Renal Safety of Sacubitril/Valsartan: A Meta-Analysis of Randomized Controlled Trials.

ABSTRACT: As a first-line therapy, sacubitril/valsartan (S/V) plays a significant role in the treatment of heart failure. However, its effect on renal function is still uncertain. We searched PubMed, EMBASE, the Cochrane Library, and Clinical Trials for randomized controlled trials to evaluate the effect of S/V on renal function in patients. The results are reported as the mean difference, relative ratio, and 95% confidence intervals. A total of 13 randomized controlled trials were included (19,367 patients). Among them, 11 studies focused on patients with heart failure, 1 on patients with acute myocardial infarction, and 1 on patients with chronic kidney disease. We found that fewer worsening renal function events, elevated creatine level events, and severe hyperkalemia events (blood potassium >6.0 mmol/L) occurred in the S/V group than those in the renin-angiotensin-aldosterone system inhibitor (RASi) group. The estimated glomerular filtration rate decreased in both the S/V group and the RASi group, but the change was more obvious in the RASi group. There was no significant difference in hyperkalemia events (blood potassium >5.5 mmol/L) between the 2 groups. Subgroup analysis showed that with the extension of follow-up time (>6 months), worsening renal function events occurred less frequently in the S/V group than in the RASi group. Existing evidence has shown that S/V is superior to RASi in general renal safety. Perhaps with the prolongation of treatment time, the advantages of S/V are more obvious.

Investigation on the interaction between myricetin and dihydromyricetin with trypsin, α-chymotrypsin, lysozyme by spectroscopy and molecular docking methods.

The interaction between myricetin and dihydromyricetin with trypsin, α-chymotrypsin and lysozyme was investigated using multispectral and molecular docking methods. The results of fluorescence quenching revealed that myricetin and dihydromyricetin could quench the intrinsic fluorescence of three different proteinases through a static quenching procedure. The binding constant and number of binding sites at different temperatures were measured. The thermodynamic parameters obtained at different temperatures showed van der Waals interactions and hydrogen bonds played the main roles in the interaction of myricetin with trypsin and lysozyme, hydrophobic force was dominant both in myricetin with α-chymotrypsin interaction and dihydromyricetin with trypsin and lysozyme interaction, as for the electrostatic forces, it was mainly the driving force in dihydromyricetin binding to α-chymotrypsin. There was non-radiative energy transfer between three proteinases and myricetin or dihydromyricetin with high probability. The microenvironment of trypsin, α-chymotrypsin and lysozyme is changed. The docking studies revealed that myricetin and dihydromyricetin entered the hydrophobic cavity of three proteinases and formed hydrogen bonds. The binding affinity of myricetin or dihydromyricetin is different with the trypsin, α-chymotrypsin and lysozyme due to the different molecular structure.

Prognosis of liver transplantation: Does postoperative ileus matter?

BACKGROUND: Nowadays, liver transplantation has become a main therapy for end-stage liver disease. However, studies show that there are high mortality and severe complications after liver transplantation. Although gastrointestinal dysfunction is a common and major complication after liver transplantation, there was rarely relative research. This study aims to elucidate the factors about ileus after liver transplantation and patients' survival. METHODS: We collected and analyzed the data (n = 318, 2016-2019) from the First Affiliated Hospital of Xi'an Jiaotong University. After excluding cases, a total of 293 patients were included for this study. The subjects were divided into a non-ileus group and an ileus group. We reviewed 38 variables (including preoperative, operative and postoperative relative factors). Additionally, other complications after liver transplantation and survival data were compared between two groups. RESULTS: Of the 293 patients, 23.2% (n = 68) experienced postoperative ileus. Ileus patients were not different with non-ileus patients in preoperative, operative and postoperative factors. HBV-positive patients with ileus had a lower MELD score (P = 0.025), and lower postoperative total bilirubin was correlated with ileus (P = 0.049). Besides, Child-Pugh score of HCC patients with ileus was low (P = 0.029). The complications after liver transplantation were not different between two groups. Compared with the patients without ileus, the patients with ileus had a higher mortality rate. CONCLUSION: According to our research, ileus-patients had a lower 1-year survival rates. The preoperative MELD score and postoperative total bilirubin of HBV-positive patients with ileus were lower, and Child-Pugh score of HCC patients with ileus was also lower.

