Colorectal Cancer Segmentation Algorithm Based on Deep Features from Enhanced CT Images

Colorectal cancer,a malignant lesion of the intestines,significantly affects human health and life,emphasizing the necessity of early detection and treatment.Accurate segmentation of colorectal cancer regions directly impacts subsequent staging,treatment methods,and prognostic outcomes.While colonoscopy is an effective method for detecting colorectal cancer,its data collection approach can cause patient discomfort.To address this,current research utilizes Computed Tomography(CT)imaging;however,conventional CT images only capture transient states,lacking sufficient representational capability to precisely locate colorectal cancer.This study utilizes enhanced CT images,constructing a deep feature network from the arterial,portal venous,and delay phases to simulate the physician’s diagnostic process and achieve accurate cancer segmentation.The innovations include:1)Utilizing portal venous phase CT images to introduce a context-aware multi-scale aggregation module for preliminary shape extraction of colorectal cancer.2)Building an image sequence based on arterial and delay phases,transforming the cancer segmentation issue into an anomaly detection problem,establishing a pixel-pairing strategy,and proposing a colorectal cancer segmentation algorithm using a Siamese network.Experiments with 84 clinical cases of colorectal cancer enhanced CT data demonstrated an Area Overlap Measure of 0.90,significantly better than Fully Convolutional Networks(FCNs)at 0.20.Future research will explore the relationship between conventional and enhanced CT to further reduce segmentation time and improve accuracy.

CFSA-Net:Efficient Large-Scale Point Cloud Semantic Segmentation Based on Cross-Fusion Self-Attention

Traditional models for semantic segmentation in point clouds primarily focus on smaller scales.However,in real-world applications,point clouds often exhibit larger scales,leading to heavy computational and memory requirements.The key to handling large-scale point clouds lies in leveraging random sampling,which offers higher computational efficiency and lower memory consumption compared to other sampling methods.Nevertheless,the use of random sampling can potentially result in the loss of crucial points during the encoding stage.To address these issues,this paper proposes cross-fusion self-attention network(CFSA-Net),a lightweight and efficient network architecture specifically designed for directly processing large-scale point clouds.At the core of this network is the incorporation of random sampling alongside a local feature extraction module based on cross-fusion self-attention(CFSA).This module effectively integrates long-range contextual dependencies between points by employing hierarchical position encoding(HPC).Furthermore,it enhances the interaction between each point’s coordinates and feature information through cross-fusion self-attention pooling,enabling the acquisition of more comprehensive geometric information.Finally,a residual optimization(RO)structure is introduced to extend the receptive field of individual points by stacking hierarchical position encoding and cross-fusion self-attention pooling,thereby reducing the impact of information loss caused by random sampling.Experimental results on the Stanford Large-Scale 3D Indoor Spaces(S3DIS),Semantic3D,and SemanticKITTI datasets demonstrate the superiority of this algorithm over advanced approaches such as RandLA-Net and KPConv.These findings underscore the excellent performance of CFSA-Net in large-scale 3D semantic segmentation.

Application of endoscopic retrograde cholangiopancreatography for treatment of obstructive jaundice after hepatoblastoma surgery:A case report

BACKGROUND We aimed to investigate the effectiveness of endoscopic retrograde cholangiopan-creatography(ERCP)for treating obstructive jaundice(OJ)post hepatoblastoma(HB)surgery(post-HB OJ)by analyzing the data of a case and performing a literature review.CASE SUMMARY Clinical data of one patient with post-HB OJ treated by ERCP were retrospectively analyzed.Furthermore,clinical characteristics and insights into the diagnosis and treatment of post-HB OJ in children were summarized via searching various databases and platforms,such as China National Knowledge Infrastructure,Wanfang database,CQVIP database,PubMed,Ringer Link,and Google Scholar.The patient reported herein underwent five chemotherapy sessions after the diagnosis of HB and right hemihepatectomy after tumor size reduction;these were followed by two postoperative chemotherapy sessions.Three months postoperatively,the patient developed icteric sclera,strong tea-colored urine,and clay-like stools,and showed signs of skin itchiness;blood analysis showed significantly an increased conjugated bilirubin(CB)level(200.3μmol/L).Following the poor efficacy of anti-jaundice and hepatoprotective treatments,the patient underwent ERCP.Intraoperative imaging showed a dilated bile duct in the porta hepatis with significant distal stenosis.A 5 Fr nasopancreatic tube was placed in the dilated area through the stricture for external drainage,and the patient was extubated on postoperative day 6.Postoperatively,the patient’s stool turned yellow,and the CB level decreased to 78.2μmol/L.Fifteen days later,ERCP was repeated due to unrelieved jaundice symptoms,wherein a 7 Fr naso-biliary drainage tube was successfully placed.Three months post-ERCP,the jaundice symptoms resolved,and the CB level was reduced to 33.2μmol/L.A follow-up examination one year postoperatively revealed no jaundice symptoms and normal CB level.CONCLUSION Post-HB OJ is rare.Compared to biliary tract reconstruction,ERCP is less invasive and has a better therapeutic effect.

