基于改进Node2vec算法的锅炉温度场分割方法研究

针对温度场特征参数差异引发的锅炉温度场分割准确性的问题,以维持温度场特征为目标,引入图结构表达场数据,通过改进Node2vec算法进行聚类分析,进而实现锅炉温度场的最佳分割。该方法基于多维度的特征信息对锅炉温度场实现分割,能够更准确地保留流场特征。在标准数据集上进行了实验验证,结果表明在具有多维度特征的数据集上,所提方法相比其他对比算法在分割效果方面有提升显著。最后将提出的方法用于分割电站锅炉温度场,结果表明该方法可以很好地捕捉温度场数据中的局部和全局特征,且结果具有较好的精确性。

基于Node.js的农业物联网应用平台设计

文章的研究目的在于提出基于Node.js的物联网应用平台设计方案,解决物联网实际运行过程中数据上传存在的问题,提出并设计一种基于Node.js的农业物联网应用平台,可以实现对农业生产环境参数的监测与控制。首先介绍农业物联网的背景与需求,明确研究目标,其次详细描述本平台的设计思路与架构,主要应用层平台软件的设计与实现。平台的实际应用显示,本平台能够在实现物联网常见功能的同时,支持更大的数据并发处理需求,对于农业物联网的应用特点有着更好的支持作用。

Analysis of the optimal target node to reduce seizure-like discharge in networks

Network approaches have been widely accepted to guide surgical strategy and predict outcome for epilepsy treatment.This study starts with a single oscillator to explore brain activity,using a phenomenological model capable of describing healthy and epileptic states.The ictal number of seizures decreases or remains unchanged with increasing the speed of oscillator excitability and in each seizure,there is an increasing tendency for ictal duration with respect to the speed.The underlying reason is that the strong excitability speed is conducive to reduce transition behaviors between two attractor basins.Moreover,the selection of the optimal removal node is estimated by an indicator proposed in this study.Results show that when the indicator is less than the threshold,removing the driving node is more possible to reduce seizures significantly,while the indicator exceeds the threshold,the epileptic node could be the removal one.Furthermore,the driving node is such a potential target that stimulating it is obviously effective in suppressing seizure-like activity compared to other nodes,and the propensity of seizures can be reduced 60%with the increased stimulus strength.Our results could provide new therapeutic ideas for epilepsy surgery and neuromodulation.

Cooperative Rate Splitting Transmit Design for Full-Duplex-Enabled Multiple Multicast Communication Systems

This paper examines the performance of Full-Duplex Cooperative Rate Splitting(FD-CRS)with Simultaneous Wireless Information and Power Transfer(SWIPT)support in Multiple Input Single Output(MISO)networks.In a Rate Splitting Multiple Access(RSMA)multicast system with two local users and one remote user,the common data stream contains the needs of all users,and all users can decode the common data stream.Therefore,each user can receive some information that other users need,and local users with better channel conditions can use this information to further enhance the reception reliability and data rate of users with poor channel quality.Even using Cell-Center-Users(CCUs)as a cooperative relay to assist the transmission of common data can improve the average system speed.To maximize the minimum achievable rate,we optimize the beamforming vector of Base Station(BS),the common streamsplitting vector,the cooperative distributed beamvector and the strong user transmission power under the power budget constraints of BS and relay devices and the service quality requirements constraints of users.Since the whole problem is not convex,we cannot solve it directly.Therefore,we propose a low complexity algorithm based on Successive Convex Approximation(SCA)technology to find the optimal solution to the problemunder consideration.The simulation results show that FD C-RSMA has better gain andmore powerful than FD C-NOMA,HD C-RSMA,RSMA and NOMA.

