The significant threat of wildfires to forest ecology and biodiversity,particularly in tropical and subtropical regions,underscores the necessity for advanced predictive models amidst shifting climate patterns.There i...The significant threat of wildfires to forest ecology and biodiversity,particularly in tropical and subtropical regions,underscores the necessity for advanced predictive models amidst shifting climate patterns.There is a need to evaluate and enhance wildfire prediction methods,focusing on their application during extended periods of intense heat and drought.This study reviews various wildfire modelling approaches,including traditional physical,semi-empirical,numerical,and emerging machine learning(ML)-based models.We critically assess these models’capabilities in predicting fire susceptibility and post-ignition spread,highlighting their strengths and limitations.Our findings indicate that while traditional models provide foundational insights,they often fall short in dynamically estimating parameters and predicting ignition events.Cellular automata models,despite their potential,face challenges in data integration and computational demands.Conversely,ML models demonstrate superior efficiency and accuracy by leveraging diverse datasets,though they encounter interpretability issues.This review recommends hybrid modelling approaches that integrate multiple methods to harness their combined strengths.By incorporating data assimilation techniques with dynamic forecasting models,the predictive capabilities of ML-based predictions can be significantly enhanced.This review underscores the necessity for continued refinement of these models to ensure their reliability in real-world applications,ultimately contributing to more effective wildfire mitigation and management strategies.Future research should focus on improving hybrid models and exploring new data integration methods to advance predictive capabilities.展开更多
BACKGROUND Colorectal laterally spreading tumors(LSTs)are best treated with endoscopic submucosal dissection or endoscopic mucosal resection.AIM To analyze the clinicopathological and endoscopic profiles of colorectal...BACKGROUND Colorectal laterally spreading tumors(LSTs)are best treated with endoscopic submucosal dissection or endoscopic mucosal resection.AIM To analyze the clinicopathological and endoscopic profiles of colorectal LSTs,determine predictive factors for high-grade dysplasia(HGD)/carcinoma(CA),submucosal invasion,and complications.METHODS We retrospectively assessed the endoscopic and histological characteristics of 375 colorectal LSTs at our hospital between January 2016 and December 2023.We performed univariate and multivariate analysis to identify risk factors associated with HGD/CA,submucosal invasion and complications.RESULTS The numbers of granular(LST-G)and non-granular LST(LST-NG)were 260 and 115,respectively.The rates of low-grade dysplasia and HGD/CA were 60.3%and 39.7%,respectively.Multivariate analysis indicated that a tumor size≥30 mm[odds ratio(OR)=1.934,P=0.032],LST granular nodular mixed type(OR=2.100,P=0.005),and LST non-granular pseudo depressed type(NG-PD)(OR=3.016,P=0.015)were independent risk factors significantly associated with higher odds of HGD/CA.NG-PD(OR=6.506,P=0.001),tumor size(20-29 mm)(OR=2.631,P=0.036)and tumor size≥30 mm(OR=3.449,P=0.016)were associated with increased odds of submucosal invasion.Tumor size≥30 mm(OR=4.888,P=0.003)was a particularly important predictor of complications.A nomogram model demonstrated a satisfactory fit,with an area under the receiver operating characteristic curve of 0.716(95%confidence interval:0.653-0.780),indicating strong predictive performance.CONCLUSION The novel nomogram incorporating tumor size,location,and morphology predicted HGD/CA during endoscopic resection for LSTs.NG-PD lesions larger than 20 mm were more likely to invade the submucosa.Tumor size≥30 mm was an important predictor of complications.展开更多
Droplet impact dynamics on solid surfaces,which are ubiquitously present in aerospace engineering,energy systems,agricultural production,etc.,involve complex fluid–structure interactions.Herein,we employ a single-cam...Droplet impact dynamics on solid surfaces,which are ubiquitously present in aerospace engineering,energy systems,agricultural production,etc.,involve complex fluid–structure interactions.Herein,we employ a single-camera high-speed threedimensional digital image correlation system to quantify the full-field deformations of flexible thin films during droplet impact dynamics.Experimental results revealed that the substrate flexibility not only reduces the maximum spreading diameter by 10%but also modulates rebound dynamics via energy competition between kinetic energy and surface adhesion energy,suggesting that coupled deformation of the solid–fluid interface plays an important role in the dynamic progress.We propose the structure-coupled response number(Sn),a governing dimensionless parameter unifying droplet spreading on both rigid and flexible films,validated by a universal 1/2 scaling law.A theoretical criterion for droplet rebound on hydrophobic flexible thin films is derived and experimentally demonstrated,which achieves the precise control of droplet rebound/non-rebound mode.This work bridges the theories of droplet impact dynamics on rigid and flexible substrates,offering a robust strategy to govern the droplet impact behaviors.展开更多
基金funding enabled and organized by CAUL and its Member Institutions.