Molecular characterization and functional analysis of TRAF6 in the spotted sea bass (Lateolabrax maculatus).

Tumor necrosis factor receptor-associated factor 6 (TRAF6) is a crucial adapter protein in the toll-like receptor signaling pathway that triggers downstream molecules involved in innate immunity. Although TRAF6 has been well studied in mammals, the molecular information and function of TRAF6 in fish is still limited. Here, we identified and analyzed a TRAF6 homolog (LmTRAF6) from the spotted sea bass (Lateolabrax maculatus). Similar to its counterparts in mammals and other fish species, LmTRAF6 shares the domain topology containing one N-terminal RING, two TRAF-type zinc fingers, a coiled-coil region and a C-terminal MATH domain. Despite a sequence similarity of 60% with mammalian TRAF6s, LmTRAF6 shares higher similarities with teleost homologs (~68%-93%). The coding region of LmTRAF6 gene contains seven exons and six introns, which is consistent to the genetic organization in grouper and rock bream, but not in zebrafish, common carp and tetrapods (the sixth intron was lost resulting in a combined exon). Quantitative real-time polymerase chain reaction analysis revealed that LmTRAF6 transcripts were ubiquitously expressed in all tested tissues and upregulated after Vibrio. harveyi and S. agalactiae infection. LmTRAF6 could assist HEK293T cells to survive by inhibiting apoptosis under both V. harveyi and S. agalactiae stimulation. Intracellular localization showed that LmTRAF6 was localized mainly in the cytoplasm. Overexpression of wild-type (WT) LmTRAF6 and the truncated form of △MATH increased the ability of NF-κB in HEK293T cells, whereas truncations, including the △RING and △coiled-coil domain, did not significantly activate NF-κB, indicating that the RING finger and coiled-coil domain play crucial roles in downstream signal transduction. In addition, overexpression of LmTRAF6-WT significantly increased the activation of NF-κB in HEK293T cells under V. harveyi and S. agalactiae stimulation. These results suggest that LmTRAF6 activates NF-κB and plays a potential role in the immune defense system against bacterial infection.

Online solid-phase extraction with high-performance liquid chromatography and mass spectrometry for the determination of five tannins in traditional Chinese medicine injections.

A rapid analytical method based on online solid-phase extraction with high-performance liquid chromatography and mass spectrometry has been established and applied to the determination of tannin compounds that may cause adverse effects in traditional Chinese medicine injections. Different solid-phase extraction sorbents have been compared and the elution buffer was optimized. The performance of the method was verified by evaluation of recovery (≥40%), repeatability (RSD ≤ 6%), linearity (r(2) ≥ 0.993), and limit of quantification (≤0.35 µg/mL). Five tannin compounds, gallic acid, cianidanol, gallocatechin gallate, ellagic acid, and penta-O-galloylglucose, were identified with concentrations ranging from 3.1-37.4 µg/mL in the analyzed traditional Chinese medicine injections.

The effect of the left stellate ganglion on sympathetic neural remodeling of the left atrium in rats following myocardial infarction.

BACKGROUND: The neural remodeling of the atrium plays an important role in the initiation of atrial fibrillation after myocardial infarction (MI); however, the effects of the left stellate ganglion (LSG) on the neural remodeling of the atrium remain incompletely understood. Thus, this study investigated the mechanism by which the LSG mediates sympathetic neural remodeling of the left atrium (LA) in rats after MI. METHODS: Sixty rats were randomly divided into a Sham group and an MI group. The expression levels of growth-associated protein-43 (GAP43) and nerve growth factor (NGF) messenger ribonucleic acid (mRNA) were measured by reverse transcription polymerase chain reaction. Immunohistochemistry was used to detect the distribution and density of GAP43- and NGF-positive nerves. The expression levels of the proteins were quantified by Western blotting. RESULTS: Compared with the Sham group, GAP43 mRNA expression in the LSG was increased in the MI group (P < 0.01), but not significantly increased in the LA. Immunohistochemical analysis demonstrated that in both the LSG and the LA, the mean densities of GAP43- and NGF-positive nerves in the MI group were increased (P < 0.01). In both the LSG and the LA, the protein levels of GAP43 and NGF in the MI group were increased relative to the Sham group (P < 0.01). CONCLUSIONS: The increased levels of NGF and GAP43 proteins can induce sympathetic nerve hyperinnervation in the LSG and the LA after MI. The increased GAP43 proteins in the LA, which may have been transported from the LSG, accelerated LA sympathetic neural remodeling in rats.

