RESUMEN
BACKGROUND: To evaluate the performances of several advanced deep convolutional neural network models (AlexNet, VGG, GoogLeNet, ResNet) based on ensemble learning for recognizing chronic gingivitis from screening oral images. METHODS: A total of 683 intraoral clinical images acquired from 134 volunteers were used to construct the database and evaluate the models. Four deep ConvNet models were developed using ensemble learning and outperformed a single model. The performances of the different models were evaluated by comparing the accuracy and sensitivity for recognizing the existence of gingivitis from intraoral images. RESULTS: The ResNet model achieved an area under the curve (AUC) value of 97%, while the AUC values for the GoogLeNet, AlexNet, and VGG models were 94%, 92%, and 89%, respectively. Although the ResNet and GoogLeNet models performed best in classifying gingivitis from images, the sensitivity outcomes were not significantly different among the ResNet, GoogLeNet, and Alexnet models (p>0.05). However, the sensitivity of the VGGNet model differed significantly from those of the other models (p < 0.001). CONCLUSION: The ResNet and GoogLeNet models show promise for identifying chronic gingivitis from images. These models can help doctors diagnose periodontal diseases efficiently or based on self-examination of the oral cavity by patients.
Asunto(s)
Gingivitis , Redes Neurales de la Computación , Humanos , Gingivitis/diagnóstico , Gingivitis/patología , Enfermedad Crónica , Adulto , Femenino , Fotografía Dental/métodos , Masculino , Aprendizaje Profundo , FotograbarRESUMEN
Low detectability and camouflage skills in the electromagnetic wave and light frequency range provide survival advantages for natural creatures and are essential for understanding the operational principles of the biosphere. Taking inspiration from natural mutualistic symbiosis, this paper proposes a symbiotic electromagnetic shadow camouflage mechanism based on a superdispersive surface, aiming to investigate its impact on the observability of specific objects. The design and experimental results indicate that the symbiotic shadow dihedral can significantly reduce overall scattering quantity, which reaches at least 10 dB shrink in the 12-18 GHz frequency range compared to the contrast object. Unlike known camouflage methods, the electromagnetic shadow technology shrinks the overall scattering without any coating and shield metal target while probably offering extensive functional design freedom for the concealed object, creature, or equipment. This also provides a hint to explore symbiosis-related camouflage phenomena in nature.
RESUMEN
Extreme scenario of lightning strikes would generate ultra-fast rotation of state-of-polarization (RSOP) up to 5.1â Mrad/s and large polarization mode dispersion (PMD) in optical ground wire (OPGW). Unfortunately, the conventional multiple modulus algorithm (MMA) cannot equalize these polarization impairments in polarization division multiplexing (PDM) probabilistic constellation shaping (PCS)-64QAM system. Moreover, due to unavoidable linearization errors and higher modulation order, the extended Kalman filter based on measurement equations of concatenated multiplication (EKF-CM) is highly unstable and fails under such scenarios. To address the above issues, we have proposed a joint equalization scheme of PMD and RSOP, which fuses probability-aware with square-root cubature Kalman filter (PA-SCKF). Firstly, according to the characteristic that the amplitude of PCS signals obeys mixed Rician distribution, the scheme combines maximum a posteriori criterion to obtain the optimal radius of constellation ring which the received symbol belongs to, for the sake of calculating the innovations of SCKF. Secondly, it performs joint equalization of PMD and RSOP impairments based on SCKF and time-frequency conversion architecture. 28GBaud PDM PCS-64QAM simulation results demonstrate that our scheme can jointly equalize maximum impairments of 8.34â Mrad/s RSOP and 90ps DGD under entropy of 4.5bits/symbol. Additionally, only 0.9â dB OSNR penalty is obtained after joint equalization of 6â Mrad/s RSOP with 70ps DGD impairments. Even under entropy of 5.5bit/symbol, it can still jointly equalize impairments of 6.05â Mrad/s RSOP with 60ps DGD. Furthermore, 16GBaud PCS-64QAM experimental results indicate that the maximum joint equalization performances of PA-SCKF scheme under entropy of 4.5bit/symbol and 5bit/symbol are 17â Mrad/s RSOP with 52ps DGD, and 9â Mrad/s RSOP with 52ps DGD, respectively. These results manifest that our PA-SCKF scheme outperforms both MMA and EKF-CM schemes. Importantly, its complexity is on an order of O(Llog2â L), which is comparable to that of EKF-CM scheme.
