Swin MoCo: Improving parotid gland MRI segmentation using contrastive learning.

BACKGROUND: Segmentation of the parotid glands and tumors by MR images is essential for treating parotid gland tumors. However, segmentation of the parotid glands is particularly challenging due to their variable shape and low contrast with surrounding structures. PURPOSE: The lack of large and well-annotated datasets limits the development of deep learning in medical images. As an unsupervised learning method, contrastive learning has seen rapid development in recent years. It can better use unlabeled images and is hopeful to improve parotid gland segmentation. METHODS: We propose Swin MoCo, a momentum contrastive learning network with Swin Transformer as its backbone. The ImageNet supervised model is used as the initial weights of Swin MoCo, thus improving the training effects on small medical image datasets. RESULTS: Swin MoCo trained with transfer learning improves parotid gland segmentation to 89.78% DSC, 85.18% mIoU, 3.60 HD, and 90.08% mAcc. On the Synapse multi-organ computed tomography (CT) dataset, using Swin MoCo as the pre-trained model of Swin-Unet yields 79.66% DSC and 12.73 HD, which outperforms the best result of Swin-Unet on the Synapse dataset. CONCLUSIONS: The above improvements require only 4 h of training on a single NVIDIA Tesla V100, which is computationally cheap. Swin MoCo provides new approaches to improve the performance of tasks on small datasets. The code is publicly available at https://github.com/Zian-Xu/Swin-MoCo.

Parotid Gland Segmentation Using Purely Transformer-Based U-Shaped Network and Multimodal MRI.

Parotid gland tumors account for approximately 2% to 10% of head and neck tumors. Segmentation of parotid glands and tumors on magnetic resonance images is essential in accurately diagnosing and selecting appropriate surgical plans. However, segmentation of parotid glands is particularly challenging due to their variable shape and low contrast with surrounding structures. Recently, deep learning has developed rapidly, and Transformer-based networks have performed well on many computer vision tasks. However, Transformer-based networks have yet to be well used in parotid gland segmentation tasks. We collected a multi-center multimodal parotid gland MRI dataset and implemented parotid gland segmentation using a purely Transformer-based U-shaped segmentation network. We used both absolute and relative positional encoding to improve parotid gland segmentation and achieved multimodal information fusion without increasing the network computation. In addition, our novel training approach reduces the clinician's labeling workload by nearly half. Our method achieved good segmentation of both parotid glands and tumors. On the test set, our model achieved a Dice-Similarity Coefficient of 86.99%, Pixel Accuracy of 99.19%, Mean Intersection over Union of 81.79%, and Hausdorff Distance of 3.87. The purely Transformer-based U-shaped segmentation network we used outperforms other convolutional neural networks. In addition, our method can effectively fuse the information from multi-center multimodal MRI dataset, thus improving the parotid gland segmentation.

CCR7 affects the tumor microenvironment by regulating the activation of naïve CD8+ T cells to promote the proliferation of oral squamous cell carcinoma.

BACKGROUND: Head and neck cancer is the sixth most common malignancy worldwide, and oral squamous cell carcinoma (OSCC) is the most common head and neck cancer, being one of the leading causes of cancer morbidity and mortality worldwide. CC Chemokine receptor 7(CCR7) is a multifunctional G protein-coupled trans-membrane chemokine that affects immune cell chemotaxis, migration, and cancer progression through its interaction with its ligands C-C motif chemokine ligand 19(CCL19) and C-C motif chemokine ligand 21(CCL21). Numerous studies have demonstrated the involvement of CCR7 in the malignant progression of a variety of cancers, reflecting the pro-cancer properties of CCR7. The Cancer Genome Atlas data suggests CCR7 has elevated expression in oral cancer. Specifically, CCR7 expression in tumor microenvironment (TME) may regulate the ability of some immune cells to engage in anti-tumor immune responses. Since CD8+ T cells have become a key immunotherapeutic target, the role of CCR7 in antitumor immune response of naïve CD8+ T cells in TME has not been thoroughly investigated. METHODS: A CCR7 knockout mouse model was constructed, and the mechanism of ccr7 on the regulation of tumor microenvironment by naïve CD8+ T cells was verified under the guidance of single-cell RNA sequencing combined with in vivo animal experiments and in vitro cell experiments. RESULTS: CCR7 is knocked out with impaired tumor growth and altered CD8+ T cell profiles, revealing the importance of this protein in OSCC. CONCLUSIONS: Inhibition of CCR7 enhances CD8+ T cell activation, proliferation, and anti-tumor function, suggesting its potential as a therapeutic target.