Assessment of liver fibrosis by Fibroscan as compared to liver biopsy in biliary atresia

AIM: To evaluate liver stiffness measurement(LSM)using non-invasive transient elastography(Fibroscan)in comparison with liver biopsy for assessment of liver fibrosis in children with biliary atresia(BA).METHODS: Thirty-one children with BA admitted to the Department of Pediatric Surgery of Beijing Children's Hospital from March 2012 to February 2013 were included in this study. Their preoperative LSM, liver biopsy findings, and laboratory results were studied retrospectively.RESULTS: The grade of liver fibrosis in all 31 patients was evaluated according to the METAVIR scoring system, which showed that 4 cases were in group F2, 20 in group F3 and 7 in group F4. There were 24non-cirrhosis cases(F2-F3) and 7 cirrhosis cases(F4).In groups F2, F3 and F4, the mean LSM was 9.10 ±3.30 kPa, 11.02 ± 3.31 kPa and 22.86 ± 12.43 kPa,respectively. LSM was statistically different between groups F2 and F4(P = 0.002), and between groups F3 and F4(P = 0.000), however, there was no statistical difference between groups F2 and F3(P = 0.593). The area under the receiver operating characteristic curve of LSM for ≥ F4 was 0.866. The cut-off value of LSM was 15.15 kPa for ≥ F4, with a sensitivity, specificity,positive predictive value and negative predictive value of 0.857, 0.917, 0.750 and 0.957, respectively.CONCLUSION: Fibroscan can be used as a noninvasive technique to assess liver fibrosis in children with BA. The cut-off value of LSM(15.15 kPa) can distinguish cirrhotic patients from non-cirrhotic patients.

Pancake-Like MOF Solid-State Electrolytes with Fast Ion Migration for High-Performance Sodium Battery

Solid-state electrolyte(SSE)of the sodium-ion battery have attracted tremendous attention in the next generation energy storage materials on account of their wide electrochemical window and thermal stability.However,the high interfacial impedance,low ion transference number and complex preparation process restrict the application of SSE.Herein,inspired by the excellent sieving function and high specific surface area of red blood cells,we obtained a solid-like electrolyte(SLE)based on the combination of the pancake-like metal-organic framework(MOF)with liquid electrolyte,possessing a high ionic conductivity of 6.60×10^(-4) S cm^(−1),and excellent sodium metal compatibility.In addition,we investigated the ion restriction effect of MOF’s apertures size and special functional groups,and the ion transference number increased from 0.16 to 0.33.Finally,the assembled Na_(0.44)MnO_(2)//SLE//Na full batteries showed no obvious capacity decrease after 160 cycles.This material design of SLE in our work is an important key to obtain fast ion migration SLE for high-performance sodium-ion batteries.

Rcm2IT: Informatization Teaching Using Refined Classroom Management of Students’ Learning Process

In this paper,we propose an iterative,problem-and task-driven refined classroom management method for informatization teaching,dubbed Rcm2IT,based on data analysis of students’learning process.More specifically,Rcm2IT firstly classifies the teaching resources and elements in the unit of classroom sections,and designs the corresponding fine-grained templates and examples of teaching elements.Meanwhile,Rcm2IT divides students into different levels according to their previous classroom performance,and teaches students at different levels according to their aptitude.Finally,we implement the refined classroom management and assessment of students’learning process via our customized precise micro classroom platform(PMcP)based on the previous two stages.The designed PMcP can carry out real-time analysis and teaching evaluation,objectively and accurately measure and feedback the students’learning results,further iteratively optimize the teaching objectives.The course practice in“digital image processing”shows that Rcm2IT can not only support teachers to dynamically adjust the teaching content,but also teach students according to their aptitude,and improve the teaching quality and efficiency.

Optimal allocation method of energy storage for integrated renewable generation plants based on power market simulation

This study designs and proposes a method for evaluating the configuration of energy storage for integrated re-newable generation plants in the power spot market,which adopts a two-level optimization model of“system simulation+plant optimization”.The first step is“system simulation”which is using the power market simu-lation model to obtain the initial nodal marginal price and curtailment of the integrated renewable generation plant.The second step is“plant optimization”which is using the operation optimization model of the integrated renewable generation plant to optimize the charge-discharge operation of energy storage.In the third step,“sys-tem simulation”is conducted again,and the combined power of renewable and energy storage inside the plant is brought into the system model and simulated again for 8,760 h of power market year-round to quantify and compare the power generation and revenue of the integrated renewable generation plant after applying energy storage.In the case analysis of the provincial power spot market,an empirical analysis of a 1 GW wind-solar-storage integrated generation plant was conducted.The results show that the economic benefit of energy storage is approximately proportional to its capacity and that there is a slowdown in the growth of economic benefits when the capacity is too large.In the case that the investment benefit of energy storage only considers the in-come of electric energy-related incomes and does not consider the income of capacity mechanism and auxiliary services,the income of energy storage cannot fulfill the economic requirements of energy storage investment.