Autonomic neural remodeling of the pulmonary vein-left atrium junction in a prolonged right atrial pacing canine model.

BACKGROUND: No evidence has been presented to show whether autonomic neural remodeling occurs in pulmonary vein-left atrium (PV-LA) junction and what an important role it may play in AF. This study aims to find out these issues in a prolonged rapid atrial pacing canine model. METHODS: Twelve healthy mongrel dogs were randomly divided into two groups, six in each: the paced group underwent rapid right atrial pacing at 400 beats per minute for 4 weeks, and the control group was not paced. The effective refractory period (ERP) of left superior pulmonary vein-left atrium (LSPV-LA) junction in all animals was determined immediately after 4 weeks. Tissues were removed from 1 cm around all PV-LA junctions. Immunohistochemical staining and western blotting were performed to examine the expression of tyrosine hydroxylase (TH) and choline acetyltransferase (ChAT). RESULTS: Compared with the control group, ERP of LSPV-LA junction of the paced group was significantly shortened ([83.33 ± 16.33] ms vs [111.67 ± 20.41] ms, P < 0.05). Spontaneous atrial fibrillation developed in two animals in the paced group, but in none of the control group. Immunohistochemistry showed that the average density and heterogeneity of both TH- and ChAT-positive nerves at LSPV-LA junction in the paced group were significantly higher compared to the control group (P < 0.01). Western blotting showed that the expression of TH and ChAT at four PV-LA junctions in the paced group also increased markedly compared with the control group (P < 0.01). CONCLUSION: Autonomic neural remodeling did exist in PV-LA junction after prolonged atrial pacing, which may contribute to the initiation of atrial fibrillation and be significant in its treatment by radiofrequency catheter ablation.

Theory of Lipid Metabolism Disorders in Rhinitis and Asthma (Lipid Droplets).

Lipid droplets are important for the storage of neutral lipids in cells; moreover, they participate in a variety of activities in cells and are multifunctional organelles. In the past few decades, lipid droplets have been extensively studied and found to play important roles in cellular energy balance, signal regulation and metabolic regulation. In particular, the formation and function of lipid droplets in adipocytes and mast cells have received much attention. This article reviews the formation, structure and function of lipid droplets in mast cells and elaborates on the relationship between lipid droplets and both adipocyte metabolism and mast cell-mediated allergic inflammation, to provide ideas for the treatment of allergic inflammation by targeting lipid droplets. This study provides important evidence for the role of lipid metabolism disorders in rhinitis and asthma.

[Communication subsystem design of tele-screening system for diabetic retinopathy].

A design scheme of a tele-screening system for diabetic retinopathy (DR) has been proposed, especially the communication subsystem. The scheme uses serial communication module consisting of ARM 7 microcontroller and relays to connect remote computer and fundus camera, and also uses C++ programming language based on MFC to design the communication software consisting of therapy and diagnostic information module, video/audio surveillance module and fundus camera control module. The scheme possesses universal property in some remote medical treatment systems which are similar to the system.

LncRNA MEG3 inhibits the proliferation and migration abilities of colorectal cancer cells by competitively suppressing MiR-31 and reducing the binding of MiR-31 to target gene SFRP1.

