Search

This study aimed to evaluate the risk factors for machine learning (ML) algorithms in predicting postoperative surgical site infection (SSI) following spine surgery. This prospective cohort study included 986 patients who underwent spine surgery at Taizhou People's Hospital Affiliated to Nanjing Medical University from January 2015 to October 2022. Supervised ML algorithms included support vector machine, logistic regression, random forest, XGboost, decision tree, k -nearest neighbour, and naïve Bayes (NB), which were tested and trained to develop a predicting model. The ML model performance was evaluated from the test dataset. We gradually analysed their accuracy, sensitivity, and specificity, as well as the positive predictive value, negative predictive value, and area under the curve. The rate of SSI was 9.33%. Using a backward stepwise approach, we identified that the remarkable risk factors predicting SSI in the multi-variate Cox regression analysis were age, body mass index, smoking, cerebrospinal fluid leakage, drain duration and pre-operative albumin level. Compared with other ML algorithms, the NB model had the highest performance in seven ML models, with an average area under the curve of 0.95, sensitivity of 0.78, specificity of 0.88, and accuracy of 0.87. The NB model in the ML algorithm had excellent calibration and accurately predicted the risk of SSI compared with the existing models, and might serve as an important tool for the early detection and treatment of SSI following spinal infection. [ABSTRACT FROM AUTHOR]

To explore the possibility of discriminating minimal residual disease (MRD) status in multiple myeloma (MM) based on magnetic resonance imaging (MRI) and identify optimal machine-learning methods to optimise the clinical treatment regimen. A total of 83 patients were analysed retrospectively. They were divided randomly into training and validation cohorts. The regions of interest were segmented and radiomics features were extracted and analysed on two sequences, including T1-weighted imaging (WI) and fat saturated (FS)-T2WI, and then radiomics models were built in the training cohort and evaluated in the validation cohort. Clinical characteristics were calculated to build a traditional model. A combined model was also built using the clinical characteristics and radiomics features. Classification accuracy was assessed using area under the curve (AUC) and F1 score. In the training cohort, only the bone marrow (BM) infiltrate ratio (p= 0.005) was retained after univariate and multivariable logistic regression analysis. In T1WI, the linear support vector machine (SVM) achieved the best performance compared to other classifiers, with AUCs of 0.811 and 0.708 and F1 scores of 0.792 and 0.696 in the training and validation cohorts, respectively. Similarly, in FS-T2WI sequence, linear SVM achieved the best performance with AUCs of 0.833 and 0.800 and F1 score of 0.833 and 0.800. The combined model constructed by the FS-T2WI-linear SVM and BM infiltrate ratio outperformed the traditional model (p= 0.050 and 0.012, Delong test), but showed no significant difference compared with the radiomics model (p= 0.798 and 0.855). The linear SVM-based machine-learning method can offer a non-invasive tool for discriminating MRD status in MM. • MRD is a new biomarker of multiple myeloma therapy and prolonged survival. • The related clinical characteristics are limited to discriminate MRD status. • Radiomics model achieves the best predictive performance for MRD status. [ABSTRACT FROM AUTHOR]

Haematopoietic stem cell transplantation (HSCT), a standard treatment for paediatric haematological diseases, is highly associated with bloodstream infection (BSI), which may increase mortality. To explore the risk factors for BSI in paediatric HSCT recipients. Three English databases and four Chinese databases were searched from inception to March 17th, 2022. Eligible studies included randomized controlled trials, cohort studies, and case–control studies that enrolled HSCT recipients aged ≤18 years and reported BSI risk factors. Two reviewers independently screened studies, extracted data, and assessed the risk of bias. Using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE), certainty of body of evidence was assessed. Fourteen studies involving 4602 persons were included. The incidences of BSI and associated mortality in paediatric HSCT recipients were approximately 10–50% and 5–15%, respectively. Meta-analysis of all studies revealed that previous BSI before HSCT (relative effect (RE): 2.28; 95% confidence interval (CI) 1.19–4.34, moderate certainty) and receiving an umbilical cord blood transplant (RE: 1.55; 95% CI: 1.22–1.97, moderate certainty) were probably associated with an increased risk of BSI. Meta-analysis of studies with low risk of bias reassured that previous BSI before HSCT probably increased the risk of BSI (RE: 2.28; 95% CI: 1.19–4.34, moderate certainty), and revealed that steroid use (RE: 2.72; 95% CI: 1.31–5.64, moderate certainty) was likely a risk factor whereas autologous HSCT was probably a protective factor of BSI (RE: 0.65; 95% CI: 0.45–0.94, moderate certainty). These findings could inform the management of paediatric HSCT recipients, helping identify who may benefit from prophylactic antibiotics. [ABSTRACT FROM AUTHOR]