RESUMEN
OBJECTIVE: This study investigates the effects of ZNF862 on the proliferation and apoptosis of human gingival fibroblasts and their related mechanisms. BACKGROUND: As a major transcription factor family, zinc finger proteins (ZFPs) regulate cell differentiation, growth, and apoptosis through their conserved zinc finger motifs, which allow high flexibility and specificity in gene regulation. In our previous study, ZNF862 mutation was associated with hereditary gingival fibromatosis. Nevertheless, little is known about the biological function of ZNF862. Therefore, this study was aimed to reveal intracellular localization of ZNF862, the influence of ZNF862 on the growth and apoptosis of human gingival fibroblasts (HGFs) and its potential related mechanisms. METHODS: Immunohistochemistry, immunofluorescence staining, and western blotting were performed to determine the intracellular localization of ZNF862 in HGFs. HGFs were divided into three groups: ZNF862 overexpression group, ZNF862 interference group, and the empty vector control group. Then, the effects of ZNF862 on cell proliferation, migration, cell cycle, and apoptosis were evaluated. qRT-PCR and western blotting were performed to further explore the mechanism related to the proliferation and apoptosis of HGFs. RESULTS: ZNF862 was found to be localized in the cytoplasm of HGFs. In vitro experiments revealed that ZNF862 overexpression inhibited HGFs proliferation and migration, induced cell cycle arrest at the G0/G1-phase and apoptosis. Whereas, ZNF862 knockdown promoted HGFs proliferation and migration, accelerated the transition from the G0/G1 phase into the S and G2/M phase and inhibited cell apoptosis. Mechanistically, the effects of ZNF862 on HGFs proliferation and apoptosis were noted to be dependent on inhibiting the cyclin-dependent kinase inhibitor 1A (p21)-retinoblastoma 1 (RB1) signaling pathway and enhancing the B-cell lymphoma-extra-large (Bcl-xL)-Caspase 3 signaling pathway. CONCLUSION: Our results for the first time reveal that ZNF862 is localized in the cytoplasm of HGFs. ZNF862 can inhibit the proliferation of HGFs by inhibiting the p21-RB1 signaling pathway, and it also promotes the apoptosis of HGFs by enhancing the Bcl-xL-Caspase 3 signaling pathway.
Asunto(s)
Apoptosis , Caspasa 3 , Proliferación Celular , Inhibidor p21 de las Quinasas Dependientes de la Ciclina , Fibroblastos , Encía , Transducción de Señal , Proteína bcl-X , Humanos , Apoptosis/genética , Proteína bcl-X/metabolismo , Caspasa 3/metabolismo , Ciclo Celular , Movimiento Celular , Células Cultivadas , Inhibidor p21 de las Quinasas Dependientes de la Ciclina/metabolismo , Proteínas de Unión al ADN/metabolismo , Proteínas de Unión al ADN/genética , Fibroblastos/metabolismo , Encía/citología , Encía/metabolismo , Factores de Transcripción/metabolismoRESUMEN
OBJECTIVES: Instance-level tooth segmentation extracts abundant localization and shape information from panoramic radiographs (PRs). The aim of this study was to evaluate the performance of a mask refinement network that extracts precise tooth edges. METHODS: A public dataset which consists of 543 PRs and 16211 labelled teeth was utilized. The structure of a typical Mask Region-based Convolutional Neural Network (Mask RCNN) was used as the baseline. A novel loss function was designed focus on producing accurate mask edges. In addition to our proposed method, 3 existing tooth segmentation methods were also implemented on the dataset for comparative analysis. The average precisions (APs), mean intersection over union (mIoU), and mean Hausdorff distance (mHAU) were exploited to evaluate the performance of the network. RESULTS: A novel mask refinement region-based convolutional neural network was designed based on Mask RCNN architecture to extract refined masks for individual tooth on PRs. A total of 3311 teeth were correctly detected from 3382 tested teeth in 111 PRs. The AP, precision, and recall were 0.686, 0.979, and 0.952, respectively. Moreover, the mIoU and mHAU achieved 0.941 and 9.7, respectively, which are significantly better than the other existing segmentation methods. CONCLUSIONS: This study proposed an efficient deep learning algorithm for accurately extracting the mask of any individual tooth from PRs. Precise tooth masks can provide valuable reference for clinical diagnosis and treatment. This algorithm is a fundamental basis for further automated processing applications.
