Comparison of SOX and CAPOX in patients with advanced gastric cancer after laparoscopic D2 gastrectomy: A randomized controlled trial.

BACKGROUND: Optimal adjuvant chemotherapy after laparoscopic surgery in gastric cancer (GC) patients is still undefined. We aimed to evaluate the efficacy of S-1 plus oxaliplatin (SOX) and capecitabine plus oxaliplatin (CAPOX) in patients with GC after laparoscopic gastrectomy. METHODS: A non-inferiority randomized controlled clinical trial was performed in China. Patients with advanced GC who underwent laparoscopic D2 gastrectomy were randomly assigned to receive SOX and CAPOX regimens. RESULTS: In total, 191 patients were screened between May 2018 and June 2019, and 140 (73.3%) were included in the modified intent-to-treat analysis (mITT), of whom 69 and 71 were assigned to the SOX and CAPOX groups, respectively. The SOX group had similar 3-year overall survival (OS) and disease-free survival to the CAPOX group. Subgroup analysis revealed significantly better OS in the SOX group for male patients ([HR] = 0.395; 95% [CI], 0.153-1.019; p = 0.045), age >60 (HR = 0.219; 95% [CI], 0.064-0.753; p = 0.016), tumors in the gastric antrum (HR = 0.273; 95% [CI], 0.076-0.981; p = 0.047), and moderately differentiated tumors (HR = 0.338; 95% [CI], 0.110-1.041; p = 0.041). There were no significant differences observed in terms of adverse events and recurrence patterns between the two groups. CONCLUSION: Adjuvant SOX was non-inferior to CAPOX treatments for patients with GC who underwent curative laparoscopic D2 gastrectomy. For male patients, aged >60 years, tumors in the gastric antrum, and moderately differentiated tumors, adjuvant SOX may achieve an improvement compared with CAPOX.

NuSEA: Nuclei Segmentation With Ellipse Annotations.

OBJECTIVE: Nuclei segmentation is a crucial pre-task for pathological microenvironment quantification. However, the acquisition of manually precise nuclei annotations for improving the performance of deep learning models is time-consuming and expensive. METHODS: In this paper, an efficient nuclear annotation tool called NuSEA is proposed to achieve accurate nucleus segmentation, where a simple but effective ellipse annotation is applied. Specifically, the core network U-Light of NuSEA is lightweight with only 0.86 M parameters, which is suitable for real-time nuclei segmentation. In addition, an Elliptical Field Loss and a Texture Loss are proposed to enhance the edge segmentation and constrain the smoothness simultaneously. RESULTS: Extensive experiments on three public datasets (MoNuSeg, CPM-17, and CoNSeP) demonstrate that NuSEA is superior to the state-of-the-art (SOTA) methods and better than existing algorithms based on point, rectangle, and text annotations. CONCLUSIONS: With the assistance of NuSEA, a new dataset called NuSEA-dataset v1.0, encompassing 118,857 annotated nuclei from the whole-slide images of 12 organs is released. The codes and the new dataset are publicly available at https://github.com/dreambamboo/NuSEA/. SIGNIFICANCE: NuSEA provides a rapid and effective annotation tool for nuclei in histopathological images, benefiting future explorations in deep learning algorithms.

Anaerobic ammonium oxidation coupled with sulfate reduction links nitrogen with sulfur cycle.

Sulfate-dependent ammonium oxidation (Sulfammox) is a critical process linking nitrogen and sulfur cycles. However, the metabolic pathway of microbes driven Sulfammox is still in suspense. The study demonstrated that ammonium was not consumed with sulfate as the sole electron acceptor during long-term enrichment, probably due to inhibition from sulfide accumulation, while ammonium was removed at âˆ¼ 10 mg N/L/d with sulfate and nitrate as electron acceptors. Ammonium and sulfate were converted into nitrogen gas, sulfide, and elemental sulfur. Sulfammox was mainly performed by Candidatus Brocadia sapporoensis and Candidatus Brocadia fulgida, both of which encoded ammonium oxidation pathway and dissimilatory sulfate reduction pathway. Not sulfide-driven autotrophic denitrifiers but Candidatus Kuenenia stuttgartiensis converted nitrate to nitrite with sulfide. The results of this study reveal the specialized metabolism of Sulfammox bacteria (Candidatus Brocadia sapporoensis and Candidatus Brocadia fulgida) and provide insight into microbial relationships during the nitrogen and sulfur cycles.

