Objective To explore value of peritumoral radiomics models for predicting lymphovascular invasion(LVI)of gastric adenocarcinoma based on portal-venous phase enhanced CT images.Methods Totally 351 patients with gastric adenocarcinoma were collected and randomly divided into training set(n=246)and test set(n=105)at a ratio of 7∶3.ROI of tumors were manually delineated on portal-venous phase enhanced CT images,then radiomics features of tumoral areas and peritumoral areas(1,3 and 5 mm expanded from lesions)were extracted,respectively.Clinical-CT,tumoral and peritumoral(1 mm,3 mm,5 mm)radiomics and comprehensive model nomograms were established,and their predictive performances for LVI were compared.Calibration curve was used to evaluate the consistency between the predicted and actual LVI of gastric adenocarcinoma,while decision curve analysis(DCA)was used to assess the net clinical benefit of each model.Results The area under the curve of clinical-CT,tumoral,peritumoral(1 mm,3 mm,5 mm)radiomics models and the comprehensive model for predicting LVI in training set was 0.741,0.732,0.713,0.728,0.708 and 0.755,respectively,which in test set was 0.748,0.725,0.759,0.724,0.704 and 0.764,respectively.The comprehensive model demonstrated the highest prediction efficiency,also good calibration and clinical practicability.Conclusion Portal-venous phase CT peritumoral radiomics models could be used to predict LVI of gastric adenocarcinoma.Combining with CT features,tumoral and optimal peritumoral radiomics features could further improve predictive efficiency.

Objective:To explore the research hotspots and trends in the radiomics field based on bibliometric for breast cancer.Methods:Publications related to radiomics for breast cancer during January 2012 to June 2024 were retrieved from the Web of Science Core Collection databases.Then,the R software's package"bibliometric"tool,VOSviewer and CiteSpace softwares were used to conduct visual analysis of radiomics-related bibliometric for breast cancer,which involved to publication trends,countries,institutions,journals,authors,keywords,and highlight keywords.Results:A total of 683 radiomics-related literatures for breast cancer were obtained,and 518 articles were finally included after non-English literatures and non-Article literatures were excluded.China and Sun Yat-sen University emerged as the leading country(326 articles)and institution(83 articles).The top three journals with the most of published articles were respectively,and.Professor Mao Ning has the largest number of articles,followed by Pinker Katja and Liang Changhong.The results of keywords and highlight keywords showed that current radiomics researches for breast cancer mainly focused on the applications of magnetic resonance imaging(MRI),machine learning,deep learning,artificial intelligence,and radiogenomics in prediction for diseases'prognosis,and assessment for treatment's response.Conclusion:The applications of machine learning,deep learning and multi-parameter imaging belongs to hot-research fields in assessment for personalized treatment.Future research directions would more and more tend to integrate artificial intelligence and multi-omics analysis,so as to promote prognostic assessment and personalized treatment.

Objective To explore value of peritumoral radiomics models for predicting lymphovascular invasion(LVI)of gastric adenocarcinoma based on portal-venous phase enhanced CT images.Methods Totally 351 patients with gastric adenocarcinoma were collected and randomly divided into training set(n=246)and test set(n=105)at a ratio of 7∶3.ROI of tumors were manually delineated on portal-venous phase enhanced CT images,then radiomics features of tumoral areas and peritumoral areas(1,3 and 5 mm expanded from lesions)were extracted,respectively.Clinical-CT,tumoral and peritumoral(1 mm,3 mm,5 mm)radiomics and comprehensive model nomograms were established,and their predictive performances for LVI were compared.Calibration curve was used to evaluate the consistency between the predicted and actual LVI of gastric adenocarcinoma,while decision curve analysis(DCA)was used to assess the net clinical benefit of each model.Results The area under the curve of clinical-CT,tumoral,peritumoral(1 mm,3 mm,5 mm)radiomics models and the comprehensive model for predicting LVI in training set was 0.741,0.732,0.713,0.728,0.708 and 0.755,respectively,which in test set was 0.748,0.725,0.759,0.724,0.704 and 0.764,respectively.The comprehensive model demonstrated the highest prediction efficiency,also good calibration and clinical practicability.Conclusion Portal-venous phase CT peritumoral radiomics models could be used to predict LVI of gastric adenocarcinoma.Combining with CT features,tumoral and optimal peritumoral radiomics features could further improve predictive efficiency.

Objective:To explore the research hotspots and trends in the radiomics field based on bibliometric for breast cancer.Methods:Publications related to radiomics for breast cancer during January 2012 to June 2024 were retrieved from the Web of Science Core Collection databases.Then,the R software's package"bibliometric"tool,VOSviewer and CiteSpace softwares were used to conduct visual analysis of radiomics-related bibliometric for breast cancer,which involved to publication trends,countries,institutions,journals,authors,keywords,and highlight keywords.Results:A total of 683 radiomics-related literatures for breast cancer were obtained,and 518 articles were finally included after non-English literatures and non-Article literatures were excluded.China and Sun Yat-sen University emerged as the leading country(326 articles)and institution(83 articles).The top three journals with the most of published articles were respectively,and.Professor Mao Ning has the largest number of articles,followed by Pinker Katja and Liang Changhong.The results of keywords and highlight keywords showed that current radiomics researches for breast cancer mainly focused on the applications of magnetic resonance imaging(MRI),machine learning,deep learning,artificial intelligence,and radiogenomics in prediction for diseases'prognosis,and assessment for treatment's response.Conclusion:The applications of machine learning,deep learning and multi-parameter imaging belongs to hot-research fields in assessment for personalized treatment.Future research directions would more and more tend to integrate artificial intelligence and multi-omics analysis,so as to promote prognostic assessment and personalized treatment.