文摘The significant threat of wildfires to forest ecology and biodiversity,particularly in tropical and subtropical regions,underscores the necessity for advanced predictive models amidst shifting climate patterns.There is a need to evaluate and enhance wildfire prediction methods,focusing on their application during extended periods of intense heat and drought.This study reviews various wildfire modelling approaches,including traditional physical,semi-empirical,numerical,and emerging machine learning(ML)-based models.We critically assess these models’capabilities in predicting fire susceptibility and post-ignition spread,highlighting their strengths and limitations.Our findings indicate that while traditional models provide foundational insights,they often fall short in dynamically estimating parameters and predicting ignition events.Cellular automata models,despite their potential,face challenges in data integration and computational demands.Conversely,ML models demonstrate superior efficiency and accuracy by leveraging diverse datasets,though they encounter interpretability issues.This review recommends hybrid modelling approaches that integrate multiple methods to harness their combined strengths.By incorporating data assimilation techniques with dynamic forecasting models,the predictive capabilities of ML-based predictions can be significantly enhanced.This review underscores the necessity for continued refinement of these models to ensure their reliability in real-world applications,ultimately contributing to more effective wildfire mitigation and management strategies.Future research should focus on improving hybrid models and exploring new data integration methods to advance predictive capabilities.
文摘BACKGROUND Colorectal laterally spreading tumors(LSTs)are best treated with endoscopic submucosal dissection or endoscopic mucosal resection.AIM To analyze the clinicopathological and endoscopic profiles of colorectal LSTs,determine predictive factors for high-grade dysplasia(HGD)/carcinoma(CA),submucosal invasion,and complications.METHODS We retrospectively assessed the endoscopic and histological characteristics of 375 colorectal LSTs at our hospital between January 2016 and December 2023.We performed univariate and multivariate analysis to identify risk factors associated with HGD/CA,submucosal invasion and complications.RESULTS The numbers of granular(LST-G)and non-granular LST(LST-NG)were 260 and 115,respectively.The rates of low-grade dysplasia and HGD/CA were 60.3%and 39.7%,respectively.Multivariate analysis indicated that a tumor size≥30 mm[odds ratio(OR)=1.934,P=0.032],LST granular nodular mixed type(OR=2.100,P=0.005),and LST non-granular pseudo depressed type(NG-PD)(OR=3.016,P=0.015)were independent risk factors significantly associated with higher odds of HGD/CA.NG-PD(OR=6.506,P=0.001),tumor size(20-29 mm)(OR=2.631,P=0.036)and tumor size≥30 mm(OR=3.449,P=0.016)were associated with increased odds of submucosal invasion.Tumor size≥30 mm(OR=4.888,P=0.003)was a particularly important predictor of complications.A nomogram model demonstrated a satisfactory fit,with an area under the receiver operating characteristic curve of 0.716(95%confidence interval:0.653-0.780),indicating strong predictive performance.CONCLUSION The novel nomogram incorporating tumor size,location,and morphology predicted HGD/CA during endoscopic resection for LSTs.NG-PD lesions larger than 20 mm were more likely to invade the submucosa.Tumor size≥30 mm was an important predictor of complications.
基金supported by the National Key R&D Program of China(grant nos.2022YFF0503500 and 2022YFA1203200)the Guangdong Basic and Applied Basic Research Foundation(grant no.2023A1515011784)+2 种基金the National Natural Science Foundation of China(grant no.12032019)the Strategic Priority Research Program of Chinese Academy of Sciences(grant nos.XDB0620101 and XDB0620103)the Youth Innovation Promotion Association,Chinese Academy of Sciences(no.2020020).
文摘Droplet impact dynamics on solid surfaces,which are ubiquitously present in aerospace engineering,energy systems,agricultural production,etc.,involve complex fluid–structure interactions.Herein,we employ a single-camera high-speed threedimensional digital image correlation system to quantify the full-field deformations of flexible thin films during droplet impact dynamics.Experimental results revealed that the substrate flexibility not only reduces the maximum spreading diameter by 10%but also modulates rebound dynamics via energy competition between kinetic energy and surface adhesion energy,suggesting that coupled deformation of the solid–fluid interface plays an important role in the dynamic progress.We propose the structure-coupled response number(Sn),a governing dimensionless parameter unifying droplet spreading on both rigid and flexible films,validated by a universal 1/2 scaling law.A theoretical criterion for droplet rebound on hydrophobic flexible thin films is derived and experimentally demonstrated,which achieves the precise control of droplet rebound/non-rebound mode.This work bridges the theories of droplet impact dynamics on rigid and flexible substrates,offering a robust strategy to govern the droplet impact behaviors.
