The operating state of bearing affects the performance of rotating machinery;thus,how to accurately extract features from the original vibration signals and recognise the faulty parts as early as possible is very crit...The operating state of bearing affects the performance of rotating machinery;thus,how to accurately extract features from the original vibration signals and recognise the faulty parts as early as possible is very critical.In this study,the one‐dimensional ternary model which has been proved to be an effective statistical method in feature selection is introduced and shapelet transformation is proposed to calculate the parameter of one‐dimensional ternary model that is usually selected by trial and error.Then XGBoost is used to recognise the faults from the obtained features,and artificial bee colony algorithm(ABC)is introduced to optimise the parameters of XGBoost.Moreover,for improving the performance of intelligent algorithm,an improved strategy where the evolution is guided by the probability that the optimal solution appears in certain solution space is proposed.The experimental results based on the failure vibration signal samples show that the average accuracy of fault signal recognition can reach 97%,which is much higher than the ones corresponding to traditional extraction strategies.And with the help of improved ABC algorithm,the performance of XGBoost classifier could be optimised;the accuracy could be improved from 97.02%to 98.60%compared with the traditional classification strategy.展开更多
Objective:To investigate the differences of apparent diffusion coefficient and transverse relaxation rate(R_(2)^(*))in magnetic resonance imaging(MRI)between micro hepatocellular carcinoma and nodules of cirrhosis.Met...Objective:To investigate the differences of apparent diffusion coefficient and transverse relaxation rate(R_(2)^(*))in magnetic resonance imaging(MRI)between micro hepatocellular carcinoma and nodules of cirrhosis.Method:The clinical data of 68 patients with micro hepatocellular carcinoma(76 lesions)and 45 patients with nodular cirrhosis(48 lesions)were retrospectively analyzed.Diffusion weighted imaging and R_(2)^(*)imaging were performed on all patients.The differences of apparent diffusion coefficient and R_(2)^(*)values in patients with micro hepatocellular carcinoma and nodular cirrhosis were compared.Receiver operating characteristic(ROC)curves were drawn to evaluate the diagnostic efficacy of apparent diffusion coefficient values and R_(2)^(*)values for microhepatocellular carcinoma.Result:Compared with nodules of cirrhosis,the mean apparent diffusion coefficient and R_(2)^(*)value of micro-hepatocellular carcinoma were significantly decreased(P<0.05).The diagnosis threshold of apparent diffusion coefficient is at 1.35×10^(-3)mm^(2)/s,and the sensitivity for the diagnosis of micro-hepatocellular carcinoma was 80.88%,and the specificity was 91.11%,and the corresponding area under the ROC curve was 0.88;The diagnostic threshold for R_(2)^(*)value was at 53.96 Hz,and the sensitivity for the diagnosis of small hepatocellular carcinoma was 91.18%,and the specificity was 77.78%,and the corresponding area under the ROC curve was 0.84.Conclusion:MRI apparent diffusion coefficient value and R_(2)^(*)value can be used to differentiate and diagnose micro hepatocellular carcinoma and nodules of cirrhosis,in which the apparent diffusion coefficient of nodules of cirrhosis was less than 1.35×10^(-3)mm^(2)/s,R_(2)^(*)values were lower than 53.96 Hz may indicate the occurrence of nodular canceration.展开更多
Objective:To evaluate the value of DWI,DCE-MRI and magnetic resonance imaging(MRI)in the diagnosis of endometrial cancer.Method:The MRI,DWI and DCE-MRI imaging data of 80 patients with suspected endometrial cancer wer...Objective:To evaluate the value of DWI,DCE-MRI and magnetic resonance imaging(MRI)in the diagnosis of endometrial cancer.Method:The MRI,DWI and DCE-MRI imaging data of 80 patients with suspected endometrial cancer were analyzed.The diagnostic value of MRI,DWI and DCE-MRI in endometrial cancer was analyzed based on the postoperative pathological diagnosis results.The diagnostic efficacy of quantitative parameters Ktrans value,Kep value,Ve value and ADC in endometrial cancer was analyzed by ROC.Result:Among the 80 patients,61 had endometrial carcinoma and 19 had benign endometrial disease.The accuracy of MRI,DWI and DCE-MRI in the diagnosis of endometrial cancer was 57.38%,63.93%and 80.33%,respectively.The specificity was 78.95%,82.41%and 84.21%,respectively.The ADC values of endometrial cancer patients were lower than those of benign patients(P<0.05),and the values of Ktrans,Kep and Ve were higher than those of benign patients(P<0.05).The ktrans and ADC in the diagnosis of endometrial cancer were higher,which was 0.922(95%CI:0.864-0.992,P=0.000)and 0.872(95%CI:0.767-0.977.P=0.000),respectively.Conclusion:DWI and DCE-MRI had high value in the diagnosis of endometrial cancer.Its parameters,Ktrans and ADC,can be used as quantitative indicators for the early diagnosis of endometrial cancer.展开更多