Analysis of the effect of CCR7 on the microenvironment of mouse oral squamous cell carcinoma by single-cell RNA sequencing technology.

BACKGROUND: Studies have shown that CCR7, an important inflammatory factor, can promote the proliferation and metastasis of oral squamous cell carcinoma (OSCC), but its role in the tumor microenvironment (TME) remains unclear. This paper explores the role of CCR7 in the TME of OSCC. METHODS: In this work, we constructed CCR7 gene knockout mice and OSCC mouse models. Single-cell RNA sequencing (scRNA-seq) and bioinformatics were used to analyze the differences in the OSCC microenvironment between three CCR7 gene knockout mice (KO) and three wild-type mice (WT). Immunohistochemistry, immunofluorescence staining, and flow cytometry were used to analyze the expression of key genes in significantly different cell types between the KO and WT groups. An in vitro experiment was used to verify the effect of CCR7 on M2 macrophage polarization. RESULTS: In the mouse OSCC models, the tumor growth rate in the KO group was significantly lower than that in the WT group. Eight main cell types (including tumor cells, fibroblasts, macrophages, granulocytes, T cells, endothelial cells, monocytes, and B cells) were identified by Seurat analysis. The scRNA-seq results showed that the proportion of tumor cells was lower, but the proportion of inflammatory cells was significantly higher in the KO group than in the WT group. CellPhoneDB analysis results indicated a strong interaction relationship between tumor cells and macrophages, T cells, fibroblasts, and endothelial cells. Functional enrichment results indicated that the expression level of the Dusp1 gene in the KO group was generally higher than that in the WT group in various cell types. Macrophage subclustering results indicated that the proportion of M2 macrophages in the KO group was lower than that in the WT group. In vitro experimental results showed that CCR7 can promote M2 macrophage polarization, thus promoting the proliferation, invasion and migration of OSCC cells. CONCLUSIONS: CCR7 gene knockout can significantly inhibit the growth of mouse oral squamous cell carcinoma by promoting the polarization of M2 macrophages.

The application of radiomics machine learning models based on multimodal MRI with different sequence combinations in predicting cervical lymph node metastasis in oral tongue squamous cell carcinoma patients.

BACKGROUND: The purpose of this study was to explore preliminary the performance of radiomics machine learning models based on multimodal MRI to predict the risk of cervical lymph node metastasis (CLNM) for oral tongue squamous cell carcinoma (OTSCC) patients. METHODS: A total of 400 patients were enrolled in this study and divided into six groups according to the different combinations of MRI sequences. Group I consisted of patients with T1-weighted images (T1WI) and FS-T2WI (fat-suppressed T2-weighted images), group II consisted of patients with T1WI, FS-T2WI, and contrast enhanced MRI (CE-MRI), group III consisted of patients with T1WI, FS-T2WI, and T2-weighted images (T2WI), group IV consisted of patients with T1WI, FS-T2WI, CE-MRI, and T2WI, group V consisted of patients with T1WI, FS-T2WI, T2WI, and apparent diffusion coefficient map (ADC), and group VI consisted of patients with T1WI, FS-T2WI, CE-MRI, T2WI, and ADC. Machine learning models were constructed. The performance of the models was compared in each group. RESULTS: The machine learning model in group IV including T1WI, FS-T2WI, T2WI, and CE-MRI presented best prediction performance, with AUCs of 0.881 and 0.868 in the two sets. The models with CE-MRI performed better than the models without CE-MRI(I vs. II, III vs. IV, V vs. VI). CONCLUSIONS: The radiomics machine learning models based on CE-MRI showed great accuracy and stability in predicting the risk of CLNM for OTSCC patients.