To explore the potential mechanism of long-chain non-coding ribonucleic acid (lncRNA) maternal expression gene 3 (MEG3) in colorectal cancer (CRC). The relationship between MEG3 and miR-31 was detected by dual-luciferase assay. Quantitative polymerase chain reaction was utilized to determine the expression of MEG3 in CRC cell lines. Cell Counting Kit-8 assay was performed to detect cell proliferation. Transwell, cell scratch wound assay, and monoclonal proliferation assay were used to detect the proliferation, migration, and invasion of cells. In addition, cell motility was evaluated by detecting the expression of cellular pseudopodia protein α-actinin via immunofluorescence assay, and cell proliferation and motility were judged by determining the expressions of Ki-67, MMP2 and MMP9 via Western blotting. The effect of MEG3 and miR-31 on the development of colorectal cancer was verified by nude mouse tumor-bearing assay and HE staining. Transient transfection with MEG3 overexpression plasmid revealed that MEG3 inhibited the proliferation and motility of cells. The results of dual-luciferase assay showed that MEG3 could specifically inhibit the expression of miR-31, which inhibits the development of colorectal cancer. Transwell, cell scratch wound assay, and monoclonal proliferation experiment showed that miR-31 enhanced cell proliferation, migration and invasion. MEG3 overexpression plasmid was capable of reversing the proliferation and motility of CRC cells enhanced by miR-31. MEG3 can inhibit the proliferation and motility of CRC cells by competitively suppressing the binding of miR-31 to the target gene SFRP1, thus playing an inhibitory role in the pathogenesis of CRC.

A Yeast Cell Wall Derived Hybrid Hydrogel with Photothermal and Immune Combined Modality Therapy for Enhanced Anti-Melanoma Efficacy.

Introduction: The effect of traditional treatment for melanoma is quite limited, especially for its recurrence. As the major components of yeast cell wall, chitin and ß-glucan exhibit good immune activation effect and are promising candidates for adjuvant. Therefore, melanoma cell membrane (CM) and indocyanine green (ICG) was loaded in a chitin and ß-glucan hybrid hydrogel to achieve an enhanced anti-melanoma therapy. Methods: The novel hybrid hydrogel was prepared, and its physicochemical properties were examined. Its effect towards melanoma prevention and treatment was evaluated via a melanoma-bearing mice model. Results: The CM-ICG-hybrid hydrogel was successfully prepared with excellent injectability, self-healing, drug loading, rheological, in vitro and in vivo photothermal stability, and retention properties. It also exhibited good cellular and in vivo safety profiles. In the primary melanoma mice model, it quickly ablated the in-situ melanoma, effectively inhibited the tumor growth, increased the survival rate of melanoma-bearing mice, and increased the level of IFN-γ and TNF-α. In the distal secondary melanoma model, it efficiently prevented the reoccurrence of melanoma and activated the memory T cells. In both models, a synergistic effect of photothermal therapy and immune therapy was found. The hydrogel effectively recruited CD3+ CD4+ T cells and CD3+ CD8+ T cells, inhibited the proliferation of melanoma cells, and induced the apoptosis of melanoma cells. Conclusion: The hybrid hydrogel was successfully prepared, and it showed excellent efficacy towards melanoma prevention and treatment due to its efficient tumor ablation and immune activation capability.

Automatic pulmonary artery-vein separation in CT images using a twin-pipe network and topology reconstruction.

Background: With the wide application of CT scanning, the separation of pulmonary arteries and veins (A/V) based on CT images plays an important role for assisting surgeons in preoperative planning of lung cancer surgery. However, distinguishing between arteries and veins in chest CT images remains challenging due to the complex structure and the presence of their similarities. Methods: We proposed a novel method for automatically separating pulmonary arteries and veins based on vessel topology information and a twin-pipe deep learning network. First, vessel tree topology is constructed by combining scale-space particles and multi-stencils fast marching (MSFM) methods to ensure the continuity and authenticity of the topology. Second, a twin-pipe network is designed to learn the multiscale differences between arteries and veins and the characteristics of the small arteries that closely accompany bronchi. Finally, we designed a topology optimizer that considers interbranch and intrabranch topological relationships to optimize the results of arteries and veins classification. Results: The proposed approach is validated on the public dataset CARVE14 and our private dataset. Compared with ground truth, the proposed method achieves an average accuracy of 90.1% on the CARVE14 dataset, and 96.2% on our local dataset. Conclusions: The method can effectively separate pulmonary arteries and veins and has good generalization for chest CT images from different devices, as well as enhanced and noncontrast CT image sequences from the same device.