Deciphering the formation of granules by n-DAMO and Anammox microorganisms.

Nitrate/nitrite-dependent anaerobic methane oxidation (n-DAMO) process is a promising wastewater treatment technology, but the slow microbial growth rate greatly hinders its practical application. Although high-level nitrogen removal and excellent biomass accumulation have been achieved in n-DAMO granule process, the formation mechanism of n-DAMO granules remains unresolved. To elucidate the role of functional microbes in granulation, this study attempted to cultivate granules dominated by n-DAMO microorganisms and granules coupling n-DAMO with anaerobic ammonium oxidation (Anammox). After long-term operation, dense granules were developed in the two systems where both n-DAMO archaea and n-DAMO bacteria were enriched, whereas granulation did not occur in the other system dominated by n-DAMO bacteria. Extracellular polymeric substances (EPS) measurement indicated the critical role of EPS production in the granulation of n-DAMO process. Metagenomic and metatranscriptomic analyses revealed that n-DAMO archaea and Anammox bacteria were active in EPS biosynthesis, while n-DAMO bacteria were inactive. Consequently, more EPS were produced in the systems containing n-DAMO archaea and Anammox bacteria, leading to the successful development of n-DAMO granules. Furthermore, EPS biosynthesis in n-DAMO systems is potentially regulated by acyl-homoserine lactones and c-di-GMP. These findings not only provide new insights into the mechanism of granule formation in n-DAMO systems, but also hint at potential strategies for management of the granule-based n-DAMO process.

MTCSNet: One-stage learning and two-point labeling are sufficient for cell segmentation.

Deep convolution neural networks have been widely used in medical image analysis, such as lesion identification in whole-slide images, cancer detection, and cell segmentation, etc. However, it is often inevitable that researchers try their best to refine annotations so as to enhance the model performance, especially for cell segmentation task. Weakly supervised learning can greatly reduce the workload of annotations, while there is still a huge performance gap between the weakly and fully supervised learning approaches. In this work, we propose a weakly-supervised cell segmentation method, namely Multi-Task Cell Segmentation Network (MTCSNet), for multi-modal medical images, including pathological, brightfield, fluorescent, phase-contrast and differential interference contrast images. MTCSNet is learnt in a single-stage training manner, where only two annotated points for each cell provide supervision information, and the first one is the centroid, the second one is its boundary. Additionally, five auxiliary tasks are elaborately designed to train the network, including two pixel-level classifications, a pixel-level regression, a local temperature scaling and an instance-level distance regression task, which is proposed to regress the distances between the cell centroid and its boundaries in eight orientations. The experimental results indicate that our method outperforms all state-of-the-art weakly-supervised cell segmentation approaches on public multi-modal medical image datasets. The promising performance also shows that a single-stage learning with two-point labeling approach are sufficient for cell segmentation, instead of fine contour delineation. The codes are available at: https://github.com/binging512/MTCSNet.

Application progress of nanocellulose in food packaging: A review.

With the increasing environmental and ecological problems caused by petroleum-based packaging materials, the focus has gradually shifted to natural resources for the preparation of functional food packaging materials. In addition to biodegradable properties, nanocellulose (NC) mechanical properties, and rich surface chemistry are also fascinating and desired to be one of the most probable green packaging materials. In this review, we firstly introduce the recent progress of novel applications of NC in food packaging, including intelligent packaging, nano(bio)sensors, and nano-paper; secondly, we focus on the modification techniques of NC to summarize the properties (antimicrobial, mechanical, hydrophobic, antioxidant, and so on) that are required for food packaging, to expand the new synthetic methods and application areas. After presenting all the latest advances related to material design and sustainable applications, an overview summarizing the safety of NC is presented to promote a continuous and healthy movement of NC toward the field of truly sustainable packaging.