Fifty-five cases of hepatic alveolar echinococcosis combined with lymph node metastasis:A retrospective study

BACKGROUND Lymph node metastasis is a specific type of metastasis in hepatic alveolar echinococcosis(AE).Currently,there is a scarcity of describing the clinical characteristics and lymph node metastasis rules of patients with hepatic AE combined with lymph node metastasis and its mechanism and management are still controversial.Radical hepatectomy combined with regional lymph node dissection is a better treatment.AIM To analyse the clinical features of hepatic AE combined with lymph node metastasis to explore its treatment and efficacy.METHODS A total of 623 patients with hepatic AE admitted to the First Affiliated Hospital of Xinjiang Medical University from 1 January 2012 to 1 January 2022 were retrospectively analysed.Fifty-five patients with combined lymph node metastasis were analysed for their clinical data,diagnosis and treatment methods,follow-up efficacy,and characteristics of lymph node metastasis.Finally,we comparatively analysed the lymph node metastasis rates at different sites.Categorical variables are expressed as frequencies and percentages,and the analysis of difference was performed using theχ2 test.The Bonferroni method was used for pairwise comparisons when statistical differences existed between multiple categorical variables.RESULTS A lymph node metastasis rate of 8.8%(55/623)was reported in patients with hepatic AE,with a female predilection(69.1%)and a statistically significant sex difference(χ2=8.018,P=0.005).Of the 55 patients with lymph node metastasis,72.7%had a parasite lesion,neighbouring organ invasion,and metastasis stage of P3N1M0 and above,of which 67.3%,78.2%,and 34.5%of hepatic AE lesions invaded the bile ducts,blood vessels,and distant metastases,respectively.Detection rates of lymph node metastasis of 16.4%,21.7%,and 34.2%were reported for a preoperative abdominal ultrasound,magnetic resonance imaging,and computed tomography examinations.All patients were intraoperatively suspected with enlarged lymph nodes and underwent radical hepatectomy combined with regional lymph node dissection.After surgery,a routine pathological examination was conducted on the resected lymph nodes.A total of 106 positive lymph nodes were detected in six groups at various sites,including 51 single-group metastasis cases and four multi-group metastasis cases.When the metastasis rates at different sites were statistically analysed,we observed that the metastasis rate in the para-hepatoduodenal ligament lymph nodes was significantly higher than that of the other sites(χ2=128.089,P=0.000<0.05).No statistical difference was observed in the metastasis rate between the five other groups.Clavien-Dindo grade IIIa complication occurred in 14 cases,which improved after administering symptomatic treatment.Additionally,lymph node dissection-related complications were not observed.Recurrence after 2 years was observed in one patient.CONCLUSION Lymph node metastasis is a rare form of metastasis in hepatic AE,which is more frequent in women.Parahepatoduodenal ligament lymph nodes are commonly observed.Radical hepatectomy combined with regional lymph node dissection is a safe,effective,and feasible treatment for liver AE combined with lymph node metastasis.

Different lymph node staging systems for predicting the prognosis of colorectal neuroendocrine neoplasms

BACKGROUND Colorectal neuroendocrine neoplasms(NENs)are a rare malignancy that primarily arises from the diffuse distribution of neuroendocrine cells in the colon and rectum.Previous studies have pointed out that the status of lymph node may be used to predict the prognosis.AIM To investigate the predictive values of lymph node ratio(LNR),positive lymph node(PLN),and log odds of PLNs(LODDS)staging systems on the prognosis of colorectal NENs treated surgically,and compare their predictive values.METHODS This cohort study included 895 patients with colorectal NENs treated surgically from the Surveillance,Epidemiology,and End Results database.The endpoint was mortality of patients with colorectal NENs treated surgically.X-tile software was utilized to identify most suitable thresholds for categorizing the LNR,PLN,and LODDS.Participants were selected in a random manner to form training and testing sets.The prognosis of surgically treating colorectal NENs was examined using multivariate cox analysis to assess the associations of LNR,PLN,and LODDS with the prognosis of colorectal NENs.C-index was used for assessing the predictive effectiveness.We conducted a subgroup analysis to explore the different lymph node staging systems’predictive values.RESULTS After adjusting all confounding factors,PLN,LNR and LODDS staging systems were linked with mortality in patients with colorectal NENs treated surgically(P<0.05).We found that LODDS staging had a higher prognostic value for patients with colorectal NENs treated surgically than PLN and LNR staging systems.Similar results were obtained in the different G staging subgroup analyses.Furthermore,the area under the receiver operating characteristic curve values for LODDS staging system remained consistently higher than those of PLN or LNR,even at the 1-,2-,3-,4-,5-and 6-year follow-up periods.CONCLUSION LNR,PLN,and LODDS were found to significantly predict the prognosis of patients with colorectal NENs treated surgically.