This paper presents a new algorithm for generation of attack signatures based on sequence alignment. The algorithm is composed of two parts: a local alignment algorithm-GASBSLA (Generation of Attack Signatures Based o...This paper presents a new algorithm for generation of attack signatures based on sequence alignment. The algorithm is composed of two parts: a local alignment algorithm-GASBSLA (Generation of Attack Signatures Based on Sequence Local Alignment) and a multi-sequence alignment algorithm-TGMSA (Tri-stage Gradual Multi-Sequence Alignment). With the inspiration of sequence alignment used in Bioinformatics, GASBSLA replaces global alignment and constant weight penalty model by local alignment and affine penalty model to improve the generality of attack signatures. TGMSA presents a new pruning policy to make the algorithm more insensitive to noises in the generation of attack signatures. In this paper, GASBSLA and TGMSA are described in detail and validated by experiments.展开更多
In this paper, we analyze the survivability of Mobile Ad Hoc Network systemically and give a detailed description of the survivability issues related to the MANET. We begin our work with analyzing the requirements of ...In this paper, we analyze the survivability of Mobile Ad Hoc Network systemically and give a detailed description of the survivability issues related to the MANET. We begin our work with analyzing the requirements of survivability of ad hoc network, and then we classify the impacts that affect survivability into three categories: dynamic topology, faults and attacks. The impacts of these factors are analyzed individually. A simulation environment for the MANET towards survivability is designed and implemented as well. Experiments that under the requirements and the impacts we de-clared are done based on this environment.展开更多
Narrowband photodetection is an important measurement technique for material analysis and sensing,for example,nondispersive infrared sensing technique.Both photoactive material engineering and nanophotonic filtering s...Narrowband photodetection is an important measurement technique for material analysis and sensing,for example,nondispersive infrared sensing technique.Both photoactive material engineering and nanophotonic filtering schemes have been explored to realize wavelength-selective photodetection,while most devices have a responsive bandwidth larger than 2%of the operating wavelength,limiting sensing performance.Near-infrared photodetection with a bandwidth of less than 0.2%of the operating wavelength was demonstrated experimentally in Au/Si Schottky nanojunctions.A minimum linewidth of photoelectric response down to 2.6 nm was obtained at a wavelength of 1550 nm by carefully tailing the absorptive and radiative loss in the nanostructures.Multiple functions were achieved on chip with the corrugated Au film,including narrowband resonance,light harvesting for sensing and photodetection,and electrodes for hot electron emission.Benefiting from such a unity integration with in situ photoelectric conversion of the optical sensing signal and the ultranarrowband resonance,self-contained on-chip biosensing via simple intensity interrogation was demonstrated with a limit of detection down to 0.0047%in concentration for glucose solution and 150 ng∕m L for rabbit IgG.Promising potential of this technique is expected for the applications in on-site sensing,spectroscopy,spectral imaging,etc.展开更多
Li-CO_(2)batteries(LCBs)suffer from high overpotentials caused by sluggish CO_(2)reaction kinetics.This work designs a Te-doped Fe_(3)O_(4)(Te-Fe_(3)O_(4))flower-like microsphere catalyst to lower the overpotential an...Li-CO_(2)batteries(LCBs)suffer from high overpotentials caused by sluggish CO_(2)reaction kinetics.This work designs a Te-doped Fe_(3)O_(4)(Te-Fe_(3)O_(4))flower-like microsphere catalyst to lower the overpotential and improve the reversibility of LCBs.Experimental results reveal that Te doping modifies the electronic structure of Fe_(3)O_(4)and reduces the overpotential.The stable Te-O bond between Te and C_(2)O^(2-)_(4)could effectively inhibit the dispro-portionation reaction of the latter,enabling the Te-Fe_(3)O_(4)cathodes to exhibit a remarkable capacity(9485 mAh g^(-1))and a long cycling life(155 cycles)with an overpotential of 1.21 V and an energy efficiency of about 80%at a high current density(2000 mA g^(-1)).Through the interaction between Te and Li_(2)C_(2)O_(4)to inhibit the dispro-portionation reaction,this work successfully achieves long-term cycling of LCBs with low overpotential at high current density.展开更多
