An experimental platform with bracket structures,cables,parallel computer and imaging system is designed for defects detecting on steel rails. Meanwhile,an improved gradient descent algorithm based on a self-adaptive ...An experimental platform with bracket structures,cables,parallel computer and imaging system is designed for defects detecting on steel rails. Meanwhile,an improved gradient descent algorithm based on a self-adaptive learning rate and a fixed momentum factor is developed to train back-propagation neural network for accurate and efficient defects classifications. Detection results of rolling scar defects show that such detection system can achieve accurate positioning to defects edges for its improved noise suppression. More precise characteristic parameters of defects can also be extracted.Furthermore,defects classification is adopted to remedy the limitations of low convergence rate and local minimum. It can also attain the optimal training precision of 0. 00926 with the least 96 iterations. Finally,an enhanced identification rate of 95% has been confirmed for defects by using the detection system. It will also be positive in producing high-quality steel rails and guaranteeing the national transport safety.展开更多
Background:Methylene blue is the most commonly used tracer for sentinel lymph node(SLN)biopsy(SLNB)in China.This study aimed to investigate the feasibility of clinical application of SLNB using methylene blue dye(MBD)...Background:Methylene blue is the most commonly used tracer for sentinel lymph node(SLN)biopsy(SLNB)in China.This study aimed to investigate the feasibility of clinical application of SLNB using methylene blue dye(MBD)for early breast cancer and the prognosis of patients with different SLN and non-SLN statuses.Methods:We retrospectively analyzed the clinicopathological data of patients with early breast cancer treated at the Peking University First Hospital between 2013 and 2018.We calculated the SLN identification rate(IR)in SLNB with MBD and the false-negative rate(FNR),and analyzed the prognosis of patients with different SLN and non-SLN statuses using Kaplan-Meier curves.Results:Between January 2013 and December 2018,1603 patients with early breast cancer underwent SLNB with MBD.The SLN IR was 95.8%(1536/1603).Two SLNs(median)were detected per patient.There were significant differences in FNR between patients with SLN micrometastasis and macrometastasis(19.0%vs.4.5%,χ^(2)=12.771,P<0.001).Chi-square test showed that there were significant differences in SLN successful detection rates among patients with different vascular tumor embolism status(96.3%vs.90.8%,χ^(2)=9.013,P=0.003)and tumor(T)stages(96.6%vs.94.1%,χ^(2)=5.189,P=0.023).Multivariate analysis showed that vascular tumor embolism was the only independent factor for SLN successful detection(odds ratio:0.440,95%confidence interval:0.224-0.862,P=0.017).Survival analysis showed a significant difference in disease-free survival(DFS)between patients with non-SLN metastasis and patients without non-SLN metastasis(P=0.006).Conclusion:Our single-center data show that,as a commonly used tracer in SLNB in China,MBD has an acceptable SLN IR and a low FNR in frozen sections.This finding is consistent with reports of dual tracer-guided SLNB.Positive SLNs with non-SLN metastasis are associated with DFS.展开更多
基金Supported by the National Natural Science Foundation of China(No.51174151)the Key Scientific Research Project of Education Department of Hubei Province(No.D20151102)the Key Scientific and Technological Project of Wuhan Technology Bureau(No.2014010202010088)
文摘An experimental platform with bracket structures,cables,parallel computer and imaging system is designed for defects detecting on steel rails. Meanwhile,an improved gradient descent algorithm based on a self-adaptive learning rate and a fixed momentum factor is developed to train back-propagation neural network for accurate and efficient defects classifications. Detection results of rolling scar defects show that such detection system can achieve accurate positioning to defects edges for its improved noise suppression. More precise characteristic parameters of defects can also be extracted.Furthermore,defects classification is adopted to remedy the limitations of low convergence rate and local minimum. It can also attain the optimal training precision of 0. 00926 with the least 96 iterations. Finally,an enhanced identification rate of 95% has been confirmed for defects by using the detection system. It will also be positive in producing high-quality steel rails and guaranteeing the national transport safety.
基金Beijing Medical Award Foundation(No.YXJL-2016-0040-0065)Beijing Medical Award Foundation"Clinical efficacy of liquid biopsy for breast cancer"(No.2017-2019)+2 种基金Beijing Medical Award Foundation"Precision medical research on breast cancer"Beijing Medical Award Foundation Youth Program(No.2018-0304)National Key R&D Program of China(No.2016YFC0901302)。
文摘Background:Methylene blue is the most commonly used tracer for sentinel lymph node(SLN)biopsy(SLNB)in China.This study aimed to investigate the feasibility of clinical application of SLNB using methylene blue dye(MBD)for early breast cancer and the prognosis of patients with different SLN and non-SLN statuses.Methods:We retrospectively analyzed the clinicopathological data of patients with early breast cancer treated at the Peking University First Hospital between 2013 and 2018.We calculated the SLN identification rate(IR)in SLNB with MBD and the false-negative rate(FNR),and analyzed the prognosis of patients with different SLN and non-SLN statuses using Kaplan-Meier curves.Results:Between January 2013 and December 2018,1603 patients with early breast cancer underwent SLNB with MBD.The SLN IR was 95.8%(1536/1603).Two SLNs(median)were detected per patient.There were significant differences in FNR between patients with SLN micrometastasis and macrometastasis(19.0%vs.4.5%,χ^(2)=12.771,P<0.001).Chi-square test showed that there were significant differences in SLN successful detection rates among patients with different vascular tumor embolism status(96.3%vs.90.8%,χ^(2)=9.013,P=0.003)and tumor(T)stages(96.6%vs.94.1%,χ^(2)=5.189,P=0.023).Multivariate analysis showed that vascular tumor embolism was the only independent factor for SLN successful detection(odds ratio:0.440,95%confidence interval:0.224-0.862,P=0.017).Survival analysis showed a significant difference in disease-free survival(DFS)between patients with non-SLN metastasis and patients without non-SLN metastasis(P=0.006).Conclusion:Our single-center data show that,as a commonly used tracer in SLNB in China,MBD has an acceptable SLN IR and a low FNR in frozen sections.This finding is consistent with reports of dual tracer-guided SLNB.Positive SLNs with non-SLN metastasis are associated with DFS.