Hierarchically structured passive radiative cooling ceramic with high solar reflectivity.

Passive radiative cooling using nanophotonic structures is limited by its high cost and poor compatibility with existing end uses, whereas polymeric photonic alternatives lack weather resistance and effective solar reflection. We developed a cellular ceramic that can achieve highly efficient light scattering and a near-perfect solar reflectivity of 99.6%. These qualities, coupled with high thermal emissivity, allow the ceramic to provide continuous subambient cooling in an outdoor setting with a cooling power of >130 watts per square meter at noon, demonstrating energy-saving potential on a worldwide scale. The color, weather resistance, mechanical robustness, and ability to depress the Leidenfrost effect are key features ensuring the durable and versatile nature of the cooling ceramic, thereby facilitating its commercialization in various applications, particularly building construction.

Reconstruction of large defects of anterior floor of mouth with free flaps using a novel individualized flap design method.

The purpose of this study was to introduce a novel individualized flap design method for large anterior floor of the mouth (AFOM) defect reconstruction, review experience with the use of this flap design method for large AFOM defect reconstruction, and assess its functional results. A retrospective study of patients who received large AFOM defect reconstruction with free flaps was conducted. There was a cohort of patients who were treated using the novel individualized flap design method and a cohort without flap design. Functional outcomes were evaluated with appropriate scales. Outcomes were analyzed, and a p-value <0.05 was considered significant. 22 patients received the individualized flap design, while 21 patients were treated without a special flap design. All flaps survived. All free flaps harvested with the novel individualized flap design method better matched AFOM defects. Relative to patients without flap design, patients in the novel individualized flap design group showed significant improvement in speech intelligibility (p = 0.036) and swallowing function (p = 0.019). Within the limitation of the study it seems that large AFOM defect reconstruction with the novel individualized flap design method can not only cover and close the wound to avoid oral-neck fistulae, but also maintains tongue mobility to achieve better functional outcomes than in patients without flap design.

A novel flap design technique for subtotal tongue reconstruction with an "Individualized and Convenient Tongue Model".

BACKGROUND: To achieve improved functional outcomes in subtotal tongue reconstruction, a flap design with sufficient volume and appropriate shape is necessary. In this study, we introduce an "Individualized and Convenient Tongue Model" (ICTM) for flap design in subtotal tongue reconstruction. METHODS: By studying the anatomical morphology of the tongue, we found a similar geometry within the dorsum and body of the tongue as well as the mouth floor. This can be used to create an ICTM through folding and splicing. We can simulate tongue defects in the ICTM and transform defect shapes into guide plates for flap design. In this study, fifty-eight patients requiring subtotal tongue reconstruction were randomly divided into two groups: an ICTM group (35 patients) and a conventional group (31 patients). In the ICTM group, we individually designed profunda artery perforator flaps (PAPFs) or anterolateral thigh flaps (ALTFs) using the ICTM method. In the conventional group, the flap was designed according to the surgeon's clinical experience. Patient demographics, operative and follow-up data were recorded. Swallowing, speech intelligibility, and cosmetic results were assessed using appropriate scales. RESULTS: All flaps survived, although there were no significant differences in tumor size, operation time, flap size, and complication rate compared to the conventional group. Patients in the ICTM group had significantly improved speech intelligibility (p = 0.019), cosmetic appearance (p = 0.009), and swallowing ability (p = 0.003). CONCLUSIONS: The ICTM technique is an effective and convenient solution for subtotal tongue reconstruction that provides an individualized flap design and improves functional outcomes compared to the conventional design.

An anatomy-aware framework for automatic segmentation of parotid tumor from multimodal MRI.