To assess the feasibility of utilizing electron density (ED) maps from a novel spectral CT simulator for dose calculation in the treatment plan. 10 patients from pelvic treatment sites were selected retrospectively, and all of them underwent spectral CT simulation. Treatment plans based on conventional CT images were transplanted to ED maps with the same structure set of planning target volume (PTV) and organ at risk (OARs), and the dose was then recalculated. The differences in dosimetric metrics of the PTV and OARs of the two types of plans were analyzed and compared. Besides, gamma analysis between these plans was performed by using MEPHYSTO Navigator software. In terms of PTV dosimetric metrics, the homogeneity index (HI), conformal index (CI), gradient index (GI), D 2% , D 98% , and D mean showed no significant difference between conventional plans and ED plans (0.921 vs. 0.929, 3.73 vs. 3.71, 0.06 vs. 0.06, 47.46 vs. 47.50 Gy, 44.53 vs. 44.53Gy, 46.33 vs. 46.34Gy, P >0.05). For OARs, a significant difference was observed in rectal D 50% between these two treatment plans (40.45 ± 3.65 vs. 40.55 ± 3.72Gy, P = 0.014), although the variance remained minor. For the rest, the dosimetric metrics exhibited no significant difference between conventional plans and ED plans. The mean gamma passing rate (GPR) of gamma analysis was 96.41%, 99.58%, and 99.66% under 1%/1mm, 2%/2mm, and 3%/2mm criteria, respectively. Compared with conventional treatment plans based on CT images, the plans based on ED maps had similar dosimetric quality. However, the latter can be directly used in dose calculation without establishing and selecting a specific HU-ED calibration curve, providing an advantage in clinical applications. [ABSTRACT FROM AUTHOR]

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.jenvman.2006.08.018 Byline: S. Tang (a)(b), J.M. Chen (c), Q. Zhu (b), X. Li (b), M. Chen (c), R. Sun (b), Y. Zhou (b), F. Deng (c), D. Xie (b) Keywords: LAI; MODIS; BRDF; Albedo; Gap fraction Abstract: Leaf area index (LAI) is an important ecological and environmental parameter. A new LAI algorithm is developed using the principles of ground LAI measurements based on canopy gap fraction. First, the relationship between LAI and gap fraction at various zenith angles is derived from the definition of LAI. Then, the directional gap fraction is acquired from a remote sensing bidirectional reflectance distribution function (BRDF) product. This acquisition is obtained by using a kernel driven model and a large-scale directional gap fraction algorithm. The algorithm has been applied to estimate a LAI distribution in China in mid-July 2002. The ground data acquired from two field experiments in Changbai Mountain and Qilian Mountain were used to validate the algorithm. To resolve the scale discrepancy between high resolution ground observations and low resolution remote sensing data, two TM images with a resolution approaching the size of ground plots were used to relate the coarse resolution LAI map to ground measurements. First, an empirical relationship between the measured LAI and a vegetation index was established. Next, a high resolution LAI map was generated using the relationship. The LAI value of a low resolution pixel was calculated from the area-weighted sum of high resolution LAIs composing the low resolution pixel. The results of this comparison showed that the inversion algorithm has an accuracy of 82%. Factors that may influence the accuracy are also discussed in this paper. Author Affiliation: (a) Key Laboratory of Radiometric Calibration and Validation for Environmental Satellites National Satellite Meteorological Center, China Meteorological Administration, Beijing 100081, China (b) Remote Sensing and GIS Research Center, Department of Geography, Beijing Normal University, Beijing 100875, China (c) Department of Geography and Program in Planning, University of Toronto, 100 St. George St., Room 5047,Toronto, ON, Canada M5S 3G3 Article History: Received 15 April 2004; Revised 6 December 2005; Accepted 23 August 2006