Asunto(s)
Algoritmos , Diente , Humanos , Radiografía Panorámica , Redes Neurales de la Computación , Diente/diagnóstico por imagenRESUMEN
Equalization-enhanced phase noise (EEPN) has emerged as one of the major impairments that cannot be ignored for a high baud rate Stokes vector direct detection (SVDD) system. When EEPN interacts with the rotation of state-of-polarization (RSOP) and chromatic dispersion (CD), the joint impairment effects become even more complicated. To achieve the joint equalization of EEPN, RSOP, and CD impairments of a high baud rate SVDD system, this paper first derives a joint impairment model of these three kinds of impairments, and then proposes a joint equalization scheme of EEPN, RSOP, and CD with a sliding window assisted extended Kalman filter (SWA-EKF). The SWA-EKF scheme first tracks RSOP in the time domain, subsequently compensates CD in the frequency domain, and finally performs EEPN mitigation in the time domain again. The effectiveness of the proposed scheme has been verified by a 60 GBaud SVDD-16QAM simulation system. The results show that when these three impairments are jointly equalized, the SWA-EKF scheme can track RSOP as fast as 3 Mrad/s, cumulative dispersion up to 1600 ps/nm, and EEPN caused by laser linewidth up to 3 MHz. In addition, with an optical signal-to-noise ratio penalty of 0.3 dB, it could increase 35 G baud rate under 3 MHz laser linewidth for the SVDD system. More importantly, its total complexity can be reduced to an order of O(N log N).
RESUMEN
Porphyromonas gingivalis (P. gingivalis) is a key pathogen of chronic periodontitis. Aryl hydrocarbon receptor (AhR) is essential in immune homeostasis via modulation of pro-inflammatory cytokines production and indoleamine 2,3-dioxygenase (IDO). In this study, it is demonstrated that P. gingivalis may regulate AhR signalling in periodontitis, which provides a potential target for further immune regulation studies in periodontitis. Experimental periodontitis was induced in C57BL/6 mice by silk ligature and P. gingivalis oral inoculation. The alveolar bone resorption was examined using Micro-CT. Histological structures were observed and related cytokines involved in AhR signalling pathway were analysed. RAW264.7 cells were pretreated with AhR agonist (FICZ) and antagonist (CH223191) and infected with P. gingivalis subsequently. The levels of IDO, AhR and other related cytokines were measured. To demonstrate IDO activity, the concentrations of tryptophan (Trp) and kynurenine (Kyn) were assessed by HPLC. Histological analysis of periodontitis mice showed distinct alveolar bone resorption and inflammatory cell infiltration. The level of AhR and its downstream target factors were significantly decreased in inflamed gingival tissue. Furthermore, RAW 264.7 cells incubated by P. gingivalis exhibited increased pro-inflammatory cytokines production and decreased AhR, CYP1A1, CYP1B1, and IDO expression. Decreased IDO activity was observed as decreased Kyn/Trp ratio in the supernatant. Moreover, FICZ decreased the pro-inflammatory cytokines levels in P. gingivalis infected cells. It is concluded that P. gingivalis may promote inflammatory responses via inhibiting the AhR signalling pathway in periodontitis.