Anchoring CeF3 nanoparticles on porous carbon nanofibers as self-supporting electrodes for highly sensitive detection of nitrite.

Designing a working electrode is crucial for the reliable electrochemistry detection, which is applied to detect toxic and harmful substances sensitively and rapidly. Here we report the polytetrafluoroethylene decomposition-assisted electrospinning, a combination method for creating nanopore and synthesizing CeF3, to prepare the self-supporting electrode of CeF3 nanoparticles-anchored on porous carbon nanofibers (CeF3/PCNFs) for highly sensitive nitrite detection. The CeF3/PCNFs exhibits remarkable electroactivity toward nitrite detection, featuring a wide concentration range (0.5 µM-6 mM), low detection limit (10 nm) and high sensitivity (2093 µA mM-1 cm-2). It also exhibits excellent selectivity, stability and reproducibility, and powerful reliability for nitrite detection in saliva, pickles, sausages, chips, river water and tap water. This study provides a facile strategy to prepare the metal fluoride-based self-supporting electrode, which overcomes the disadvantages of chemically modified electrodes unstable and poorly reproducible, and is significant for clinical diagnosis, food safety and environmental monitoring.

Phlorizin Limits Bovine Viral Diarrhea Virus Infection in Mice via Regulating Gut Microbiota Composition.

Phlorizin (PHZ) is one of the main pharmacologically active ingredients in Lithocarpus polystachyus. We have previously shown that PHZ inhibits the replication of bovine viral diarrhea virus (BVDV), but the exact antiviral mechanism, especially in vivo, is still unknown. Here, we further confirm that PHZ has good protective effects in BVDV-infected mice. We analyzed BVDV-induced CD3+, CD4+, and CD8+ T cells among peripheral blood lymphocytes and found that PHZ significantly restored their percentage. Metagenomic analyses revealed that PHZ markedly improved the richness and diversity of intestinal microbiota and increased the abundance of potentially health-related microbes (families Lachnosipiraceae, Ruminococcaceae, and Oscillospiraceae). Specifically, the relative abundance of short chain fatty acid (SCFA)-producing bacteria, including Lachnospiraceae_UCG-006, unclassified_f_Ruminococcaceae, Oscillibacter, Intestinimonas, Blautia, and Lachnoclostridium increased significantly after PHZ treatment. Interestingly, BVDV-infected mice that received fecal microbiota from PHZ-treated mice (PHZ-FMT) had a significantly lower viral load in the duodenum and jejunum than untreated mice. Pathological lesions of duodenum and jejunum were also greatly reduced in the PHZ-FMT group, confirming a significant antiviral effect. These findings show that gut microbiota play an important role in PHZ's antiviral activity and suggest that their targeted intervention might be a promising endogenous strategy to prevent and control BVDV.

Deep insights into the biofilm formation mechanism and nitrogen-transformation network in a nitrate-dependent anaerobic methane oxidation biofilm.

Nitrate/nitrite-dependent anaerobic methane oxidation (n-DAMO) process offers a promising solution for simultaneously achieving methane emissions reduction and efficient nitrogen removal in wastewater treatment. Although nitrogen removal at a practical rate has been achieved by n-DAMO biofilm process, the mechanisms of biofilm formation and nitrogen transformation remain to be elucidated. In this study, n-DAMO biofilms were successfully developed in the membrane aerated moving bed biofilm reactor (MAMBBR) and removed nitrate at a rate of 159 mg NO3--N L-1 d-1. The obvious increase in the content of extracellular polymeric substances (EPS) indicated that EPS production was important for biofilm development. n-DAMO microorganisms dominated the microbial community, and n-DAMO bacteria were the most abundant microorganisms. However, the expression of biosynthesis genes for proteins and polysaccharides encoded by n-DAMO archaea was significantly more active compared to other microorganisms, suggesting the central role of n-DAMO archaea in EPS production and biofilm formation. In addition to nitrate reduction, n-DAMO archaea were revealed to actively express dissimilatory nitrate reduction to ammonium and nitrogen fixation. The produced ammonium was putatively converted to dinitrogen gas through the joint function of n-DAMO archaea and n-DAMO bacteria. This study revealed the biofilm formation mechanism and nitrogen-transformation network in n-DAMO biofilm systems, shedding new light on promoting the application of n-DAMO process.