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.jenvman.2006.08.019 Byline: J.M. Chen (a), S.C. Thomas (b), Y. Yin (c), V. Maclaren (a), J. Liu (d), J. Pan (e), G. Liu (d), Q. Tian (f), Q. Zhu (g), J.-J. Pan (h), X. Shi (i), J. Xue (j), E. Kang (k) Keywords: Carbon Sequestration; China's forests; Integrated assessment; Remote sensing Abstract: This article serves as an introduction to this special issue, 'China's Forest Carbon Sequestration', representing major results of a project sponsored by the Canadian International Development Agency and the Chinese Academy of Sciences. China occupies a pivotal position globally as a principle emitter of carbon dioxide, as host to some of the world's largest reforestation efforts, and as a key player in international negotiations aimed at reducing global greenhouse gas emission. The goals of this project are to develop remote sensing approaches for quantifying forest carbon balance in China in a transparent manner, and information and tools to support land-use decisions for enhanced carbon sequestration (CS) that are science based and economically and socially viable. The project consists of three components: (i) remote sensing and carbon modeling, (ii) forest and soil assessment, and (iii) integrated assessment of the socio-economic implications of CS via forest management. Articles included in this special issue are highlights of the results of each of these components. Author Affiliation: (a) Department of Geography, University of Toronto, 100 St. George Street, Room 5047, Toronto, ON, Canada M5S 3G3 (b) Faculty of Forestry, University of Toronto, 33 Willcocks Street, Toronto, ON, Canada M5S 3B3 (c) Adaptation and Impacts Research Division, Environment Canada (d) Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, PR China (e) Center for Environment, University of Toronto, Toronto, ON, Canada M5S 3B3 (f) International Earth System Science Institute, Nanjing University, PR China (g) College of Geography and Remote Sensing, Beijing Normal University, PR China (h) Nanjing Agricultural University, PR China (i) Nanjing Institute of Soil Science, Chinese Academy of Sciences, PR China (j) Nanjing Forestry University, PR China (k) Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou, PR China Article History: Received 10 April 2006; Revised 23 April 2006; Accepted 9 August 2006

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.jenvman.2006.08.017 Byline: P. Wang (a)(b)(c), R. Sun (a)(b), J. Hu (a)(b), Q. Zhu (a)(b), Y. Zhou (e), L. Li (d), J.M. Chen (f) Keywords: LAI; NPP; Forest; BEPS; Carbon cycle Abstract: Large scale process-based modeling is a useful approach to estimate distributions of global net primary productivity (NPP). In this paper, in order to validate an existing NPP model with observed data at site level, field experiments were conducted at three sites in northern China. One site is located in Qilian Mountain in Gansu Province, and the other two sites are in Changbaishan Natural Reserve and Dunhua County in Jilin Province. Detailed field experiments are discussed and field data are used to validate the simulated NPP. Remotely sensed images including Landsat Enhanced Thematic Mapper plus (ETM+, 30m spatial resolution in visible and near infrared bands) and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER, 15m spatial resolution in visible and near infrared bands) are used to derive maps of land cover, leaf area index, and biomass. Based on these maps, field measured data, soil texture and daily meteorological data, NPP of these sites are simulated for year 2001 with the boreal ecosystem productivity simulator (BEPS). The NPP in these sites ranges from 80 to 800gCm.sup.-2 a.sup.-1. The observed NPP agrees well with the modeled NPP. This study suggests that BEPS can be used to estimate NPP in northern China if remotely sensed images of high spatial resolution are available. Author Affiliation: (a) State Key Laboratory of Remote Sensing Science, Jointly Sponsored by Beijing Normal University and the Institute of Remote Sensing Applications of Chinese Academy of Sciences, Beijing, China (b) Beijing Key Laboratory of Enviromental Remote Sensing and City Digitalization, School of Geography, Beijing Normal University, Beijing, China (c) Chinese Academy of Meteorological Sciences, Beijing 100081, China (d) Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing 100101, China (e) Department of Natural Resources Science, University of Rhode Island, USA (f) Department of Geography and Program in Panning, University of Toronto, 100 St. George Street, Room 5047,Toronto, Ont., Canada M5S 3G3 Article History: Received 29 April 2005; Revised 30 November 2005; Accepted 9 August 2006