Robotic-assisted retroperitoneal lymph node dissection for stage II testicular cancer

Objective:To evaluate the perioperative as well as early oncological outcomes of patients undergoing robotic retroperitoneal lymph node dissection for treatment of testicular cancer.Methods:We conducted a prospective consecutive case series of patients undergoing robotic assisted retroperitoneal lymph node dissection for metastatic testicular cancer between May 2018 and July 2021 at our institution.Data were collected on patient and tumour characteristics,intraoperative and postoperative parameters,and functional and oncological outcomes.Descriptive statistics are presented.Results:Nineteen patients were identified;18(94.7%)completed the procedure robotically and one was converted to open surgery;78.9%of patients had stage≥IIB and 12(63.2%)patients had undergone prior chemotherapy.The median operative time was 300(interquartile range[IQR]240-315)min.Median blood loss was 100(IQR 50-175)mL.Median length of stay was 2(range 1-11)days.All robotically completed patients commenced diet and passed flatus on Day 1 and were discharged by Day 3.The median lymph node yield was 40.5(IQR 38-51)nodes.All patients undergoing nerve-sparing procedures recovered antegrade ejaculatory function.One patient had a Clavien-Dindo III complication(chylous ascites requiring drainage).At a median follow-up of 22.3(IQR 16.3-24.9)months,one patient developed retroperitoneal recurrence,which was successfully treated with second-line chemotherapy;no other patients have had recurrences.Conclusion:Robotic retroperitoneal lymph node dissection is a safe and feasible alternative to open surgery in appropriately selected patients,offering low morbidity.Early oncological outcomes are promising.Larger cohorts and longer follow-ups are required to validate our institution's findings.

A Layered Energy-EfficientMulti-Node Scheduling Mechanism for Large-Scale WSN

In recent years, target tracking has been considered one of the most important applications of wireless sensornetwork (WSN). Optimizing target tracking performance and prolonging network lifetime are two equally criticalobjectives in this scenario. The existing mechanisms still have weaknesses in balancing the two demands. Theproposed heuristic multi-node collaborative scheduling mechanism (HMNCS) comprises cluster head (CH)election, pre-selection, and task set selectionmechanisms, where the latter two kinds of selections forma two-layerselection mechanism. The CH election innovatively introduces the movement trend of the target and establishesa scoring mechanism to determine the optimal CH, which can delay the CH rotation and thus reduce energyconsumption. The pre-selection mechanism adaptively filters out suitable nodes as the candidate task set to applyfor tracking tasks, which can reduce the application consumption and the overhead of the following task setselection. Finally, the task node selection is mathematically transformed into an optimization problem and thegenetic algorithm is adopted to form a final task set in the task set selection mechanism. Simulation results showthat HMNCS outperforms other compared mechanisms in the tracking accuracy and the network lifetime.

“Half of the Node Records Are Forged?”: The Problemof Node Records Forgery in Ethereum Network

Ethereum, currently the most widely utilized smart contracts platform, anchors the security of myriad smartcontracts upon its own robustness. Its foundational peer-to-peer network facilitates a dependable node connectionmechanism, whereas an efficient data-sharing protocol constitutes as the bedrock of Blockchain network security.In this paper, we propose NodeHunter, an Ethereum network detector implemented through the application ofsimulation technology, which is capable of aggregating all node records within the network and the interconnectednessbetween them. Utilizing this connection information, NodeHunter can procure more comprehensive insightsfor network status analysis compared to preceding detection methodologies. Throughout a three-month period ofunbroken surveillance of the Ethereum network, we obtained an excess of two million node records along with overone hundred million node acquaintances. Analysis of the gathered data revealed that an alarming 49% or more ofthese node records were maliciously forged.