Automatic center identification of electron diffraction with multi-scale transformer networks.

Selected area electron diffraction (SAED) is a widely used technique for characterizing the structure and measuring lattice parameters of materials. An autonomous analytic method has become an urgent demand for the large-scale SAED data produced from in-situ experiments. In this work, we realize the automatic processing for center identification with a proposed deep segmentation model named the multi-scale Transformer (MS-Trans) network. This algorithm enables robust segmentation of the central spots by combining a novel gated axial-attention module and multi-scale feature fusion. The proposed MS-Trans model shows high precision and robustness, enabling autonomous processing of SAED patterns without any prior knowledge. The application on in-situ SAED data of the oxidation process of FeNi alloy demonstrates its capability of implementing autonomous quantitative processing. © 2017 Elsevier Inc. All rights reserved.

Cellulose Nanofiber-Based Triboelectric Nanogenerators for Efficient Air Filtration in Harsh Environments.

Advanced portable healthcare devices with high efficiencies, small pressure drops, and high-temperature resistance are urgently desired in harsh environments with high temperatures, high humidities, and high levels of atmospheric pollution. Triboelectric nanogenerators (TENGs), which serve as energy converters in a revolutionary self-powered sensor device, present a sustainable solution for meeting these requirements. In this work, we developed a porous negative triboelectric material by synthesizing ZIF-8 on the surface of a cellulose/graphene oxide aerogel, grafting it with trimethoxy(1H,1H,2H,2H-heptadecafluorodecyl)silane, and adding a negative corona treatment, and it was combined with a positive triboelectric material to create a cellulose nanofiber-based TENG self-powered filter. The devices achieved a balance between a small pressure drop (53 Pa) and high filtration efficiency (98.97%, 99.65%, and 99.93% for PM0.3, PM0.5, and PM1, respectively), demonstrating robust filtration properties at high temperatures and high humidities. Our work provides a new approach for developing self-powered wearable healthcare devices with excellent air filtration properties.

The gut microbiota contributes to the infection of bovine viral diarrhea virus in mice.

Bovine viral diarrhea virus (BVDV) is prevalent worldwide and causes significant economic losses. Gut microbiota is a large microbial community and has a variety of biological functions. However, whether there is a correlation between gut microbiota and BVDV infection and what kind of relation between them have not been reported. Here, we found that gut microbiota composition changed in normal mice after infecting with BVDV, but mainly the low abundance microbe was affected. Interestingly, BVDV infection significantly reduced the diversity of gut microbiota and changed its composition in gut microbiota-dysbiosis mice. Furthermore, compared with normal mice of BVDV infection, there were more viral loads in the duodenum, jejunum, spleen, and liver of the gut microbiota-dysbiosis mice. However, feces microbiota transplantation (FMT) reversed these effects. The data above indicated that the dysbiosis of gut microbiota was a key factor in the high infection rate of BVDV. It is found that the IFN-I signal was involved by investigating the underlying mechanisms. The inhibition of the proliferation and increase in the apoptosis of peripheral blood lymphocytes (PBL) were also observed. However, FMT treatment reversed these changes by regulating PI3K/Akt, ERK, and Caspase-9/Caspase-3 pathways. Furthermore, the involvement of butyrate in the pathogenesis of BVDV was also further confirmed. Our results showed for the first time that gut microbiota acts as a key endogenous defense mechanism against BVDV infection; moreover, targeting regulation of gut microbiota structure and abundance may serve as a new strategy to prevent and control the disease.IMPORTANCEWhether the high infection rate of BVDV is related to gut microbiota has not been reported. In addition, most studies on BVDV focus on in vitro experiments, which limits the study of its prevention and control strategy and its pathogenic mechanism. In this study, we successfully confirmed the causal relationship between gut microbiota and BVDV infection as well as the potential molecular mechanism based on a mouse model of BVDV infection and a mouse model of gut microbiota dysbiosis. Meanwhile, a mouse model which is more susceptible to BVDV provided in this study lays an important foundation for further research on prevention and control strategy of BVDV and its pathogenesis. In addition, the antiviral effect of butyrate, the metabolites of butyrate-producing bacteria, has been further revealed. Overall, our findings provide a promising prevention and control strategy to treat this infectious disease which is distributed worldwide.