In mammals,the neonatal heart can regenerate upon injury within a short time after birth,while adults lose this ability.Metabolic reprogramming has been demonstrated to be critical for cardiomyocyte proliferation in t...In mammals,the neonatal heart can regenerate upon injury within a short time after birth,while adults lose this ability.Metabolic reprogramming has been demonstrated to be critical for cardiomyocyte proliferation in the neonatal heart.Here,we reveal that cardiac metabolic reprogramming could be regulated by altering global protein lactylation.By performing 4D label-free proteomics and lysine lactylation(Kla)omics analyses in mouse hearts at postnatal days 1,5,and 7,2297 Kla sites from 980 proteins are identified,among which 1262 Kla sites from 409 proteins are quantified.Functional clustering analysis reveals that the proteins with altered Kla sites are mainly involved in metabolic processes.The expression and Kla levels of proteins in glycolysis show a positive correlation while a negative correlation in fatty acid oxidation.Furthermore,we verify the Kla levels of several differentially modified proteins,including ACAT1,ACADL,ACADVL,PFKM,PKM,and NPM1.Overall,our study reports a comprehensive Kla map in the neonatal mouse heart,which will help to understand the regulatory network of metabolic reprogramming and cardiac regeneration.展开更多
This study investigates the bulk density,sintering behaviour,and microwave dielectric properties of the MgO-2B_(2)O_(3) series ceramics synthesised by solid-state reaction.According to the X-ray diffraction and micros...This study investigates the bulk density,sintering behaviour,and microwave dielectric properties of the MgO-2B_(2)O_(3) series ceramics synthesised by solid-state reaction.According to the X-ray diffraction and microstructural analyses,the as-prepared MgO-2B_(2)O_(3) ceramics possess a single-phase structure with a rod-like morphology.The effects of different quantities of H_(3)BO_(3) and BaCu(B_(2)O5)(BCB)on the bulk density,sintering behaviour,and microwave dielectric properties of the MgO-2B_(2)O_(3) ceramics were investigated.Accordingly,the optimal sintering temperature was obtained by adding 30 wt%H_(3)BO_(3) and 8 wt%BCB.We also reduced the sintering temperature to 825°C.Furthermore,the addition of 40 wt%H_(3)BO_(3) and 4 wt%BCB increased the quality factor,permittivity,and temperature coefficient of resonance frequency of MgO-2B_(2)O_(3) to 44,306 GHz(at 15 GHz),5.1,and-32 ppm/℃,respectively.These properties make MgO-2B_(2)O_(3) a viable low-temperature co-fired ceramic with broad applications in microwave dielectrics.展开更多
Glycerol is a simple polyhydroxy-three-carbon compound that is highly hydrophilic and is considered a central intermediate metabolite for cellular metabolism.It is found at the crossroads among several metabolic pathw...Glycerol is a simple polyhydroxy-three-carbon compound that is highly hydrophilic and is considered a central intermediate metabolite for cellular metabolism.It is found at the crossroads among several metabolic pathways,such as glycolysis and gluconeogenesis(for carbohydrates),lipogenesis,lipolysis,and glyceroneogenesis(for lipids),pentose-phosphate pathway(for nucleic acids),amino acid synthesis,and even energy generation by oxidative phosphorylation[1](Fig.1).展开更多
Rapid antimicrobial susceptibility testing(AST)is urgently needed to slow down the emergence of antibioticresistant bacteria and treat infections with correct antibiotics.Stimulated Raman scattering(SRS)microscopy is ...Rapid antimicrobial susceptibility testing(AST)is urgently needed to slow down the emergence of antibioticresistant bacteria and treat infections with correct antibiotics.Stimulated Raman scattering(SRS)microscopy is a technique that enables rapid chemical-bond imaging with sub-cellular resolution.It can obtain the AST results with a single bacterium resolution.Although the SRS imaging assay is relatively fast,taking less than 2 h,the calculation of single-cell metabolism inactivation concentration(SC-MIC)is performed manually and takes a long time.The bottleneck tasks that hinder the SC-MIC throughput include bacterial segmentation and intensity thresholding.To address these issues,we devised a hybrid algorithm to segment single bacteria from SRS images with automatic thresholding.Our proposed method comprises a U-Net convolutional neural network(CNN),DropBlock,and secondary segmentation post-processing.Our results show that SC-MIC calculation can be accomplished within 1 min and more accurate segmentation results using deep learning-based bacterial segmentation method,which is essential for its clinical applications.展开更多