Management of lateral pelvic lymph nodes in rectal cancer:Is it time to reach an Agreement?

In this editorial,we proceed to comment on the article by Chua et al,addressing the management of metastatic lateral pelvic lymph nodes(mLLN)in stage II/III rectal cancer patients below the peritoneal reflection.The treatment of this nodal area sparks significant controversy due to the strategic differences followed by Eastern and Western physicians,albeit with a higher degree of convergence in recent years.The dissection of lateral pelvic lymph nodes without neoadjuvant therapy is a standard practice in Eastern countries.In contrast,in the West,preference leans towards opting for neoadjuvant therapy with chemoradiotherapy or radiotherapy,that would cover the treatment of this area without the need to add the dissection of these nodes to the total mesorectal excision.In the presence of high-risk nodal characteristics for mLLN related to radiological imaging and lack of response to neoadjuvant therapy,the risk of lateral local recurrence increases,suggesting the appropriate selection of strategies to reduce the risk of recurrence in each patient profile.Despite the heterogeneous and retrospective nature of studies addressing this area,an international consensus is necessary to approach this clinical scenario uniformly.

Can we triumph over locally advanced cervical cancer with colossal para-aortic lymph nodes? A case report

BACKGROUND Para-aortic lymph nodes(PALNs)are common sites for the regional spread of cervical squamous cell carcinoma(SCC).CASE SUMMARY We report the case of a 36-year-old woman who presented with cervical SCC with multiple bulky PALNs,largest measured 4.5 cm×5 cm×10 cm.The patient was treated with radical intent with definitive chemoradiation using sequential doseescalated adaptive radiotherapy,followed by maintenance chemotherapy.The patient achieved a complete response;she has been doing well since the completion of treatment with no evidence of the disease for 2 years.CONCLUSION Regardless of the size of PALN metastases of cervical carcinoma origin,it is still treatable(with radical intent)via concurrent chemoradiation.Adaptive radiotherapy allows dose escalation with minimal toxicity.

Feasibility and limitations of combined treatment for lateral pelvic lymph node metastases in rectal cancer

Colorectal cancer ranks among the most commonly diagnosed cancers globally,and is associated with a high rate of pelvic recurrence after surgery.In efforts to mitigate recurrence,pelvic lymph node dissection(PLND)is commonly advocated as an adjunct to radical surgery.Neoadjuvant chemoradiotherapy(NACRT)is a therapeutic approach employed in managing locally advanced rectal cancer,and has been found to increase the survival rates.Chua et al have proposed a combination of NACRT with selective PLND for addressing lateral pelvic lymph node metastases in rectal cancer patients,with the aim of reducing recurrence and improving survival outcomes.Nevertheless,certain studies have indicated that the addition of PLND to NACRT and total mesorectal excision did not yield a significant reduction in local recurrence rates or improvement in survival.Consequently,meticulous patient selection and perioperative chemotherapy may prove indispensable in ensuring the efficacy of PLND.

An Optimal Node Localization in WSN Based on Siege Whale Optimization Algorithm

Localization or positioning scheme in Wireless sensor networks (WSNs) is one of the most challenging andfundamental operations in various monitoring or tracking applications because the network deploys a large areaand allocates the acquired location information to unknown devices. The metaheuristic approach is one of themost advantageous ways to deal with this challenging issue and overcome the disadvantages of the traditionalmethods that often suffer from computational time problems and small network deployment scale. This studyproposes an enhanced whale optimization algorithm that is an advanced metaheuristic algorithm based on thesiege mechanism (SWOA) for node localization inWSN. The objective function is modeled while communicatingon localized nodes, considering variables like delay, path loss, energy, and received signal strength. The localizationapproach also assigns the discovered location data to unidentified devices with the modeled objective functionby applying the SWOA algorithm. The experimental analysis is carried out to demonstrate the efficiency of thedesigned localization scheme in terms of various metrics, e.g., localization errors rate, converges rate, and executedtime. Compared experimental-result shows that theSWOA offers the applicability of the developed model forWSNto perform the localization scheme with excellent quality. Significantly, the error and convergence values achievedby the SWOA are less location error, faster in convergence and executed time than the others compared to at least areduced 1.5% to 4.7% error rate, and quicker by at least 4%and 2% in convergence and executed time, respectivelyfor the experimental scenarios.