The activation of liver X receptors in Madin-Darby bovine kidney cells and mice restricts infection by bovine viral diarrhea virus.

Bovine viral diarrhea virus (BVDV) is prevalent worldwide and is an important pathogen that represents a serious threat to the development of the cattle industry by causing significant economic losses. Liver X receptors (LXRs) are members of the nuclear receptor superfamily and have become attractive therapeutic targets for cardiovascular disease. In the present study, we found that LXRs in both Madin-Darby bovine kidney (MDBK) cells and mice were associated with BVDV infection. GW3965, an agonist for LXRs, significantly inhibited BVDV RNA and protein levels in MDBK cells. In vivo studies in a mouse model also confirmed the inhibitory role of GW3965 in BVDV replication and the ameliorating effect of GW3965 on pathological injury to the duodenum. In vitro investigations of the potential mechanisms involved showed that GW3965 significantly inhibited BVDV-induced increases in cholesterol levels and viral internalization. Furthermore, the antiviral activity of GW3965 was significantly reduced following cholesterol replenishment, thus demonstrating that cholesterol was involved in the resistance of GW3965 to BVDV replication. Further studies indicated the role of ATP-binding cassette transporter A1 (ABCA1) and cholesterol-25-hydroxylase (CH25H) in the antiviral activity of GW3965. We also demonstrated the significant antiviral effect of 25hydroxycholesterol (25HC), a product of the catalysis of cholesterol by CH25H. In addition, the anti-BVDV effects of demethoxycurcumin (DMC), cyanidin-3-O-glucoside (C3G), and saikosaponin-A (SSA), three natural agonizts of LXRs, were also confirmed in both MDBK cells and mice. However, the antiviral activities of these agents were weakened by SR9243, a synthetic inhibitor of LXRs. For the first time, our research demonstrated that the activation of LXRs can exert significant anti-BVDV effects in MDBK cells and mice.

Improved Electrical Output Performance of Cellulose-Based Triboelectric Nanogenerators Enabled by Negative Triboelectric Materials.

Cellulose-based triboelectric nanogenerators (TENGs) have attracted widespread attention due to the low cost and environmentally friendly characteristics of cellulose. However, achieving high electrical energy output from these generators still presents significant challenges. Here, cellulose is dissolved-regenerated to form a composite aerogel with high specific surface area, in which cellulose-based composites with excellent negative triboelectric properties are developed by coupling the rich 3D network structure of the regenerated cellulose aerogel, modified barium titanate, and poly(vinylidene fluoride). The TENGs assembled from the composite materials exhibit an output voltage of 1040 V and a current of 1.165 mA at an external force of 8 N and a frequency of 4 Hz, outperforming all cellulose-based negative triboelectric materials. In addition, the nanogenerators have a stable electrical energy output capacity, with no significant property degradation in 100 000 contact-separation tests. The excellent electrical output property of the composite materials enables them to harvest energy from human movement and waterdrops, demonstrating their great application prospects in wearable devices, energy harvesting devices, self-powered sensors, and other fields.

Preventive effect and mechanism of compound Danshen dripping pills on contrast-induced nephropathy after percutaneous coronary interventional.