Asunto(s)
Pérdida de Hueso Alveolar , Periodontitis , Animales , Ratones , Porphyromonas gingivalis/metabolismo , Receptores de Hidrocarburo de Aril , Ratones Endogámicos C57BL , Periodontitis/patología , Inflamación , Citocinas/metabolismo , Pérdida de Hueso Alveolar/patologíaRESUMEN
Moderate or strong shaping conditions reduce the occurrence probability of the outermost ring constellation points of probabilistically shaped (PS)-M quadrature amplitude modulation (QAM) signals, which easily causes the peaks in the 4th power periodogram of received signals be submerged, accordingly the classical frequency offset estimation (FOE) scheme using 4th power fast Fourier transform (FFT) cannot be applied in PS-MQAM system. To solve this issue, we have proposed an optimal decision threshold assisted quadrature phase shift keying (QPSK)-partition blind FOE scheme. Firstly, the proposed scheme utilizes an optimal decision threshold assisted method for the symbol decision of received symbols, then chooses the symbols on multiple specific QPSK-shape rings. Secondly, the amplitude of each symbol selected above is normalized and uniformly augmented to 18. Finally, it carries out FOE using an improved time-domain 4th power feedforward method that eliminates the time interval. The effectiveness of the proposed scheme has been verified by 28 GBaud polarization division multiplexing (PDM) PS-16/64QAM simulations and 28/8 GBaud PS-16/64QAM experiments. The results obtained by this scheme present that under moderate or strong shaping conditions, the generalized mutual information (GMI) increases with optical signal-to-noise ratio (OSNR) and eventually exceeds the corresponding GMI threshold. Besides that, the FOE range can reach [-Rs/8, Rs/8], where Rs denotes the baud rate. When OSNRs are higher than 16â dB and 19.5â dB, the NMSEs of PS-16QAM-3/3.6 are lower than 1e-7, respectively. For PS-64QAM-4.4/5, the NMSEs achieve lower than 1e-6 after OSNR increases to 20.3â dB and 23.4â dB, respectively. More importantly, the overall complexity can be reduced to O(N), which is at most as 26.5% as that of FFT FOE scheme.
RESUMEN
OBJECTIVE: To explore the clinical and genetic characteristics of a Chinese pedigree affected with hereditary dentinogenesis imperfecta (DGI) type II. METHODS: Clinical data of the pedigree members were collected. Genomic DNA was extracted from peripheral blood samples and subjected to whole exome sequencing. RESULTS: Clinical characteristics of the affected family members have included amber teeth along with significant attrition, constricted roots and dentine hypertrophy leading to pulpal obliteration, which were suggestive of DGI type II. All of the affected members were found to have harbored a novel heterozygous c.2837delA (p.Asp946Valfs*368) variant of the DSPP gene which was predicted to be likely pathogenic. CONCLUSION: The c.2837delA variant of the DSPP gene probably underlay the disease in this pedigree. Above finding has expanded the variant spectrum of DSPP gene and provided a basis for molecular diagnosis and genetic counseling for this pedigree.
Asunto(s)
Dentinogénesis Imperfecta , Dentinogénesis Imperfecta/genética , Proteínas de la Matriz Extracelular/genética , Humanos , Mutación , Linaje , Fosfoproteínas/genética , Sialoglicoproteínas/genéticaRESUMEN
[This corrects the article DOI: 10.3389/fcimb.2021.820535.].
RESUMEN
Human dental pulp stem cells (hDPSCs) have attracted tremendous attention in tissue regeneration engineering due to their excellent multidirectional differentiation potential. Photobiomodulation (PBM) using low-level light-emitting diodes (LEDs) or lasers has been proved to promote the osteogenesis of mesenchymal stem cells. However, the effect of LEDs on osteogenic differentiation of hDPSCs has little published data. In this work, the effect of blue LEDs with different energy densities of 2, 4, 6, 8, 10 J/cm2 on osteogenic differentiation of hDPSCs was examined by using in vitro ALP staining, ALP activity, mineralization, and real-time PCR. The results showed that compared with the control group, osteogenic differentiation was significantly enhanced in blue LEDs treated groups. As the energy density increased, the level of osteogenesis initially increased and then decreased reaching the highest level at 6 J/cm2. Transient receptor potential vanilloid 1 (TRPV1), a Ca2+ ion channel, was believed to be a potential player in osteogenesis by photobiomodulation. By immunofluorescence assay, calcium influx assay, PCR, and ALP staining, it was shown that blue LEDs irradiation can increase the activity of TRPV1 and intracellular calcium levels similarly to the agonist of TRPV1 capsaicin. Additionally, pretreatment with capsazepine, a selective TRPV1 inhibitor, was able to abrogate the osteogenic effect of blue LEDs. In conclusion, these findings proposed that blue LEDs can promote the osteogenesis of hDPSCs within the appropriate range (4-8 J/cm2) during culture of osteogenic medium, and TRPV1/Ca2+ may be an essential signaling pathway involved in blue LEDs-induced osteogenesis, providing new insights for the use of hDPSCs in tissue regeneration engineering.