Magnetic resonance imaging-based lymph node radiomics for predicting the metastasis of evaluable lymph nodes in rectal cancer

BACKGROUND Lymph node(LN)staging in rectal cancer(RC)affects treatment decisions and patient prognosis.For radiologists,the traditional preoperative assessment of LN metastasis(LNM)using magnetic resonance imaging(MRI)poses a challenge.AIM To explore the value of a nomogram model that combines Conventional MRI and radiomics features from the LNs of RC in assessing the preoperative metastasis of evaluable LNs.METHODS In this retrospective study,270 LNs(158 nonmetastatic,112 metastatic)were randomly split into training(n=189)and validation sets(n=81).LNs were classified based on pathology-MRI matching.Conventional MRI features[size,shape,margin,T2-weighted imaging(T2WI)appearance,and CE-T1-weighted imaging(T1WI)enhancement]were evaluated.Three radiomics models used 3D features from T1WI and T2WI images.Additionally,a nomogram model combining conventional MRI and radiomics features was developed.The model used univariate analysis and multivariable logistic regression.Evaluation employed the receiver operating characteristic curve,with DeLong test for comparing diagnostic performance.Nomogram performance was assessed using calibration and decision curve analysis.RESULTS The nomogram model outperformed conventional MRI and single radiomics models in evaluating LNM.In the training set,the nomogram model achieved an area under the curve(AUC)of 0.92,which was significantly higher than the AUCs of 0.82(P<0.001)and 0.89(P<0.001)of the conventional MRI and radiomics models,respectively.In the validation set,the nomogram model achieved an AUC of 0.91,significantly surpassing 0.80(P<0.001)and 0.86(P<0.001),respectively.CONCLUSION The nomogram model showed the best performance in predicting metastasis of evaluable LNs.

Value of glucose transport protein 1 expression in detecting lymph node metastasis in patients with colorectal cancer

BACKGROUND There are limited data on the use of glucose transport protein 1(GLUT-1)expre-ssion as a biomarker for predicting lymph node metastasis in patients with colorectal cancer.GLUT-1 and GLUT-3,hexokinase(HK)-II,and hypoxia-induced factor(HIF)-1 expressions may be useful biomarkers for detecting primary tumors and lymph node metastasis when combined with fluorodeoxyglucose(FDG)uptake on positron emission tomography/computed tomography(PET/CT).AIM To evaluate GLUT-1,GLUT-3,HK-II,and HIF-1 expressions as biomarkers for detecting primary tumors and lymph node metastasis with 18F-FDG-PET/CT.METHODS This retrospective study included 169 patients with colorectal cancer who underwent colectomy and preoperative 18F-FDG-PET/CT at Chungbuk National University Hospital between January 2009 and May 2012.Two tissue cores from the central and peripheral areas of the tumors were obtained and were examined by a dedicated pathologist,and the expressions of GLUT-1,GLUT-3,HK-II,and HIF-1 were determined using immunohisto-chemical staining.We analyzed the correlations among their expressions,various clinicopathological factors,and the maximum standardized uptake value(SUVmax)of PET/CT.RESULTS GLUT-1 was found at the center or periphery of the tumors in 109(64.5%)of the 169 patients.GLUT-1 positivity was significantly correlated with the SUVmax of the primary tumor and lymph nodes,regardless of the biopsy site(tumor center,P<0.001 and P=0.012;tumor periphery,P=0.030 and P=0.010,respectively).GLUT-1 positivity and negativity were associated with higher and lower sensitivities of PET/CT,respectively,for the detection of lymph node metastasis,regardless of the biopsy site.GLUT3,HK-II,and HIF-1 expressions were not significantly correlated with the SUVmax of the primary tumor and lymph nodes.CONCLUSION GLUT-1 expression was significantly correlated with the SUVmax of 18F-FDG-PET/CT for primary tumors and lymph nodes.Clinicians should consider GLUT-1 expression in preoperative endoscopic biopsy in interpreting PET/CT findings.

