A novel difficult-to-segment samples focusing network for oral CBCT image segmentation.

Using deep learning technology to segment oral CBCT images for clinical diagnosis and treatment is one of the important research directions in the field of clinical dentistry. However, the blurred contour and the scale difference limit the segmentation accuracy of the crown edge and the root part of the current methods, making these regions become difficult-to-segment samples in the oral CBCT segmentation task. Aiming at the above problems, this work proposed a Difficult-to-Segment Focus Network (DSFNet) for segmenting oral CBCT images. The network utilizes a Feature Capturing Module (FCM) to efficiently capture local and long-range features, enhancing the feature extraction performance. Additionally, a Multi-Scale Feature Fusion Module (MFFM) is employed to merge multiscale feature information. To further improve the loss ratio for difficult-to-segment samples, a hybrid loss function is proposed, combining Focal Loss and Dice Loss. By utilizing the hybrid loss function, DSFNet achieves 91.85% Dice Similarity Coefficient (DSC) and 0.216 mm Average Symmetric Surface Distance (ASSD) performance in oral CBCT segmentation tasks. Experimental results show that the proposed method is superior to current dental CBCT image segmentation techniques and has real-world applicability.

CTA-UNet: CNN-transformer architecture UNet for dental CBCT images segmentation.

In view of the limitations of current deep learning models in segmenting dental cone-beam computed tomography (CBCT) images, specifically dealing with complex root morphological features, fuzzy boundaries between tooth roots and alveolar bone, and the need for costly annotation of dental CBCT images. We collected dental CBCT data from 200 patients and annotated 45 of them for network training, and proposed a CNN-Transformer Architecture UNet network, which combines the advantages of CNN and Transformer. The CNN component effectively extracts local features, while the Transformer captures long-range dependencies. Multiple spatial attention modules were included to enhance the network's ability to extract and represent spatial information. Additionally, we introduced a novel Masked image modeling method to pre-train the CNN and Transformer modules simultaneously, mitigating limitations due to a smaller amount of labeled training data. Experimental results demonstrate that the proposed method achieved superior performance (DSC of 87.12%, IoU of 78.90%, HD95 of 0.525 mm, ASSD of 0.199 mm), and provides a more efficient and effective approach to automatically and accurately segment dental CBCT images, has real-world applicability in orthodontics and dental implants.

Coating 3D-Printed Bioceramics with Histatin Promotes Adhesion and Osteogenesis of Stem Cells.

Mesenchymal stem cell and 3D printing-based bone tissue engineering present a promising technique to repair large-volume bone defects. Its success is highly dependent on cell attachment, spreading, osteogenic differentiation, and in vivo survival of stem cells on 3D-printed scaffolds. In this study, we applied human salivary histatin-1 (Hst1) to enhance the interactions of human adipose-derived stem cells (hASCs) on 3D-printed ß-tricalcium phosphate (ß-TCP) bioceramic scaffolds. Fluorescent images showed that Hst1 significantly enhanced the adhesion of hASCs to both bioinert glass and 3D-printed ß-TCP scaffold. In addition, Hst1 was associated with significantly higher proliferation and osteogenic differentiation of hASCs on 3D-printed ß-TCP scaffolds. Moreover, coating 3D-printed ß-TCP scaffolds with histatin significantly promotes the survival of hASCs in vivo. The ERK and p38 but not JNK signaling was found to be involved in the superior adhesion of hASCs to ß-TCP scaffolds with the aid of Hst1. In conclusion, Hst1 could significantly promote the adhesion, spreading, osteogenic differentiation, and in vivo survival of hASCs on 3D-printed ß-TCP scaffolds, bearing a promising application in stem cell/3D printing-based constructs for bone tissue engineering.

Cationic antimicrobial peptide NRC-03 induces oral squamous cell carcinoma cell apoptosis via CypD-mPTP axis-mediated mitochondrial oxidative stress.