Asunto(s)
Células Madre Mesenquimatosas , Osteogénesis , Calcio/metabolismo , Diferenciación Celular/efectos de la radiación , Proliferación Celular/efectos de la radiación , Células Cultivadas , Pulpa Dental , Humanos , Osteogénesis/efectos de la radiaciónRESUMEN
Periodontal Ehlers-Danlos syndrome (pEDS) is a rare hereditary disorder characterized by severe early-onset periodontitis with premature tooth loss, pretibial hyperpigmentation, and skin fragility. It is caused by mutant variants in the C1R and C1S genes that result in C4 cleavage and local complement cascade activation, as well as other possible consequences. However, the exact functional consequences of this activation remain unclear. To shed light on molecular mechanisms underlying pEDS and to identify novel molecular targets that may expand treatment strategies, we performed transcriptome profiling by RNA sequencing of monocytes and gingival fibroblasts from two patients with pEDS. Compared to normal controls, differential expression of genes was found only in monocytes but not gingival fibroblasts. Most of the significant genes were enriched in biological processes such as neutrophil-mediated immunity, response to bacterium, TNF-α and IL-17 pathway which are related to inflammation response and immune response. In disease ontology enrichment analysis, genes related to periodontal host defense, inflammatory response, skin disease, and vascular development, including MMP9, VEGFA, IL10, IL1A, IL1B, IL2RA, and IL6, were significantly enriched and also validated by qPCR and ELISA. Overall, the present study provides the transcriptomic data of pEDS for the first time and the distinct molecular features in monocytes of pEDS might serve as a tool to better understand the disease.
RESUMEN
In this study, we propose and verify a joint multi-parameter optical performance monitoring (OPM) scheme based on trajectory information for the Stokes vector direct detection (SVDD) system, for the first time, to the best of our knowledge. Here, the proposed scheme first performs quantification of the trajectory to construct trajectory information, which not only presents diversity of the received symbols in spatial dimension, but also records the jump pattern among symbols in time dimension. Subsequently, eigenanalysis is introduced to extract critical features hidden in trajectory information and simultaneously achieve the purpose of dimensionality reduction. The effectiveness of the scheme is verified through 14/28 GBaud SVDD binary phase shift keying/quadrature phase shift keying/-8 quadrature amplitude modulation (QAM)/-16QAM/-32QAM/-64QAM simulation systems. Under the scenario of joint modulation format (MF) identification and optical signal to noise ratio (OSNR) monitoring, the identification rates of all six kinds of MFs achieve 100% within their corresponding reasonable OSNR ranges. Besides that, the average mean absolute error (MAE) of the monitored OSNRs are obtained as 0.03 dB, 0.22 dB, 0.36 dB, 0.41 dB, 0.46 dB, and 0.49 dB for those six kinds of MFs, respectively. Under the scenario of multi-parameter OPM, SVDD-8QAM/-16QAM/-32QAM signals are 100% successfully identified when residual chromatic dispersion (RCD) is located in the ranges of 0-200 ps/nm, 0-190 ps/nm, and 0-160 ps/nm, respectively. The average MAE of OSNR monitoring and RCD estimation for these three commonly used MFs are 1.08 dB and 3.23 ps/nm, respectively. Moreover, the study also demonstrates the robustness for baud rates and a relatively simpler calculation complexity about the proposed OPM scheme.
RESUMEN
OBJECTIVE: This study investigated the peripheral Th17/Treg balance and its potential controlling factor indoleamine 2,3 dioxygenase (IDO) in patients with periodontitis and atherosclerosis (AS), as well as its correlation with Porphyromonas gingivalis infection. METHODS: In this retrospective study, P. gingivalis-infected atherosclerotic patients (Pg-AS), atherosclerotic patients (AS), P. gingivalis-infected periodontitis patients (Pg), and healthy controls (HCs) were selected after clinical examination, subgingival plaque examination, and plasma anti-P. gingivalis antibody analysis. Treg and Th17 cell percentages, related transcription factors, and functional cytokines in peripheral blood were analysed. Plasma tryptophan (Trp) and kynurenine (Kyn) were measured to determine IDO activity. RESULTS: Atherosclerotic patients (Pg-AS and AS groups) had significantly lower IDO activity and higher Th17/Treg ratio than those in the Pg and HC groups. The Th17/Treg ratio was higher and IDO activity was lower in the Pg-AS group compared with the AS group. Transcription factors and cytokines exhibited the same trend as the Th17 and Treg cells. Additionally, IDO activity was negatively correlated with the plasma anti-P. gingivalis antibody titre and the Th17/Treg ratio in the atherosclerotic group. CONCLUSIONS: P. gingivalis may reduce IDO activity and further promote Th17/Treg imbalance to facilitate AS development. IDO may be a novel molecular marker to predict periodontitis-associated AS.