基金National Nature Science Foundation of China,Grant/Award Number:U1813201the Key Scientific Research Projects of Henan Province,Grant/Award Number:22A413011+2 种基金the Training Program for Young Teachers in Universities of Henan Province,Grant/Award Number:2020GGJS137Henan Province Science and Technology R&D projects,Grant/Award Number:202102210135,212102310547 and 212102210080High‐end foreign expert program of Ministry of Science and Technology,Grant/Award Number:G2021026006L。
文摘The operating state of bearing affects the performance of rotating machinery;thus,how to accurately extract features from the original vibration signals and recognise the faulty parts as early as possible is very critical.In this study,the one‐dimensional ternary model which has been proved to be an effective statistical method in feature selection is introduced and shapelet transformation is proposed to calculate the parameter of one‐dimensional ternary model that is usually selected by trial and error.Then XGBoost is used to recognise the faults from the obtained features,and artificial bee colony algorithm(ABC)is introduced to optimise the parameters of XGBoost.Moreover,for improving the performance of intelligent algorithm,an improved strategy where the evolution is guided by the probability that the optimal solution appears in certain solution space is proposed.The experimental results based on the failure vibration signal samples show that the average accuracy of fault signal recognition can reach 97%,which is much higher than the ones corresponding to traditional extraction strategies.And with the help of improved ABC algorithm,the performance of XGBoost classifier could be optimised;the accuracy could be improved from 97.02%to 98.60%compared with the traditional classification strategy.
文摘Objective:To investigate the differences of apparent diffusion coefficient and transverse relaxation rate(R_(2)^(*))in magnetic resonance imaging(MRI)between micro hepatocellular carcinoma and nodules of cirrhosis.Method:The clinical data of 68 patients with micro hepatocellular carcinoma(76 lesions)and 45 patients with nodular cirrhosis(48 lesions)were retrospectively analyzed.Diffusion weighted imaging and R_(2)^(*)imaging were performed on all patients.The differences of apparent diffusion coefficient and R_(2)^(*)values in patients with micro hepatocellular carcinoma and nodular cirrhosis were compared.Receiver operating characteristic(ROC)curves were drawn to evaluate the diagnostic efficacy of apparent diffusion coefficient values and R_(2)^(*)values for microhepatocellular carcinoma.Result:Compared with nodules of cirrhosis,the mean apparent diffusion coefficient and R_(2)^(*)value of micro-hepatocellular carcinoma were significantly decreased(P<0.05).The diagnosis threshold of apparent diffusion coefficient is at 1.35×10^(-3)mm^(2)/s,and the sensitivity for the diagnosis of micro-hepatocellular carcinoma was 80.88%,and the specificity was 91.11%,and the corresponding area under the ROC curve was 0.88;The diagnostic threshold for R_(2)^(*)value was at 53.96 Hz,and the sensitivity for the diagnosis of small hepatocellular carcinoma was 91.18%,and the specificity was 77.78%,and the corresponding area under the ROC curve was 0.84.Conclusion:MRI apparent diffusion coefficient value and R_(2)^(*)value can be used to differentiate and diagnose micro hepatocellular carcinoma and nodules of cirrhosis,in which the apparent diffusion coefficient of nodules of cirrhosis was less than 1.35×10^(-3)mm^(2)/s,R_(2)^(*)values were lower than 53.96 Hz may indicate the occurrence of nodular canceration.
文摘Objective:To evaluate the value of DWI,DCE-MRI and magnetic resonance imaging(MRI)in the diagnosis of endometrial cancer.Method:The MRI,DWI and DCE-MRI imaging data of 80 patients with suspected endometrial cancer were analyzed.The diagnostic value of MRI,DWI and DCE-MRI in endometrial cancer was analyzed based on the postoperative pathological diagnosis results.The diagnostic efficacy of quantitative parameters Ktrans value,Kep value,Ve value and ADC in endometrial cancer was analyzed by ROC.Result:Among the 80 patients,61 had endometrial carcinoma and 19 had benign endometrial disease.The accuracy of MRI,DWI and DCE-MRI in the diagnosis of endometrial cancer was 57.38%,63.93%and 80.33%,respectively.The specificity was 78.95%,82.41%and 84.21%,respectively.The ADC values of endometrial cancer patients were lower than those of benign patients(P<0.05),and the values of Ktrans,Kep and Ve were higher than those of benign patients(P<0.05).The ktrans and ADC in the diagnosis of endometrial cancer were higher,which was 0.922(95%CI:0.864-0.992,P=0.000)and 0.872(95%CI:0.767-0.977.P=0.000),respectively.Conclusion:DWI and DCE-MRI had high value in the diagnosis of endometrial cancer.Its parameters,Ktrans and ADC,can be used as quantitative indicators for the early diagnosis of endometrial cancer.