Magnetic Resonance Imaging (MRI) plays an important role in diagnosing the parotid tumor, where accurate segmentation of tumors is highly desired for determining appropriate treatment plans and avoiding unnecessary surgery. However, the task remains nontrivial and challenging due to ambiguous boundaries and various sizes of the tumor, as well as the presence of a large number of anatomical structures around the parotid gland that are similar to the tumor. To overcome these problems, we propose a novel anatomy-aware framework for automatic segmentation of parotid tumors from multimodal MRI. First, a Transformer-based multimodal fusion network PT-Net is proposed in this paper. The encoder of PT-Net extracts and fuses contextual information from three modalities of MRI from coarse to fine, to obtain cross-modality and multi-scale tumor information. The decoder stacks the feature maps of different modalities and calibrates the multimodal information using the channel attention mechanism. Second, considering that the segmentation model is prone to be disturbed by similar anatomical structures and make wrong predictions, we design anatomy-aware loss. By calculating the distance between the activation regions of the prediction segmentation and the ground truth, our loss function forces the model to distinguish similar anatomical structures with the tumor and make correct predictions. Extensive experiments with MRI scans of the parotid tumor showed that our PT-Net achieved higher segmentation accuracy than existing networks. The anatomy-aware loss outperformed state-of-the-art loss functions for parotid tumor segmentation. Our framework can potentially improve the quality of preoperative diagnosis and surgery planning of parotid tumors.

Swin MAE: Masked autoencoders for small datasets.

The development of deep learning models in medical image analysis is majorly limited by the lack of large-sized and well-annotated datasets. Unsupervised learning does not require labels and is more suitable for solving medical image analysis problems. However, most unsupervised learning methods must be applied to large datasets. To make unsupervised learning applicable to small datasets, we proposed Swin MAE, a masked autoencoder with Swin Transformer as its backbone. Even on a dataset of only a few thousand medical images, Swin MAE can still learn useful semantic features purely from images without using any pre-trained models. It can equal or even slightly outperform the supervised model obtained by Swin Transformer trained on ImageNet in the transfer learning results of downstream tasks. Compared to MAE, Swin MAE brought a performance improvement of twice and five times for downstream tasks on BTCV and our parotid dataset, respectively. The code is publicly available at https://github.com/Zian-Xu/Swin-MAE.

The role of p21-activated kinase 4 in the progression of oral squamous cell carcinoma by targeting PI3K-AKT signaling pathway.

BACKGROUND: Oral squamous carcinoma (OSCC), the most common head and neck malignancy, has a strong propensity for malignant proliferation and metastasis, which will decrease the survival of patients. P21-activated kinase 4 (PAK4), a classical serine/threonine protein kinase with multiple cellular functions, has an essential role in cancer cell migration and invasion. Here, we elucidated the function and possible molecular mechanisms of the effect of PAK4 on the biological behaviors of OSCC. METHODS: The expression of genes and protein was detected by real-time PCR and western blotting. We used oral squamous carcinoma cell lines, Tca8117, Cal 27, SCC 4, and SCC 9 for validation of our cell function data. Flow cytometry, 3D cultures, and clone formation assay were used to detect proliferation of cells. RNA sequencing and bioinformatic analysis was performed to determine the potential function of PAK4. RESULTS: Immunohistochemistry, western blotting and real-time PCR demonstrated that PAK4 expression was up-regulated in OSCC tissues. Overexpression of PAK4 promoted the proliferation, migration and invasion of OSCC cell lines. RNA sequencing (RNA-seq) for the transcriptome-wide analysis of differential gene expression followed by bioinformatic analysis was performed to determine the potential function of PAK4. Based on the KEGG enrichment analysis and GO analysis of differential expression genes (DEGs) we found that PAK4 promotes the cell-cycle machinery, which associated with 44 regulated genes, thereby promoting cancer cell differentiation. CONCLUSIONS: This study demonstrates that the PAK4 regulates the biological behaviors of OSCC by PI3K-AKT signaling pathway, and these findings might provide a novel strategy for OSCC treatment.

CC chemokine receptor 7 promotes macrophage recruitment and induces M2-polarization through CC chemokine ligand 19&21 in oral squamous cell carcinoma.

