental Ancestry Group ; Bone Diseases, Metabolic ; Bone Remodeling ; Bone Resorption ; Femur ; Humans ; Mice ; Mice, Transgenic ; Osteitis Deformans* ; Osteoblasts ; Osteoclasts ; Osteogenesis ; Phosphorylation ; Phosphotransferases ; Tacrolimus Binding Proteins ; Transcription Factors

Objective To analyze the correlation of real-time fluorescent quantitation PCR(FQ-PCR)for detecting HCV-RNA loading and the chemiluminescence immunoassay(CLIA)for detecting anti-HCV antibody.Methods 587 samples of anti-HCV an-tibody positive detected by CLIA were furteher detected HCV-RNA by FQ-PCR.Results Among 587 samples of anti-HCV anti-body positive by the CLIA screening,225 samples were HCV-RNA negative and 362 samples were HCV-RNA positive detected by FQ-PCR,and the positive rate was 61 .67%,moreover,which was positively correlated with the S/CO ratio detected by CLIA.Con-clusion The positive rate of HCV-RNA is positively correlated with the S/CO ratio detected by CLIA.The result of HCV-RNA can be predicted according to the S/CO ratio.

Objective To explore the association between HLA-Cw alleles with systemic lupuserythematosus. Methods Polymerase chain reaction-sequence specific primer method was used to analyze thedistribution of HLA-Cw01-08 alleles among 108 patients with SLE and 102 healthy controls. The allelefrequencies was compared between various patient groups and the control population. Results The frequencyof HLA-Cw07 alleles in patients with SLE was significantly increased in patients with SLE. Conclusion Theresults indicate that HLA-Cw07 may be the susceptible alleles or may be closely linked to the susceptiblegenes for SLE.

Objective To investigate the alterations in killer cell immunoglobulin-like receptors (KIRs)2D and their specific HLA-Cw ligands in patients with ankylosing spondylitis(AS)and determine whether the changes were correlate to the pathogenesis of AS.Methods Polymerase chain reaction of sequence specific primerB(PCR-SSP)was employed for genotyping the presence or absence of five KIR2D genes(KIR2DL1,2DS1,KIR2DL2,2DL3,2DS2)as well as HLA-Cw01-08 alleles from genomic DNA in 105 individuals with AS,together with 51 individuals with osteoarthritis(OA)and 120 healthy controls.Then HLA-C10-08 was divided into two groups.HLA-Cwasn and HLA-Cwlys to calculate the frequency of KIRID genotype.HLA-Gu alleles and KIR/HLA-Cw genotypes.Results The frequencies of HLA-Cwlys genes were significandy higher in patients with AS(0.269 7)compared with those in OA controls(0.148 2)and healthy controls(0.138 8,P=0.024,P=0.001,respectively).The frequency of KIR2DS1/HLA-Cwlys combination Was also markedly higher in AS group(26.67%)than that in OA controls(11.76%)and healthy controls(13.33%,P=0.039,P=0.018,respectively).Condusion The data suggest that the HLA-Cwlys allele may be associated with genetic susceptibility to AS and moreover.in the existence of HLA-Cwlys.the individuals with KIR2DS1 gene are likely to be at increased risk of AS.

Objective To investigate the relationship of the killer cell immunoglobulin-like receptor (KIR) gene polymorphism with Hashimoto′s thyroiditis(HT). Methods One hundred HT patients and 260 randomly matched healthy controls were enrolled to detect the KIR genotype. The genomic DNA were extracted, and 15 selected KIR genes, KIR2DL1-5, KIR3DL1-3, KIR2DS1-5, KIR3DS1 and pseudogene KIR2DP1, were determined by a polymerase chain reaction using sequence-specific primers (PCR-SSP). Results The frequency of KIR2DL5 gene was significantly lower of the patient group than that of the control group (0.200 vs 0.312, RR=0.64, P<0.01). Conclusion There may be an association between pathogenesis of HT and KIR2DL5 gene.