Objective To use analytic hierarchy process(AHP)-entropy weight method to evaluate the quality of symbiotic Armillaria in Gastrodia elata Bl.Methods The physiological,growth and chemical indexes of 6 strains of G.elata symbiotic Armillaria were used as evaluation factors,and the weight coefficients of each index were calculated by the AHP-entropy weight method.A comprehensive quality evaluation system was established to evaluate the quality of Armillaria strains,and the results were verified by red G.elata and hybrid G.elata combined planting experiments.Results The quality evaluation results showed that extracellular xylanase activity,extracellular cellulase activity and total sugar weight were relatively high,among which strains N6 and N3 ranked the top.The results showed that strains N5 and N4 combined with red G.elata had better quality,that strains N6 and N3 combined with hybrid G.elata had better quality.Conclusion The correlation of extracellular xylanase activity and total sugar of cellulase activity should be paid attention to in the quality evaluation of symbiotic Armillaria in G.elata.The quality evaluation system constructed in this artilcle is basically consistent with the results of hybrid G.elata combined with hybrid G.elata experiment,which can provide references for the quality screening of symbiotic Armillaria in G.elata.

Objective To observe changes of CT quantitative indexes in patients with untreated chronic obstructive pulmonary disease(COPD)and relationships with pulmonary function indicators.Methods Totally 99 patients with untreated COPD were retrospectively enrolled.According to the degrees of airflow obstruction,the patients were divided into group A(GOLD grade 1 with mild obstruction,n=36),group B(GOLD 2 with moderate obstruction,n=37)and group C(GOLD 3 or 4 with obvious obstruction,n=26).The results of chest CT and pulmonary function tests conducted at the first diagnosis and the follow-up,as well as their correlations were analyzed.CT quantitative indicators included the whole lung volume,low-attenuation areas less than-950 percentage(LAA％),total number of vessels per 1 cm2 of lung surface area(Ntotal/LSA),total number of vessels with area less than 5 mm2 per 1 cm2 of lung surface area(N＜5mm2/LSA),the square root of the wall area of a hypothetical airway with a 10 mm internal perimeter(Pi10),the entire volume,wall thickness and wall area percentage(WA％)of airway wall,etc.,while results of pulmonary function tests included the forced expiratory volume in one second(FEV1)after administration of a bronchodilator,forced vital capacity(FVC),the ratio FEV1/FVC and FEV1 expressed as percent predicted(FEV1％).Results Compared with those at the first diagnosis,the follow-up results of FVC,FEV1,FEV1/FVC,Ntotal/LSA and N＜5mm2/LSA were lower,whereas LAA％,Pi10 and entire volume of airway were all higher in each group(all P＜0.05).Compared with those in group A,group B and C had decreased LAA％and increased Pi10,and the magnitude increased with the severity of airflow obstruction(all P＜0.05).LAA％,Pi1o and entire volume of airway wall were negatively correlated with pulmonary function indicators(all P＜0.05),while Ntotal/LSA and N＜5 mm2/LSA were positively correlated with pulmonary function indicators(all P＜0.05).Conclusion CT quantitative parameters,including LAA％,Ntotal/LSA,N＜5mm2/LSA,Pi10 and entire volume of airway wall were related to pulmonary function,which might reflect the longitudinal changes of airways and blood vessels in COPD patients.

Objective To observe the reliability of regional 4D flow and whole heart 4D flow cardiac MRI(CMRI)for measuring hemodynamic parameters of left ventricle.Methods Heart ultrasonography and CMRI were prospectively obtained in 31 healthy subjects.Hemodynamic parameters of left ventricle were measured using heart ultrasound,3-chamber 4D flow CMRI(based on inflow and outflow channel of left ventricle)and whole heart 4D flow CMRI,respectively.Intra-class correlation coefficient(ICC)was performed to evaluate the consistencies of the measured left ventricle hemodynamic parameters among the above 3 methods.Results Good consistencies of peak systolic velocity in aortic supravalvular/subvalvular,E peak diastolic velocity of mitral valve,supravalvular/subvalvular aortic pressure and aortic valve pressure gradient(all ICC＞0.75),while moderate consistency of A peak diastolic velocity of mitral valve(ICC=0.718)were found between heart ultrasound and 3-chamber 4D flow CMRI.Good consistencies of peak systolic velocity in aortic supravalvular/subvalvular,A peak diastolic velocity of mitral valve and supravalvular/subvalvular aortic pressure(all ICC＞0.75),while moderate consistencies of E peak diastolic velocity of mitral valve and aortic valve pressure gradient(ICC=0.600,0.628)were found between heart ultrasound and whole heart 4D flow CMRI.Meanwhile,good consistencies of the above parameters were found between 3-chamber 4D flow CMRI and whole heart 4D flow CMRI(all ICC＞0.75).Conclusion Measuring left ventricular hemodynamic parameters using local regional 4D flow and whole heart 4D flow CMRI were reliable,with good consistency with cardiac ultrasound.