PURPOSE: This study aimed to investigate the impact of CC chemokine receptor 7 (CCR7) on the recruitment and polarization of tumor-associated macrophages (TAMs) in oral squamous cell carcinoma (OSCC). METHODS: We analyzed CCR7 expression pattern, clinicopathological significance, and its association with M2 macrophage infiltration in OSCC by bioinformatic methods. Small interfering RNA (siRNA) was utilized to silence CCR7 in OSCC cells. Conditioned media (CM) was harvested from transfected OSCC cells to establish a co-culture model of THP-1 derived macrophages and OSCC cells. Transwell assay and cell adhesion assay were performed to examine the effect of CCR7 on macrophages recruitment and adhesion. Cytoskeleton was labelled by phalloidin to observe macrophage morphological changes. Moreover, phenotypic alteration of macrophages was measured using quantitative real-time PCR (qRT-PCR), flow cytometry, and immunofluorescence (IF) staining. Ultimately, recombinant human CCL19 and CCL21 were added into the medium of THP-1 derived macrophages to explore their effects on polarization in vitro. RESULTS: In OSCC patients, the overexpression of CCR7 positively correlated with lymph node metastasis and M2 macrophage infiltration. Macrophage not only exhibited enhanced migration, invasion and adhesion abilities, but also appeared more spindle and branched in vitro when treated with CM from OSCC cells. However, these phenomena were abrogated with knockdown of CCR7. We also discovered that inhibition of CCR7 in OSCC cells suppressed TAMs polarization to an M2 phenotype. In addition, recombinant human CCL19 and CCL21 promoted macrophage M2-polarization in vitro. CONCLUSION: CCR7 in OSCC cells promoted recruitment and M2-polarization of THP-1 derived macrophages in vitro by regulating production of CCL19 and CCL21.

The Overexpression of Fibronectin 1 Promotes Cancer Progression and Associated with M2 Macrophages Polarization in Head and Neck Squamous Cell Carcinoma Patients.

Purpose: This study aimed to investigate the biological roles of fibronectin 1 (FN1) in head and neck squamous cell carcinoma (HNSCC) and its effects on macrophage M2 polarization. Methods: We analyzed FN1 expression pattern and examined its clinical relevance in HNSCC progression by bioinformatic analysis. Small interfering RNA (siRNA) was utilized to silence FN1 in HNSCC cells. Cell counting kit-8 (CCK-8) assay, colony formation assay, Transwell assay and wound healing assay were performed to reveal the effect of FN1 on malignant behaviors of HNSCC cells. Moreover, a co-culture model of macrophages and HNSCC cells was established to investigate whether FN1 induce macrophage M2 polarization. Finally, we used bioinformatic methods to explore the possible FN1-related pathways in HNSCC. Results: FN1 is significantly overexpressed in HNSCC patients and has been obviously correlated with higher pathological stage and poor prognosis. Downregulation of FN1 suppressed the proliferation, migration and invasion of HNSCC cells, and inhibited macrophage M2 polarization in vitro. In addition, "PI3K-Akt" and "MAPK" signaling pathways may be involved in the malignant process of FN1 in HNSCC. Conclusion: The overexpression of FN1 promotes HNSCC progression and induces macrophages M2 polarization. FN1 may serve as a promising prognostic biomarker and therapeutic target in HNSCC.

Inhibiting the Leidenfrost effect above 1,000 °C for sustained thermal cooling.

The Leidenfrost effect, namely the levitation of drops on hot solids1, is known to deteriorate heat transfer at high temperature2. The Leidenfrost point can be elevated by texturing materials to favour the solid-liquid contact2-10 and by arranging channels at the surface to decouple the wetting phenomena from the vapour dynamics3. However, maximizing both the Leidenfrost point and thermal cooling across a wide range of temperatures can be mutually exclusive3,7,8. Here we report a rational design of structured thermal armours that inhibit the Leidenfrost effect up to 1,150 °C, that is, 600 °C more than previously attained, yet preserving heat transfer. Our design consists of steel pillars serving as thermal bridges, an embedded insulating membrane that wicks and spreads the liquid and U-shaped channels for vapour evacuation. The coexistence of materials with contrasting thermal and geometrical properties cooperatively transforms normally uniform temperatures into non-uniform ones, generates lateral wicking at all temperatures and enhances thermal cooling. Structured thermal armours are limited only by their melting point, rather than by a failure in the design. The material can be made flexible, and thus attached to substrates otherwise challenging to structure. Our strategy holds the potential to enable the implementation of efficient water cooling at ultra-high solid temperatures, which is, to date, an uncharted property.