Predictive value of positive lymph node ratio in patients with locally advanced gastric remnant cancer

BACKGROUND Traditional lymph node stage(N stage)has limitations in advanced gastric remnant cancer(GRC)patients;therefore,establishing a new predictive stage is necessary.AIM To explore the predictive value of positive lymph node ratio(LNR)according to clinicopathological characteristics and prognosis of locally advanced GRC.METHODS Seventy-four patients who underwent radical gastrectomy and lymphadenectomy for locally advanced GRC were retrospectively reviewed.The relationship between LNR and clinicopathological characteristics was analyzed.The survival analysis was performed using Kaplan-Meier survival curves and Cox regression model.RESULTS Number of metastatic LNs,tumor diameter,depth of tumor invasion,Borrmann type,serum tumor biomarkers,and tumor-node-metastasis(TNM)stage were correlated with LNR stage and N stage.Univariate analysis revealed that the factors affecting survival included tumor diameter,anemia,serum tumor biomarkers,vascular or neural invasion,combined resection,LNR stage,N stage,and TNM stage(all P<0.05).The median survival time for those with LNR0,LNR1,LNR2 and LNR3 stage were 61,31,23 and 17 mo,respectively,and the differences were significant(P=0.000).Anemia,tumor biomarkers and LNR stage were independent prognostic factors for survival in multivariable analysis(all P<0.05).CONCLUSION The new LNR stage is uniquely based on number of metastatic LNs,with significant prognostic value for locally advanced GRC,and could better differentiate overall survival,compared with N stage.

Assessment of programmed death-ligand 1 expression in primary tumors and paired lymph node metastases of gastric adenocarcinoma

BACKGROUND Anti-programmed death-1/programmed death-ligand 1(PD-1/PD-L1)immuno-therapy has demonstrated promising results on gastric cancer(GC).However,PD-L1 can express differently between metastatic sites and primary tumors(PT).AIM To compare PD-L1 status in PT and matched lymph node metastases(LNM)of GC patients and to determine the correlation between the PD-L1 status and clinicopathological characteristics.METHODS We retrospectively reviewed 284 GC patients who underwent D2-gastrectomy.PD-L1 was evaluated by immunohistochemistry(clone SP142)using the com-bined positive score.All PD-L1+PT staged as pN+were also tested for PD-L1 expression in their LNM.PD-L1(-)GC with pN+served as the comparison group.RESULTS Among 284 GC patients included,45 had PD-L1+PT and 24 of them had pN+.For comparison,44 PD-L1(-)cases with pN+were included(sample loss of 4 cases).Of the PD-L1+PT,54.2%(13/24 cases)were also PD-L1+in the LNM.Regarding PD-L1(-)PT,9.1%(4/44)had PD-L1+in the LNM.The agreement between PT and LNM had a kappa value of 0.483.Larger tumor size and moderate/severe peritumoral inflammatory response were associated with PD-L1 positivity in both sites.There was no statistical difference in overall survival for PT and LNM according to the PD-L1 status(P=0.166 and P=0.837,respectively).CONCLUSION Intra-patient heterogeneity in PD-L1 expression was observed between the PT and matched LNM.This disagreement in PD-L1 status may emphasize the importance of considering different tumor sites for analyses to select patients for immunotherapy.

Target layer state estimation in multi-layer complex dynamical networks considering nonlinear node dynamics

In many engineering networks, only a part of target state variables are required to be estimated.On the other hand,multi-layer complex network exists widely in practical situations.In this paper, the state estimation of target state variables in multi-layer complex dynamical networks with nonlinear node dynamics is studied.A suitable functional state observer is constructed with the limited measurement.The parameters of the designed functional observer are obtained from the algebraic method and the stability of the functional observer is proven by the Lyapunov theorem.Some necessary conditions that need to be satisfied for the design of the functional state observer are obtained.Different from previous studies, in the multi-layer complex dynamical network with nonlinear node dynamics, the proposed method can estimate the state of target variables on some layers directly instead of estimating all the individual states.Thus, it can greatly reduce the placement of observers and computational cost.Numerical simulations with the three-layer complex dynamical network composed of three-dimensional nonlinear dynamical nodes are developed to verify the effectiveness of the method.