Asunto(s)
Aterosclerosis , Periodontitis , Humanos , Indolamina-Pirrol 2,3,-Dioxigenasa , Periodontitis/complicaciones , Estudios Retrospectivos , Linfocitos T Reguladores , Células Th17RESUMEN
Hereditary gingival fibromatosis (HGF) is the most common genetic form of gingival fibromatosis which is featured as a localized or generalized overgrowth of gingivae. Currently two genes (SOS1 and REST), as well as four loci (2p22.1, 2p23.3-p22.3, 5q13-q22, and 11p15), have been identified as associated with HGF in a dominant inheritance pattern. Here, we report 13 individuals with autosomal-dominant HGF from a four-generation Chinese family. Whole-exome sequencing followed by further genetic co-segregation analysis was performed for the family members across three generations. A novel heterozygous missense mutation (c.2812G > A) in zinc finger protein 862 gene (ZNF862) was identified, and it is absent among the population as per the Genome Aggregation Database. The functional study supports a biological role of ZNF862 for increasing the profibrotic factors particularly COL1A1 synthesis and hence resulting in HGF. Here, for the first time we identify the physiological role of ZNF862 for the association with the HGF.
Asunto(s)
Fibromatosis Gingival/genética , Predisposición Genética a la Enfermedad , Proliferación Celular , Regulación hacia Abajo , Femenino , Fibromatosis Gingival/patología , Regulación de la Expresión Génica , Técnicas de Silenciamiento del Gen , Pruebas Genéticas , Humanos , Masculino , Mutación , Linaje , Interferencia de ARN , Análisis de Secuencia de ARN , Regulación hacia ArribaRESUMEN
As a chronic infectious disease, periodontitis can cause gum recession, loss of alveolar bone, loosening of teeth, and even loss of teeth. Dental plaque biofilm is the initiating factor for the occurrence and development of periodontitis. Fusobacterium nucleatum (F. nucleatum) plays a vital role in the structure and ecology of dental plaque biofilms. It is a bridge between early and late colonization bacteria in dental plaque. Understanding the molecular mechanism of F. nucleatum during biofilm development is essential to control periodontitis. This study aimed to determine gene expression profiles of the F. nucleatum strain, ATCC 25586, in the planktonic and biofilm phase through RNA-sequencing approach. The results were confirmed by quantitative reverse transcriptase PCR (RT-qPCR). The results clearly illustrate the difference in gene expression of F. nucleatum under planktonic and biofilms. A total of 110 genes were differentially expressed by F. nucleatum in the biofilm state compared with the planktonic state. The 25 upregulated genes in the biofilm state were mainly related to carbohydrate and amino acid metabolism, while the 85 downregulated genes were primarily associated with cell growth, division, and oxidative stress; most of the upregulated genes of F. nucleatum involved in virulence and oral malodor. Furthermore, the transcriptome analysis and antibacterial activity test also identified Lysine might exhibit the antibacterial and antibiofilm activity of F. nucleatum for the first time. These new findings could provide caveats for future studies on the regulation and maintenance of plaque biofilm and the development of biomarkers for periodontitis.