Background: Contrast-induced nephropathy (CIN) is one of the most common complications after coronary stent implantation due to the extensive development of coronary catheterization technology. Compound Danshen dripping pills (CDDP) are clinically used as cardiovascular drugs, relieving systemic inflammatory response. Previous studies have observed that CDDP can decrease CIN incidence after coronary stent implantation with uncertain effectiveness. Methods: We conducted a prospective, randomized, single-center, single-blind, controlled trial. We enrolled patients 18 years and older with unstable angina pectoris and NSTEMI who underwent PCI at the Tianjin Chest Hospital between November 1, 2021, and November 31, 2022, and followed for 30 days. Patients were randomized to CDDP and hydration therapy (10 capsules three times/day; N = 411) or hydration only (N = 411). The primary outcome was the contrast nephropathy incidence, defined as an elevation in serum creatinine by more than 25% or 44 µmol/L from baseline within 48-72 h of contrast exposure. Secondary outcomes included major adverse cardiovascular events post-surgery and during follow-up. Results: After 48 h of operation, the two groups had statistical significance in Scr and BUN values (80.0 ± 12.59 vs. 84.43 ± 13.49, P < 0.05; 6.22 ± 1.01 vs. 6.40 ± 0.93, P < 0.05). The difference in Scr in 72 h between the two groups was statistically significant (76.42 ± 10.92 vs. 79.06 ± 11.58, P < 0.05). The CIN incidence was significantly lower in the CDDP group than in the hydration group. The CIN risk was significantly elevated in patients with LVEF <50%, contrast volume ≥160 ml, and hypertension, after 48 and 72 h of operation. The serum inflammation index levels NGAL, TNF-α, oxidative stress indexes SOD, and MDA significantly differed between the two groups. However, there was no significant difference in serum apoptosis indexes Bax, Bcl-2, and Casepase-9. Conclusions: CDDP pre-treatment could prevent contrast-induced nephropathy. Inflammatory response and oxidative stress could be significant in the CDDP mechanism.

Evidence for Nitrous Oxide Emissions by Nitrite-Dependent Anaerobic Methane Oxidizing Bacteria.

Nitrite-dependent anaerobic methane oxidizing (n-DAMO) bacteria generally convert nitrite to dinitrogen and bypass the nitrous oxide (N2O) formation step. However, N2O is often detected in n-DAMO bacteria dominated cultures and it remains an open question as to the microbial origin of N2O in these enrichments. Using a stable nitrite consuming microbial community enriched for n-DAMO bacteria, we demonstrated that N2O production was coupled to methane oxidation and the higher initial nitrite concentrations led to increased quantities of N2O being formed. Moreover, continuous exposure of the enrichment culture to about 5 mg of N L-1 nitrite resulted in constant N2O being produced (12.5% of nitrite was reduced to N2O). Metatranscriptomic analyses revealed that nitrite reductase (nirS) and nitric oxide reductase (norZ) transcripts from n-DAMO bacteria increased in response to nitrite exposure. No other bacteria significantly expressed nor genes under these conditions, suggesting n-DAMO bacteria are responsible for N2O being produced. In a 35-day bioreactor experiment, N2O produced by the n-DAMO bacteria accumulated when nitrite was in excess; this was found to be up to 3.2% of the nitrogen that resulted from nitrite removal. Together, these results suggested that excess nitrite is an important driver of N2O production by n-DAMO bacteria. To this end, proper monitoring and control of nitrite levels in wastewater treatment plants would be effective strategies for mitigating N2O emissions to the atmosphere.

SwinDAE: Electrocardiogram Quality Assessment Using 1D Swin Transformer and Denoising AutoEncoder.

OBJECTIVE: Electrocardiogram (ECG) signals have wide-ranging applications in various fields, and thus it is crucial to identify clean ECG signals under different sensors and collection scenarios. Despite the availability of a variety of deep learning algorithms for ECG quality assessment, these methods still lack generalization across different datasets, hindering their widespread use. METHODS: In this paper, an effective model named Swin Denoising AutoEncoder (SwinDAE) is proposed. Specifically, SwinDAE uses a DAE as the basic architecture, and incorporates a 1D Swin Transformer during the feature learning stage of the encoder and decoder. SwinDAE was first pre-trained on the public PTB-XL dataset after data augmentation, with the supervision of signal reconstruction loss and quality assessment loss. Specially, the waveform component localization loss is proposed in this paper and used for joint supervision, guiding the model to learn key information of signals. The model was then fine-tuned on the finely annotated BUT QDB dataset for quality assessment. RESULTS: SwinDAE achieved 0.02-0.13 mean F1 score improvement on the BUT QDB dataset compared to multiple deep learning methods, and demonstrated applicability on two other datasets. CONCLUSION: The proposed SwinDAE shows strong generalization ability on different datasets, and surpasses other state-of-the-art deep learning methods on multiple evaluation metrics. In addition, the statistical analysis for SwinDAE prove the significance of the performance and the rationality of the prediction. SIGNIFICANCE: SwinDAE can learn the commonality between high-quality ECG signals, exhibiting excellent performance in the application of cross-sensors and cross-collection scenarios.