Mesoporous Hollow Manganese Doped Ceria Nanoparticle for Effectively Prevention of Hepatic Ischemia Reperfusion Injury.

Introduction: Hepatic ischemia-reperfusion injury (HIRI) is the main reason for liver dysfunction or failure after liver resection and liver transplantation. As excess accumulation of reactive oxygen species (ROS) is the leading factor, ceria nanoparticle, a cyclic reversible antioxidant, is an excellent candidate for HIRI. Methods: Manganese doped mesoporous hollow ceria nanoparticles (MnOx-CeO2 NPs) were prepared, and the physicochemical characteristics, such as particle size, morphology, microstructure, etc. were elucidated. The in vivo safety and liver targeting effect were examined after i.v. injection. The anti-HIRI was determined by a mouse HIRI model. Results: MnOx-CeO2 NPs with 0.40% Mn doped exhibited the strongest ROS-scavenging capability, which may due to the increased specific surface area and surface oxygen concentration. The nanoparticles accumulated in the liver after i.v. injection and exhibited good biocompatibility. In the HIRI mice model, MnOx-CeO2 NPs significantly reduced the serum ALT and AST level, decreased the MDA level and increased the SOD level in the liver, prevent pathological damages in the liver. Conclusion: MnOx-CeO2 NPs were successfully prepared and it could significantly inhibit the HIRI after i.v. injection.

Radiomics approach with deep learning for predicting T4 obstructive colorectal cancer using CT image.

OBJECTIVES: Patients with T4 obstructive colorectal cancer (OCC) have a high mortality rate. Therefore, an accurate distinction between T4 and T1-T3 (NT4) in OCC is an important part of preoperative evaluation, especially in the emergency setting. This paper introduces three models of radiomics, deep learning, and deep learning-based radiomics to identify T4 OCC. METHODS: We established a dataset of computed tomography (CT) images of 164 patients with pathologically confirmed OCC, from which 2537 slides were extracted. First, since T4 tumors penetrate the bowel wall and involve adjacent organs, we explored whether the peritumoral region contributes to the assessment of T4 OCC. Furthermore, we visualized the radiomics and deep learning features using the t-distributed stochastic neighbor embedding technique (t-SNE). Finally, we built a merged model by fusing radiomic features with deep learning features. In this experiment, the performance of each model was evaluated by the area under the receiver operating characteristic curve (AUC). RESULTS: In the test cohort, the AUC values predicted by the radiomics model in the dilated region of interest (dROI) was 0.770. And the AUC value of the deep learning model with the patches extended 20-pixel reached 0.936. Combining the characteristics of radiomics and deep learning, our method achieved an AUC value of 0.947 in the T4 and non-T4 (NT4) classification, and increased the AUC value to 0.950 after the addition of clinical features. CONCLUSION: The prediction results of our merged model of deep learning radiomics outperformed the deep learning model and significantly outperformed the radiomics model. The experimental results demonstrate that combining the peritumoral region improves the prediction performance of the radiomics model and the deep learning model.

Learning multi-view and centerline topology connectivity information for pulmonary artery-vein separation.

BACKGROUND: Automatic pulmonary artery-vein separation has considerable importance in the diagnosis and treatment of lung diseases. However, insufficient connectivity and spatial inconsistency have always been the problems of artery-vein separation. METHODS: A novel automatic method for artery-vein separation in CT images is presented in this work. Specifically, a multi-scale information aggregated network (MSIA-Net) including multi-scale fusion blocks and deep supervision, is proposed to learn the features of artery-vein and aggregate additional semantic information, respectively. The proposed method integrates nine MSIA-Net models for artery-vein separation, vessel segmentation, and centerline separation tasks along with axial, coronal, and sagittal multi-view slices. First, the preliminary artery-vein separation results are obtained by the proposed multi-view fusion strategy (MVFS). Then, centerline correction algorithm (CCA) is used to correct the preliminary results of artery-vein separation by the centerline separation results. Finally, the vessel segmentation results are utilized to reconstruct the artery-vein morphology. In addition, weighted cross-entropy and dice loss are employed to solve the class imbalance problem. RESULTS: We constructed 50 manually labeled contrast-enhanced computed CT scans for five-fold cross-validation, and experimental results demonstrated that our method achieves superior segmentation performance of 97.7%, 85.1%, and 84.9% on ACC, Pre, and DSC, respectively. Additionally, a series of ablation studies demonstrate the effectiveness of the proposed components. CONCLUSION: The proposed method can effectively solve the problem of insufficient vascular connectivity and correct the spatial inconsistency of artery-vein.