Asunto(s)
Proteínas Bacterianas/metabolismo , Biopelículas/crecimiento & desarrollo , Fusobacterium nucleatum/metabolismo , Plancton/metabolismo , Transcriptoma , Proteínas Bacterianas/genética , Fusobacterium nucleatum/genética , Perfilación de la Expresión Génica , Regulación Bacteriana de la Expresión Génica , Humanos , Plancton/genética , Plancton/crecimiento & desarrollo , VirulenciaRESUMEN
OBJECTIVES: To develop and evaluate the performance of a deep learning system based on convolutional neural network (ConvNet) to detect dental caries from oral photographs. METHODS: 3,932 oral photographs obtained from 625 volunteers with consumer cameras were included for the development and evaluation of the model. A deep ConvNet was developed by adapting from Single Shot MultiBox Detector. The hard negative mining algorithm was applied to automatically train the model. The model was evaluated for: (i) classification accuracy for telling the existence of dental caries from a photograph and (ii) localization accuracy for locations of predicted dental caries. RESULTS: The system exhibited a classification area under the curve (AUC) of 85.65% (95% confidence interval: 82.48% to 88.71%). The model also achieved an image-wise sensitivity of 81.90%, and a box-wise sensitivity of 64.60% at a high-sensitivity operating point. The hard negative mining algorithm significantly boosted both classification (p < .001) and localization (p < .001) performance of the model by reducing false-positive predictions. CONCLUSIONS: The deep learning model is promising to detect dental caries on oral photographs captured with consumer cameras. It can be useful for enabling the preliminary and cost-effective screening of dental caries among large populations.
Asunto(s)
Aprendizaje Profundo , Caries Dental , Algoritmos , Área Bajo la Curva , Caries Dental/diagnóstico por imagen , Humanos , Redes Neurales de la ComputaciónRESUMEN
Blind modulation format identification (MFI) is indispensable for correct signal demodulation and optical performance monitoring in future elastic optical networks (EON). Existing MFI schemes based on a clustering algorithm in Stokes space have gained good performance, while only limited types of modulation formats could be correctly identified, and the complexities are relatively high. In this work, we have proposed an MFI scheme with a low computational complexity, which combines an improved particle swarm optimization (I-PSO) clustering algorithm with a 2D Stokes plane. The main idea of I-PSO is to add a new field of view on each particle and limit each particle to only communicate with its neighbor particles, so as to realize the correct judgment of the number of multiple clusters (local extrema) on the density images of the s2-s3 plane. The effectiveness has been verified by 28 GBaud polarization division multiplexing (PDM)-BPSK/PDM-QPSK/PDM-8QAM/PDM-16QAM/PDM-32QAM/PDM-64QAM simulation EON systems and 28 GBaud PDM-QPSK/PDM-8QAM/PDM-16QAM/PDM-32QAM proof-of-concept transmission experiments. The results show that, using this MFI scheme, the minimum optical signal-to-noise ratio (OSNR) values to achieve 100% MFI success rate are all equal to or lower than those of the corresponding 7% forward error correction (FEC) thresholds. At the same time, the MFI scheme also obtains good tolerance to residual chromatic dispersion and differential group delay. Besides that, the proposed scheme achieves 100% MFI success rate within a maximum launch power range of -2â¼+6 dBm. More importantly, its computational complexity can be denoted as O(N).
RESUMEN
OBJECTIVE: Chronic periodontitis (CP), aggressive periodontitis (AP), and peri-implantitis (PI) are chronic inflammatory diseases. Tumor necrosis factor-α (TNF-a) is an effective immune inflammatory mediator. Several studies have been conducted to explore the association between the TNF-α (G-308A) polymorphism and susceptibility to CP, AP, and PI. Our objective was to examine whether the TNF-α (G-308A) polymorphism is related to these diseases. METHODS: We conducted a meta-analysis to investigate the association between the TNF-α (G-308A) polymorphism and CP, AP, and PI. The PubMed, Embase, CNKI, and Web of Science electronic databases were searched for studies published from inception to August 11, 2020; the reference lists of included studies were also searched. The included studies were assessed in the following genetic models: dominant model, recessive model, allelic model, heterozygous model, and homozygous model. RESULTS: Forty articles (50 comparisons) with 2243 CP, 824 AP, 615 PI, 795 healthy peri-implant, and 3575 healthy controls were considered for the TNF-α (G-308A) polymorphism in this meta-analysis. Variant A of TNF-α (G-308A) was associated with increased AP risk in the general population, especially in Asians, and this polymorphism was significantly associated with elevated risk of CP in Asians and Caucasians. There was no association between the A allele and PI risk. None of the contrasts of the genetic model yielded a significant finding in Latin Americans. Different genotyping methods may affect the association between the TNF-α (G-308A) polymorphism and these diseases. CONCLUSION: These findings supported that variant A of the TNF-α (G-308A) polymorphism may contribute to CP and AP susceptibility, particularly in Asians and Caucasians. More efforts and further studies with larger sample sizes will be required to validate the risk of CP, AP, and PI.