Meta-BN Net for few-shot learning

In this paper, we propose a lightweight network with an adaptive batch normalization module, called Meta-BN Net, for few-shot classification. Unlike existing few-shot learning methods, which consist of complex models or algorithms, our approach extends batch normalization, an essential part of current deep neural network training, whose potential has not been fully explored. In particular, a meta-module is introduced to learn to generate more powerful affine transformation parameters, known as and , in the batch normalization layer adaptively so that the representation ability of batch normalization can be activated. The experimental results on miniImageNet demonstrate that Meta-BN Net not only outperforms the baseline methods at a large margin but also is competitive with recent state-of-the-art few-shot learning methods. We also conduct experiments on Fewshot-CIFAR100 and CUB datasets, and the results show that our approach is effective to boost the performance of weak baseline networks. We believe our findings can motivate to explore the undiscovered capacity of base components in a neural network as well as more efficient few-shot learning methods.

Tailoring Meta-Liquid Crystal for Larger Tunability

Liquid crystals are pivotal innovations in the field of tunable materials.However,in the millimeter wave band,the relatively low tunability of liquid crystal limits its applicability.A design method utilizing metamaterial characteristics to enhance the tunability of a liquid crystal is proposed.This design method is based on two theoretical models.Both models embed metamaterial structures into a liquid crystal,thus tailoring a new tunable material,namely,a meta-liquid crystal.By appropriately designing the metamaterials to meet the corresponding conditions derived in different models,the tunability of the meta-liquid crystal can be higher than that of the original liquid crystal.An application of beam scanning is provided to verify this method.Two beam steerable antenna arrays are designed,fabricated and investigated.The beam scanning range of the meta-liquid crystal antenna array is much wider than that of the original liquid crystal antenna array,which confirms the enhancement of tunability.Consequently,this work fills the research gap in improving the performance of liquid crystals in the millimeter wave band.Simultaneously,a bridge is established between two significant materials:tunable materials and metamaterials.The results imply that the performance of tunable materials can be manipulated by metamaterials.

骨质疏松性椎体压缩骨折椎体强化术后骨水泥血管渗漏的危险因素分析

目的探讨骨质疏松性椎体压缩骨折(OVCF)椎体强化术后骨水泥血管渗漏的危险因素。方法采用病例对照研究分析2019年10月至2020年10月昆明医科大学第一及第二附属医院收治的217例OVCF行椎体强化术[经皮椎体成形术(PVP)或经皮椎体后凸成形术(PKP)]患者的临床资料, 其中男79例, 女138例;年龄58~88岁[(73.1±6.9)岁]。根据是否出现骨水泥血管渗漏分为血管渗漏组(39例)和无血管渗漏组(178例)。记录两组患者性别、年龄、骨密度、受伤至手术时间、伤椎解剖位置、压缩程度、后壁完整性、椎内裂隙征、椎基静脉孔、手术入路、手术方式, 以及骨水泥注入时期、注入速度、注射量、注射区域。采用单因素分析上述指标与椎体强化术后骨水泥血管渗漏的相关性, 再通过多因素Logistic回归分析明确椎体强化术后骨水泥血管渗漏的独立危险因素。结果单因素分析结果显示, 受伤至手术时间、压缩程度、后壁完整性、椎内裂隙征、椎基静脉孔、手术方式, 以及骨水泥注入时期、注入速度、注射量、注射区域与椎体强化术后骨水泥血管渗漏相关(P<0.05);性别、年龄、骨密度、伤椎解剖位置、手术入路与椎体强化术后骨水泥血管渗漏不相关(P>0.05)。多因素Logistic 回归分析结果显示, 椎内裂隙征(OR=7.00, 95%CI 1.57~31.30, P<0.05)、椎基静脉孔(OR=7.52, 95%CI 1.94~29.16, P<0.01)、PVP(OR=10.98, 95%CI 2.51~47.94, P<0.01)、骨水泥稀薄期注入(OR=5.91, 95%CI 1.45~24.15, P<0.05)、骨水泥注射量大(OR=3.60, 95%CI 1.70~7.65, P<0.01)、骨水泥边缘区注射(OR=24.80, 95%CI 5.28~116.37, P<0.01)与OVCF椎体强化术后骨水泥血管渗漏显著相关。结论椎内裂隙征、椎基静脉孔、PVP, 以及骨水泥稀薄期注入、注射量大、边缘区注射是OVCF椎体强化术后骨水泥血管渗漏的独立危险因素。