文摘This paper presents a new algorithm for generation of attack signatures based on sequence alignment. The algorithm is composed of two parts: a local alignment algorithm-GASBSLA (Generation of Attack Signatures Based on Sequence Local Alignment) and a multi-sequence alignment algorithm-TGMSA (Tri-stage Gradual Multi-Sequence Alignment). With the inspiration of sequence alignment used in Bioinformatics, GASBSLA replaces global alignment and constant weight penalty model by local alignment and affine penalty model to improve the generality of attack signatures. TGMSA presents a new pruning policy to make the algorithm more insensitive to noises in the generation of attack signatures. In this paper, GASBSLA and TGMSA are described in detail and validated by experiments.
文摘In this paper, we analyze the survivability of Mobile Ad Hoc Network systemically and give a detailed description of the survivability issues related to the MANET. We begin our work with analyzing the requirements of survivability of ad hoc network, and then we classify the impacts that affect survivability into three categories: dynamic topology, faults and attacks. The impacts of these factors are analyzed individually. A simulation environment for the MANET towards survivability is designed and implemented as well. Experiments that under the requirements and the impacts we de-clared are done based on this environment.
基金financial support from the National Natural Science Foundation of China(Grant Nos.62220106001 and 92050108)the Key Research and Development Program of Guangdong Province(Grant No.2023B0101200009)+3 种基金the Guangdong Science and Technology Program International Cooperation Program(Grant No.2021A0505030038)the Guangdong Basic and Applied Basic Research Foundation(Grant No.2022B1515020069)the Pearl River Talent Plan Program of Guangdong(Grant No.2019QN01X120)the technical support from Nano Fabrication Facility and Nano-X of SINANO,CAS
文摘Narrowband photodetection is an important measurement technique for material analysis and sensing,for example,nondispersive infrared sensing technique.Both photoactive material engineering and nanophotonic filtering schemes have been explored to realize wavelength-selective photodetection,while most devices have a responsive bandwidth larger than 2%of the operating wavelength,limiting sensing performance.Near-infrared photodetection with a bandwidth of less than 0.2%of the operating wavelength was demonstrated experimentally in Au/Si Schottky nanojunctions.A minimum linewidth of photoelectric response down to 2.6 nm was obtained at a wavelength of 1550 nm by carefully tailing the absorptive and radiative loss in the nanostructures.Multiple functions were achieved on chip with the corrugated Au film,including narrowband resonance,light harvesting for sensing and photodetection,and electrodes for hot electron emission.Benefiting from such a unity integration with in situ photoelectric conversion of the optical sensing signal and the ultranarrowband resonance,self-contained on-chip biosensing via simple intensity interrogation was demonstrated with a limit of detection down to 0.0047%in concentration for glucose solution and 150 ng∕m L for rabbit IgG.Promising potential of this technique is expected for the applications in on-site sensing,spectroscopy,spectral imaging,etc.
基金supported by the National Natural Science Foundation of China(22162004)the Natural Science Foundation of Guangxi Province(2022JJD120011)the Innovation Project of Guangxi Graduate Education(YCBZ2023012,YCSW2023115,YCBZ2023048).
文摘Li-CO_(2)batteries(LCBs)suffer from high overpotentials caused by sluggish CO_(2)reaction kinetics.This work designs a Te-doped Fe_(3)O_(4)(Te-Fe_(3)O_(4))flower-like microsphere catalyst to lower the overpotential and improve the reversibility of LCBs.Experimental results reveal that Te doping modifies the electronic structure of Fe_(3)O_(4)and reduces the overpotential.The stable Te-O bond between Te and C_(2)O^(2-)_(4)could effectively inhibit the dispro-portionation reaction of the latter,enabling the Te-Fe_(3)O_(4)cathodes to exhibit a remarkable capacity(9485 mAh g^(-1))and a long cycling life(155 cycles)with an overpotential of 1.21 V and an energy efficiency of about 80%at a high current density(2000 mA g^(-1)).Through the interaction between Te and Li_(2)C_(2)O_(4)to inhibit the dispro-portionation reaction,this work successfully achieves long-term cycling of LCBs with low overpotential at high current density.