The prevalence of extra-salivary gland complications in immunoglobulin G4-related sialadenitis.

OBJECTIVES: To estimate the prevalence of extra-glandular lesions in patients with immunoglobulin G4-related sialadenitis (IgG4-RS). METHODS: Six electronic databases (PubMed, EMBASE, Science Direct, Scopus, Web of Science, and China National Knowledge Infrastructure) were systematically searched from the date of inception of each database to March 2021. The Strengthening the Reporting of Observational Studies in Epidemiology statement was used to conduct methodological quality assessment, and a random-effect meta-analysis model was applied to estimate the prevalence. Publication bias was visually assessed using a funnel plot and calculated via Begg's and Egger's tests. The Stata 15 software was used to perform data analysis. RESULTS: A total of 43 articles comprising 1,864 patients with IgG4-RS were considered to be eligible for this study. The pooled prevalence of extra-salivary gland lesions in IgG4-RS was 76.53% with a confidence interval (CI) of (69.39%, 83.04%). A higher prevalence was associated with studies published before or during 2015 (84.38%, CI [74.23%, 92.58%]) than those published after 2015 (68.55%, CI [58.44%, 77.88%]). Lacrimal gland involvement (54.68%, CI [45.61%, 63.60%]) and lymph node swelling (56.96%, CI [48.16%, 65.56%]) were the most frequent lesions. CONCLUSIONS: Extra-glandular lesions were common in patients with IgG4-RS. More high-quality prospective studies with less heterogeneity are required to determine the accurate prevalence.

Specificity protein 1/microRNA-92b forms a feedback loop promoting the migration and invasion of head and neck squamous cell carcinoma.

In this study we report a novel specificity protein 1 (SP1)/microRNA-92b (miR-92b) feedback loop regulating the migration and invasion of head and neck squamous cell carcinoma (HNSCC). Microarray and real-time Polymerase Chain Reaction (PCR) were used to detect gene expression in HNSCC tissues and cell lines. Transwell migration, invasion, wound healing and cell counting kit - 8 (CCK-8) cell assays were used to compare cell migration, invasion and proliferation abilities. Chromatin Immunoprecipitation (ChIP) assays were used to detect SP1 binding to the miR-92b promoter. Western blot was used to detect protein levels. An in vivo tumorigenesis experiment was used to evaluate the effect of SP1 knockdown on tumor growth and protein levels were evaluated by immunohistochemistry. We found that the miR-92b expression level was elevated in HNSCC primary focus tissue compared with adjacent normal tissue, and a higher level of miR-92b was related to a higher clinical stage and worse prognosis of HNSCC patients. MiR-92b and SP1 mutually promoted each expression and cooperatively facilitated the migration, invasion and proliferation of HNSCC cells. A decreased level of SP1/miR-92b resulted in a restraint of in vivo tumor growth. In conclusion, our results suggest that the SP1/miR-92b feedback loop generally promotes HNSCC invasion and metastasis, thus presenting a possible therapeutic target in the treatment of HNSCC patients.

TransMed: Transformers Advance Multi-Modal Medical Image Classification.