Pleurocidin-family cationic antimicrobial peptide NRC-03 exhibits potent and selective cytotoxicity towards cancer cells. However, the anticancer effect of NRC-03 in oral squamous cell carcinoma (OSCC) and the molecular mechanism of NRC-03 induced cancer cell death is still unclear. This study focused to investigate mitochondrial oxidative stress-mediated altered mitochondrial function involved in NRC-03-induced apoptosis of OSCC cells. NRC-03 entered the OSCC cells more easily than that of normal cells and bound to mitochondria as well as the nucleus, causing cell membrane blebbing, mitochondria swelling, and DNA fragmentation. NRC-03 induced high oxygen consumption, reactive oxygen species (ROS) release, mitochondrial dysfunction, and apoptosis in OSCC cells. Non-specific antioxidant N-acetyl-l-cysteine (NAC), or mitochondria-specific antioxidant mitoquinone (MitoQ) alleviated NRC-03-induced apoptosis and mitochondrial dysfunction indicated that NRC-03 exerts a cytotoxic effect in cancer cells via inducing cellular and mitochondrial oxidative stress. Moreover, the expression of cyclophilin D (CypD), the key component of mitochondrial permeability transition pore (mPTP), was upregulated in NRC-03-treated cancer cells. Blockade of CypD by siRNA-mediated depletion or pharmacological inhibitor cyclosporine A (CsA) significantly suppressed NRC-03-induced mitochondrial oxidative stress, mitochondrial dysfunction, and apoptosis. NRC-03 also activated MAPK/ERK and NF-κB pathways. Importantly, intratumoral administration of NRC-03 inhibited the growth of CAL-27 cells-derived tumors on xenografted animal models. Taken together, our study indicates that NRC-03 induces apoptosis in OSCC cells via the CypD-mPTP axis mediated mitochondrial oxidative stress.

Big Data Fusion Method Based on Internet of Things Collection.

With the development of information collection technology, the data that people need to deal with are also increasing, which brings problems such as isomerization of data types, poor data quality, and fast data generation speed. At present, as an important method of data fusion technology, data fusion method based on deep learning has become an effective way of data fusion under the background of big data, which has important research significance. There is a problem with heterogeneous data types between time series data and text data, and it is difficult to fuse them effectively by traditional data fusion methods. In order to make full use of the information contained in text data and improve the accuracy of time series prediction, this paper proposes a data fusion model based on FC-SAE. In this model, GloVe and CNN are used to extract the features of text data, FC neural network is used to extract the potential features of time series data, and then, the SEA model is used to fuse the data, which fully discovers the relationship between data and greatly improves the prediction accuracy.

Neuro-regenerative imidazole-functionalized GelMA hydrogel loaded with hAMSC and SDF-1α promote stem cell differentiation and repair focal brain injury.

Brain tissues that are severely damaged by traumatic brain injury (TBI) is hardly regenerated, which leads to a cavity or a repair with glial scarring. Stem-cell therapy is one viable option to treat TBI-caused brain tissue damage, whose use is, whereas, limited by the low survival rate and differentiation efficiency of stem cells. To approach this problem, we developed an injectable hydrogel using imidazole groups-modified gelatin methacrylate (GelMA-imid). In addition, polydopamine (PDA) nanoparticles were used as carrier for stromal-cell derived factor-1 (SDF-1α). GelMA-imid hydrogel loaded with PDA@SDF-1α nanoparticles and human amniotic mesenchymal stromal cells (hAMSCs) were injected into the damaged area in an in-vivo cryogenic injury model in rats. The hydrogel had low module and its average pore size was 204.61 ± 41.41 nm, which were suitable for the migration, proliferation and differentiation of stem cells. In-vitro cell scratch and differentiation assays showed that the imidazole groups and SDF-1α could promote the migration of hAMSCs to injury site and their differentiation into nerve cells. The highest amount of nissl body was detected in the group of GelMA-imid/SDF-1α/hAMSCs hydrogel in the in-vivo model. Additionally, histological analysis showed that GelMA-imid/SDF-1α/hAMSCs hydrogel could facilitate the regeneration of regenerate endogenous nerve cells. In summary, the GelMA-imid/SDF-1α/hAMSCs hydrogel promoted homing and differentiation of hAMSCs into nerve cells, and showed great application potential for the physiological recovery of TBI.