Electrochemical behavior of o-sec-butylphenol at glassy carbon electrode modified with multiwalled carbon nanotubes and 1-butyl-3-methylimidazolium hexafluorophosphate.

A sensitive electrochemical sensor based on immobilized multiwalled carbon nanotubes (MWCNTs) and 1-butyl-3-methylimidazolium hexafluorophosphate (BMIM·PF(6)) on a glassy carbon electrode (GCE) for o-sec-butylphenol (osBP) was proposed. The electro-oxidation behavior was studied, the experimental conditions were optimized and kinetic parameters were calculated. The results indicated that this electrochemical sensor has the advantages of fast electron-transfer rate, minimal fouling of electrodes, high sensitivity and stability for o-sec-butylphenol. Upon comparison with a glassy carbon electrode, this senor would effectively minimize the over-potential and increase the electrochemical response to o-sec-butylphenol. Under the optimum conditions, the peak current was linear to the osBP concentration range from 1 × 10(-7) to 2.5 × 10(-5) M with the detection limit of 8.65 × 10(-9) M (S/N = 3). The proposed method was applied to the determination of spiked water samples with satisfactory results.

Cucurbitacin E reduces the cognitive dysfunction induced by sevoflurane in rats by regulating NF-κB pathway.

BACKGROUND: Perioperative neurocognitive disorders (PND) occur frequently and refer to alterations in cognitive function after surgery, especially in elderly patients. PND is characterized as abnormalities of learning, memory, language, and emotions. Cucurbitacin E has been reported to possess various pharmacological properties, including anticancer, antiviral, and anti-inflammatory effects. In this study, we investigated whether cucurbitacin E could alleviate sevoflurane-induced cognitive dysfunction in rats. METHODS: Sprague-Dawley male rats (~6 weeks old) were randomly assigned to three groups: the control group, the Sevoflurane group, and the Sevoflurane + Cucurbitacin E group. Subsequently, the cognitive dysfunction of the rats was evaluated through the morris water maze test. Hematoxylin and eosin (HE) staining was used to measure the pathological change in brain tissues. Enzyme-linked immunosorbent assay (ELISA) kits were used for determinations of S-100 calcium binding protein B (S-100ß) and neuron-specific enolase (NSE) and cytokine. Cell apoptosis was analyzed by TdT-Mediated Nick-End Labeling (TUNEL) staining. Protein levels were confirmed by Western blotting. RESULTS: Cucurbitacin E relieved brain injury in rats induced by sevoflurane. Cucurbitacin E alleviated sevoflurane-induced S-100ß and NSE levels. Additionally, the Morris water maze task revealed that cucurbitacin E attenuated cognition impairment in sevoflurane-induced rats. Sevoflurane increased levels of IL-6, TNF-α and IL-1ß levels, and decreased the level of IL-10. However, cucurbitacin E exhibited opposite effects on these cytokines, which were induced by sevoflurane. Furthermore, cucurbitacin E inhibited sevoflurane-induced neuron apoptosis and NF-κB pathway in rats. CONCLUSION: These findings indicate that cucurbitacin E can improve sevoflurane-induced cognitive dysfunction in rats by regulating NF-κB pathway, which provided a new strategy for PND treatment.

A two-stage network with prior knowledge guidance for medullary thyroid carcinoma recognition in ultrasound images.