Predictive value of machine learning models for lymph node metastasis in gastric cancer: A two-center study

BACKGROUND Gastric cancer is one of the most common malignant tumors in the digestive system,ranking sixth in incidence and fourth in mortality worldwide.Since 42.5%of metastatic lymph nodes in gastric cancer belong to nodule type and peripheral type,the application of imaging diagnosis is restricted.AIM To establish models for predicting the risk of lymph node metastasis in gastric cancer patients using machine learning(ML)algorithms and to evaluate their pre-dictive performance in clinical practice.METHODS Data of a total of 369 patients who underwent radical gastrectomy at the Depart-ment of General Surgery of Affiliated Hospital of Xuzhou Medical University(Xuzhou,China)from March 2016 to November 2019 were collected and retro-spectively analyzed as the training group.In addition,data of 123 patients who underwent radical gastrectomy at the Department of General Surgery of Jining First People’s Hospital(Jining,China)were collected and analyzed as the verifi-cation group.Seven ML models,including decision tree,random forest,support vector machine(SVM),gradient boosting machine,naive Bayes,neural network,and logistic regression,were developed to evaluate the occurrence of lymph node metastasis in patients with gastric cancer.The ML models were established fo-llowing ten cross-validation iterations using the training dataset,and subsequently,each model was assessed using the test dataset.The models’performance was evaluated by comparing the area under the receiver operating characteristic curve of each model.RESULTS Among the seven ML models,except for SVM,the other ones exhibited higher accuracy and reliability,and the influences of various risk factors on the models are intuitive.CONCLUSION The ML models developed exhibit strong predictive capabilities for lymph node metastasis in gastric cancer,which can aid in personalized clinical diagnosis and treatment.

Combining machine and deep transfer learning for mediastinal lymph node evaluation in patients with lung cancer

Background The prognosis and survival of patients with lung cancer are likely to deteriorate with metastasis.Using deep-learning in the detection of lymph node metastasis can facilitate the noninvasive calculation of the likelihood of such metastasis,thereby providing clinicians with crucial information to enhance diagnostic precision and ultimately improve patient survival and prognosis.Methods In total,623 eligible patients were recruited from two medical institutions.Seven deep learning models,namely Alex,GoogLeNet,Resnet18,Resnet101,Vgg16,Vgg19,and MobileNetv3(small),were utilized to extract deep image histological features.The dimensionality of the extracted features was then reduced using the Spearman correlation coefficient(r≥0.9)and Least Absolute Shrinkage and Selection Operator.Eleven machine learning methods,namely Support Vector Machine,K-nearest neighbor,Random Forest,Extra Trees,XGBoost,LightGBM,Naive Bayes,AdaBoost,Gradient Boosting Decision Tree,Linear Regression,and Multilayer Perceptron,were employed to construct classification prediction models for the filtered final features.The diagnostic performances of the models were assessed using various metrics,including accuracy,area under the receiver operating characteristic curve,sensitivity,specificity,positive predictive value,and negative predictive value.Calibration and decision-curve analyses were also performed.Results The present study demonstrated that using deep radiomic features extracted from Vgg16,in conjunction with a prediction model constructed via a linear regression algorithm,effectively distinguished the status of mediastinal lymph nodes in patients with lung cancer.The performance of the model was evaluated based on various metrics,including accuracy,area under the receiver operating characteristic curve,sensitivity,specificity,positive predictive value,and negative predictive value,which yielded values of 0.808,0.834,0.851,0.745,0.829,and 0.776,respectively.The validation set of the model was assessed using clinical decision curves,calibration curves,and confusion matrices,which collectively demonstrated the model's stability and accuracy.Conclusion In this study,information on the deep radiomics of Vgg16 was obtained from computed tomography images,and the linear regression method was able to accurately diagnose mediastinal lymph node metastases in patients with lung cancer.