Over the past decade, convolutional neural networks (CNN) have shown very competitive performance in medical image analysis tasks, such as disease classification, tumor segmentation, and lesion detection. CNN has great advantages in extracting local features of images. However, due to the locality of convolution operation, it cannot deal with long-range relationships well. Recently, transformers have been applied to computer vision and achieved remarkable success in large-scale datasets. Compared with natural images, multi-modal medical images have explicit and important long-range dependencies, and effective multi-modal fusion strategies can greatly improve the performance of deep models. This prompts us to study transformer-based structures and apply them to multi-modal medical images. Existing transformer-based network architectures require large-scale datasets to achieve better performance. However, medical imaging datasets are relatively small, which makes it difficult to apply pure transformers to medical image analysis. Therefore, we propose TransMed for multi-modal medical image classification. TransMed combines the advantages of CNN and transformer to efficiently extract low-level features of images and establish long-range dependencies between modalities. We evaluated our model on two datasets, parotid gland tumors classification and knee injury classification. Combining our contributions, we achieve an improvement of 10.1% and 1.9% in average accuracy, respectively, outperforming other state-of-the-art CNN-based models. The results of the proposed method are promising and have tremendous potential to be applied to a large number of medical image analysis tasks. To our best knowledge, this is the first work to apply transformers to multi-modal medical image classification.

MKP-1 is required to limit myeloid-cell mediated oral squamous cell carcinoma progression and regional extension.

Mitogen-activated protein kinases (MAPKs) require MAPK phosphatases (MKPs) for deactivation of MAPK intracellular signaling. MKP-1 (encoded by Dusp1) is a key negative regulator of MAPKs and prior reports have indicated that MKP-1 regulates oral cancer-associated inflammation and leukocyte infiltration. OBJECTIVE: To determine the significance of myeloid-based expression of MKP-1 in oral cancer. METHODS: The Cancer Genome Atlas (TCGA) was used to address DUSP1 expression in oral squamous cell carcinoma (OSCC). Syngeneic and carcinogen-induced mouse models using global and myeloid-specific Dusp-1 deficient mice with immunophenotypic, histologic, and transcriptomic analyses and in vitro migration assays. RESULTS: Data from TCGA indicates the DUSP1 expression is inversely related to oral cancer burden and nodal involvement. Using murine models of OSCC, the role of MKP-1 signaling in tumor associated macrophages (TAMs) was assessed. Dusp1-deficient mice had increased tumor burden and TAM infiltrate with increased M2 macrophage polarization. Transcriptomic signatures of TAMs from Dusp1-deficent mice indicated a pro-metastatic phenotype as well as concomitant differences in myeloid-associated genes, cytokine/chemokine signaling, and Notch signaling consistent with tumor progression. In vitro and in vivo assays revealed mouse OSCC cells had a higher migration rate using TAM cell-free supernatant from Dusp1 deficiency mice compared to controls with enhanced regional cervical lymph node metastasis, respectively. To validate TAM studies using implantable mouse models, an OSCC progression model with conditional myeloid-specific Dusp-1 deficient mice demonstrated enhanced OSCC disease progression, characterized by advanced onset, histological stage, and tumor burden. CONCLUSION: Myeloid-based Dusp1-deficiency increases OSCC burden and metastasis through alteration in TAM recruitment, gene profile, and polarity suggesting that MKP-1 could be a viable target to reprogram TAM to limit local/regional OSCC extension.

Digital navigation and 3D model technology in mandibular reconstruction with fibular free flap: A comparative study.

OBJECTIVE: The low accuracy limits the use of fibular free flap (FFF). We apply digital navigation and 3D printing model technology in mandibular reconstruction to improve FFF's accuracy. METHODS: 34 patients who underwent with FFF to repair mandibular defects were divided into Navigation Group (13 cases, using digital navigation and 3D printing model) and Control Group (21 cases, only 3D printing model). We retrospectively reviewed patients' hospitalization information and evaluated patients by subjective and objective items, such as UW-Qol scale, CT data. RESULTS: The operation time of Navigation Group was higher significantly than Control Group (10.36 ± 1.87vs9.00 ± 1.34 h).There were no significant differences in early postoperative complications. The Qol score of appearance, motion, anxiety were higher significantly in Navigation Group. The CT results showed that mandibular angle deviation and chin deflection of Navigation Group were better significantly than Control Group (1.72 ± 1.29° vs 3.69 ± 1.67°, 2.45 ± 1.39 vs 5.19 ± 2.13 mm). CONCLUSIONS: The digital navigation can improve FFF's accuracy in mandibular reconstruction. It doesn't significantly increase complications. The digital navigation's installation and operation methods should be simplified to shorter operation time and expand its application.