文摘In mammals,the neonatal heart can regenerate upon injury within a short time after birth,while adults lose this ability.Metabolic reprogramming has been demonstrated to be critical for cardiomyocyte proliferation in the neonatal heart.Here,we reveal that cardiac metabolic reprogramming could be regulated by altering global protein lactylation.By performing 4D label-free proteomics and lysine lactylation(Kla)omics analyses in mouse hearts at postnatal days 1,5,and 7,2297 Kla sites from 980 proteins are identified,among which 1262 Kla sites from 409 proteins are quantified.Functional clustering analysis reveals that the proteins with altered Kla sites are mainly involved in metabolic processes.The expression and Kla levels of proteins in glycolysis show a positive correlation while a negative correlation in fatty acid oxidation.Furthermore,we verify the Kla levels of several differentially modified proteins,including ACAT1,ACADL,ACADVL,PFKM,PKM,and NPM1.Overall,our study reports a comprehensive Kla map in the neonatal mouse heart,which will help to understand the regulatory network of metabolic reprogramming and cardiac regeneration.
基金supported by the National Natural Science Foundation of China(Nos.61761015 and 12064007)the Natural Science Foundation of Guangxi(Nos.2018GXNSFFA050001,2017GXNSFDA198027,and 2017GXNSFFA198011)High Level Innovation Team and Outstanding Scholar Program of Guangxi Institutes.
文摘This study investigates the bulk density,sintering behaviour,and microwave dielectric properties of the MgO-2B_(2)O_(3) series ceramics synthesised by solid-state reaction.According to the X-ray diffraction and microstructural analyses,the as-prepared MgO-2B_(2)O_(3) ceramics possess a single-phase structure with a rod-like morphology.The effects of different quantities of H_(3)BO_(3) and BaCu(B_(2)O5)(BCB)on the bulk density,sintering behaviour,and microwave dielectric properties of the MgO-2B_(2)O_(3) ceramics were investigated.Accordingly,the optimal sintering temperature was obtained by adding 30 wt%H_(3)BO_(3) and 8 wt%BCB.We also reduced the sintering temperature to 825°C.Furthermore,the addition of 40 wt%H_(3)BO_(3) and 4 wt%BCB increased the quality factor,permittivity,and temperature coefficient of resonance frequency of MgO-2B_(2)O_(3) to 44,306 GHz(at 15 GHz),5.1,and-32 ppm/℃,respectively.These properties make MgO-2B_(2)O_(3) a viable low-temperature co-fired ceramic with broad applications in microwave dielectrics.
基金supported by the National Key Research and Development Program of China (2018YFC1004700)
文摘Glycerol is a simple polyhydroxy-three-carbon compound that is highly hydrophilic and is considered a central intermediate metabolite for cellular metabolism.It is found at the crossroads among several metabolic pathways,such as glycolysis and gluconeogenesis(for carbohydrates),lipogenesis,lipolysis,and glyceroneogenesis(for lipids),pentose-phosphate pathway(for nucleic acids),amino acid synthesis,and even energy generation by oxidative phosphorylation[1](Fig.1).
基金This research was supported by the Natural Science Foundation of Beijing(7204274)the National Natural Science Foundation of China(81901790).
文摘Rapid antimicrobial susceptibility testing(AST)is urgently needed to slow down the emergence of antibioticresistant bacteria and treat infections with correct antibiotics.Stimulated Raman scattering(SRS)microscopy is a technique that enables rapid chemical-bond imaging with sub-cellular resolution.It can obtain the AST results with a single bacterium resolution.Although the SRS imaging assay is relatively fast,taking less than 2 h,the calculation of single-cell metabolism inactivation concentration(SC-MIC)is performed manually and takes a long time.The bottleneck tasks that hinder the SC-MIC throughput include bacterial segmentation and intensity thresholding.To address these issues,we devised a hybrid algorithm to segment single bacteria from SRS images with automatic thresholding.Our proposed method comprises a U-Net convolutional neural network(CNN),DropBlock,and secondary segmentation post-processing.Our results show that SC-MIC calculation can be accomplished within 1 min and more accurate segmentation results using deep learning-based bacterial segmentation method,which is essential for its clinical applications.