Different Methods of Minimally Invasive Esophagojejunostomy After Total Gastrectomy for Gastric Cancer: Outcomes from Two Experienced Centers.

BACKGROUND: Esophagojejunostomy after minimally invasive total gastrectomy (MITG) for gastric cancer (GC) is technically challenging. Failure of the esophagojejunal anastomosis can lead to significant morbidity, leading to short- and long-term quality of life (QoL) impairment or mortality. The optimal reconstruction method following MITG remains controversial. We evaluated outcomes of minimally invasive esophagojejunostomy after laparoscopic or robotic total gastrectomies. METHODS: We retrospectively reviewed MITG patients between 2015 and 2020 at two high-volume centers in China and the United States. Eligible patients were divided into groups by different reconstruction methods. We compared clinicopathologic characteristics, postoperative outcomes, including complication rates, overall survival rate (OS), disease-free survival rate (DFS), and patient-reported QoL. RESULTS: GC patients (n = 105) were divided into intracorporeal esophagojejunostomy (IEJ, n = 60) and extracorporeal esophagojejunostomy (EEJ, n = 45) groups. EEJ had higher incidence of wound infection (8.3% vs 13.3%, P = 0.044) and pneumonia (21.7% vs 40.0%, P = 0.042) than IEJ. The linear stapler (LS) group was inferior to the circular stapler (CS) group in reflux [50.0 (11.1-77.8) vs 44.4 (0.0-66.7), P = 0.041] and diarrhea [33.3 (0.0-66.7) vs 0.0 (0.0-66.7), P = 0.045] while LS was better than CS for dysphagia [22.2 (0.0-33.3) vs 11.1 (0.0-33.3), P = 0.049] and eating restrictions [33.3 (16.7-58.3) vs 41.7 (16.7-66.7), P = 0.029] at 1 year. OS and DFS did not differ significantly between LS and CS. CONCLUSIONS: IEJ anastomosis generated better results than EEJ. LS was associated with a better patient eating experience, but more diarrhea and reflux compared with CS. Clinical and patient-reported outcomes show the superiority of IEJ with the LS reconstruction method in MITG for GC.

A Deep Learning-Based System Trained for Gastrointestinal Stromal Tumor Screening Can Identify Multiple Types of Soft Tissue Tumors.

The accuracy and timeliness of the pathologic diagnosis of soft tissue tumors (STTs) critically affect treatment decision and patient prognosis. Thus, it is crucial to make a preliminary judgement on whether the tumor is benign or malignant with hematoxylin and eosin-stained images. A deep learning-based system, Soft Tissue Tumor Box (STT-BOX), is presented herein, with only hematoxylin and eosin images for malignant STT identification from benign STTs with histopathologic similarity. STT-BOX assumed gastrointestinal stromal tumor as a baseline for malignant STT evaluation, and distinguished gastrointestinal stromal tumor from leiomyoma and schwannoma with 100% area under the curve in patients from three hospitals, which achieved higher accuracy than the interpretation of experienced pathologists. Particularly, this system performed well on six common types of malignant STTs from The Cancer Genome Atlas data set, accurately highlighting the malignant mass lesion. STT-BOX was able to distinguish ovarian malignant sex-cord stromal tumors without any fine-tuning. This study included mesenchymal tumors that originated from the digestive system, bone and soft tissues, and reproductive system, where the high accuracy of migration verification may reveal the morphologic similarity of the nine types of malignant tumors. Further evaluation in a pan-STT setting would be potential and prospective, obviating the overuse of immunohistochemistry and molecular tests, and providing a practical basis for clinical treatment selection in a timely manner.