元自步学习

自步学习是近年来机器学习领域提出的一种启发于人和动物"由易到难"学习过程的学习机制.尽管自步学习已取得可喜的理论与应用进展,但是当前的自步学习算法仍存在超参数选择的瓶颈问题.针对该问题当前主要采用一些启发式的手工设计方法或者交叉验证方法,然而此类方法效率很低,缺乏理论性指导,难以推广应用到广泛的实践问题中.针对这一挑战性问题,本文提出一种基于元学习机理的自步学习算法,该方法能使自步学习中涉及的超参数以数据驱动的方式自动习得,从而大大减弱了自步学习的这一核心问题.特别地,我们针对3种典型的自步学习实现格式,将所提元学习策略实质性嵌入,通过回归和分类实验验证了所提算法的准确性和泛化性,特别验证了相比于传统超参设置方法的显著优越性.

Effect of Cerium on High-Temperature Oxidation Resistance of 00Cr17NbTi Ferritic Stainless Steel

The influence of cerium addition on the isothermal oxidation behavior of 00Cr17NbTi ferritic stainless steel was studied at temperature up to 1,000 ℃ for 100 h in air. The results show that cerium additions can reduce the grain size of this ferritic stainless steel, improve the diffusion of chromium and decrease the critical concentration of chromium to form protective Cr2O3 layer. With the increasing of cerium addition, the oxide particles become smaller and this can increase the rupture strength and spalling resistance of oxide layers. The transport mechanism through the oxide layer is varied from metal transport outward from steel to principally oxygen transport inward with the increase of cerium content, which leads to the lower oxidation rate and the better scale adherence of 00Cr17NbTi ferritic stainless steel.

Construction of cDNA Library from NPC Tissue and Screening of Antigenic Genes

To construct cDNA library of nasopharyngeal carcinoma （NPC） and obtain the NPC associated or specific antigens from it, we used a powerful new method to identify the antigens eliciting humoral immune response, which is SEREX （serological identification of antigen by recombinant cDNA expression library）. Autologous serum of NPC patient was used to screen the reactive clones in the human NPC tissue cDNA library consisted of 3.64 × 10^6 recombinants. The 23 exact positive clones were subcloned to monoclonality and the size of cDNA inserts was identified by PCR. Then the nucleotide sequence of cDNA inserts was determined, and the sequence alignments were performed with BLAST software on GenBank database. They represented 16 different antigens. A detailed sequence analysis showed that 10 of 16 genes were high homologous to genes known in GenBank, such as RPL31, SI00 A2, MT2A, etc. However, there were also 6 genes with low homology to genes in GenBank. Furthermore, 3 of 6 genes may be novel genes. The associations of these genes to NPC and the roles that they played in the occurrence and development of NPC should be further revealed.

Deficiency of two-pore segment channel 2 contributes to systemic lupus erythematosus via regulation of apoptosis and cell cycle

Background:Systemic lupus erythematosus(SLE)is a complex autoimmune disease,and the mechanism of SLE is yet to be fully elucidated.The aim of this study was to explore the role of two-pore segment channel 2(TPCN2)in SLE pathogenesis.Methods:Quantitative reverse transcription polymerase chain reaction(qRT-PCR)was used to detect the expression ofTPCN2 in SLE.We performed a loss-of-function assay by lentiviral construct in Jurkat and THP-1 cell.Knockdown ofTPCN2 were confirmed at the RNA level by qRT-PCR and protein level by Western blotting.Cell Count Kit-8 and flow cytometry were used to analyze the cell proliferation,apoptosis,and cell cycle ofTPCN2-deficient cells.In addition,gene expression profile ofTPCN2-deficient cells was analyzed by RNA sequencing(RNA-seq).Results:TPCN2 knockdown with short hairpin RNA(shRNA)-mediated lentiviruses inhibited cell proliferation,and induced apoptosis and cell-cycle arrest of G2/M phase in both Jurkat and THP-1 cells.We analyzed the transcriptome of knockdown-TPCN2-Jurkat cells,and screened the differential genes,which were enriched for the G2/M checkpoint,complement,and interleukin-6-Janus kinase-signal transducer and activator of transcription pathways,as well as changes in levels of forkhead box O,phosphatidylinositol 3-kinase/protein kinase B/mechanistic target of rapamycin,and T cell receptor pathways;moreover,TPCN2 significantly influenced cellular processes and biological regulation.Conclusion:TPCN2 might be a potential protective factor against SLE.