PURPOSE: Accurate recognition of medullary thyroid carcinoma (MTC) is of great importance in medical diagnosis, as MTC is rare but second-most malignant thyroid cancers with a high case-fatality ratio.1 But there is a lower recognition rate on distinguishing MTC from other thyroid nodules in ultrasound images, even by experienced experts. This paper introduces the computer-aided method to tackle the challenge of recognizing MTC from ultrasound images, including limited MTC samples, and ambiguities among MTC, benign nodules, and papillary thyroid carcinoma (PTC). METHODS: The recognition of MTC based on large MTC samples of ultrasound images has never been explored, as only one existing work presented a relevant dataset with a limited 21 MTC samples. This study proposes a novel method for primarily differentiating MTC samples from benign nodules and PTC that is the most common thyroid cancer. Our method is a two-stage schema with two important components including a cascaded coarse-to-fine segmentation network and a knowledge-based classification network. The cascaded coarse-to-fine segmentation network incorporates two U-Net++ networks for improving the segmentation results of thyroid nodules. Meanwhile, our knowledge-based classification network extracts and fuses semantic features of solid tissues and calcification for better recognizing the segmented nodules from the ultrasound images. In our experiments, dice similarity coefficient (DSC), intersection over union (IoU), precision, recall, and Hausdorff distance (HD) are adopted for evaluating the segmentation results of thyroid nodules, and accuracy, precision, recall, and F1-score are used for classification evaluation. RESULTS: We present a well-annotated dataset including samples of 248 MTC, 240 benign nodules, and 239 PTC. For thyroid nodule segmentation, our designed cascaded segmentation network attains values of 0.776 DSC, 0.689 IoU, 0.778 precision, and 0.821 recall, respectively. By incorporating prior knowledge, our method achieves a mean accuracy of 82.1% in classifying thyroid nodules of MTC, PTC, and benign ones. Especially, our method gains the higher performance in recognizing MTC with an accuracy of 86.8%, compared to nearly 70% diagnosis accuracy of experienced doctors. The experimental results on our Fujian Provincial Hospital dataset further validate the efficiency of our proposed method. CONCLUSIONS: Our proposed two-stage method incorporates pipelines of thyroid nodules segmentation and classification of MTC, individually. Quantitative and qualitative results indicate that our proposed model achieves accurate segmentation of thyroid nodules. The results also validate that our learning-based framework facilitates the recognition of MTC, which gains better classification accuracy than experienced doctors.

Coarse-to-fine airway segmentation using multi information fusion network and CNN-based region growing.

BACKGROUND AND OBJECTIVES: Automatic airway segmentation from chest computed tomography (CT) scans plays an important role in pulmonary disease diagnosis and computer-assisted therapy. However, low contrast at peripheral branches and complex tree-like structures remain as two mainly challenges for airway segmentation. Recent research has illustrated that deep learning methods perform well in segmentation tasks. Motivated by these works, a coarse-to-fine segmentation framework is proposed to obtain a complete airway tree. METHODS: Our framework segments the overall airway and small branches via the multi-information fusion convolution neural network (Mif-CNN) and the CNN-based region growing, respectively. In Mif-CNN, atrous spatial pyramid pooling (ASPP) is integrated into a u-shaped network, and it can expend the receptive field and capture multi-scale information. Meanwhile, boundary and location information are incorporated into semantic information. These information are fused to help Mif-CNN utilize additional context knowledge and useful features. To improve the performance of the segmentation result, the CNN-based region growing method is designed to focus on obtaining small branches. A voxel classification network (VCN), which can entirely capture the rich information around each voxel, is applied to classify the voxels into airway and non-airway. In addition, a shape reconstruction method is used to refine the airway tree. RESULTS: We evaluate our method on a private dataset and a public dataset from EXACT09. Compared with the segmentation results from other methods, our method demonstrated promising accuracy in complete airway tree segmentation. In the private dataset, the Dice similarity coefficient (DSC), Intersection over Union (IoU), false positive rate (FPR), and sensitivity are 93.5%, 87.8%, 0.015%, and 90.8%, respectively. In the public dataset, the DSC, IoU, FPR, and sensitivity are 95.8%, 91.9%, 0.053% and 96.6%, respectively. CONCLUSION: The proposed Mif-CNN and CNN-based region growing method segment the airway tree accurately and efficiently in CT scans. Experimental results also demonstrate that the framework is ready for application in computer-aided diagnosis systems for lung disease and other related works.