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.jenvman.2006.04.024 Byline: Y. Zhou (a)(b), Q. Zhu (a), J.M. Chen (c), Y.Q. Wang (b), J. Liu (d), R. Sun (a), S. Tang (a) Keywords: BEPS; Biomass; Leaf area index; Net primary productivity; Western China Abstract: We modeled net primary productivity (NPP) at high spatial resolution using an advanced spaceborne thermal emission and reflection radiometer (ASTER) image of a Qilian Mountain study area using the boreal ecosystem productivity simulator (BEPS). Two key driving variables of the model, leaf area index (LAI) and land cover type, were derived from ASTER and moderate resolution imaging spectroradiometer (MODIS) data. Other spatially explicit inputs included daily meteorological data (radiation, precipitation, temperature, humidity), available soil water holding capacity (AWC), and forest biomass. NPP was estimated for coniferous forests and other land cover types in the study area. The result showed that NPP of coniferous forests in the study area was about 4.4tCha.sup.-1 y.sup.-1. The correlation coefficient between the modeled NPP and ground measurements was 0.84, with a mean relative error of about 13.9%. Author Affiliation: (a) College of Geography and Remote Sensing, Beijing Normal University, China (b) Department of Natural Resources Science, University of Rhode Island, Kingston, RI 02881, USA (c) Department of Geography and Program in Planning, University of Toronto, 100 St. George Street, Room 5047, Toronto, Ont., Canada M5S 3G3 (d) Department of Physics, University of Toronto, Canada Article History: Received 4 April 2005; Revised 12 December 2005; Accepted 3 April 2006

• Composites containing N-doped lotus seedpods biocarbon (NALC) and NiCo 2 S 4 (NALC@NCS) was developed. • Excellent electrocatalytic activity of the composites for reduction from I 3 − to I − was achieved. • Dye-sensitized solar cells assembled with the NALC@NCS counter electrodes achieved impressive photovoltaic conversion efficiency of 8.58%. A facile and green approach to composites containing nitrogen doped lotus seedpods biocarbon (NALC) and NiCo 2 S 4 (NCS) is reported. The composites present excellent electrocatalytic activity for reduction from I 3 − to I − due to the synergistic effect between the nanocrystalline NCS and the NALC. Dye-sensitized solar cells (DSSCs) assembled with the as-prepared NALC@NCS counter electrodes (CEs) achieve impressive photovoltaic conversion efficiency (PCE) of 8.58%, which is higher than that by the same structured cell with Pt (7.85%) and NCS (7.53%) CEs. Compared with other NCS based CEs, the NALC@NCS CEs show good durability and stability, suggesting that combining nanocrystalline catalyst with biocarbon is beneficial to stimulate the synergistic effect, and is a feasible strategy to optimal CEs for high performance DSSCs. Low cost NALC@NCS composites developed in this work are environmentally friendly, stable and efficient, which are believed to be good candidates for Pt-free CE materials in DSSCs and electrocatalysis applications. [Display omitted] [ABSTRACT FROM AUTHOR]

Aim: To investigate whether whole-lesion histogram analysis of apparent diffusion coefficient (ADC) values derived from mono-exponential and bi-exponential diffusion-weighted imaging (DWI) can differentiate lung cancer from benign pulmonary lesions.Materials and Methods: Thirty-two patients with lung cancer and 17 patients with benign pulmonary lesions were included retrospectively. All patients underwent DWI before surgery or biopsy. ADC histogram parameters, including mean, percentile values (10th and 90th), kurtosis, and skewness, were calculated independently by two radiologists. The histogram parameters were compared between patients with lung cancer and benign lesions. Receiver operating characteristic curves were constructed to evaluate the diagnostic performance.Results: The ADCMean, ADC10th, DMean, D10th were significantly lower in lung cancer (1.187 ± 0.144 × 10-3; 0.440 ± 0.062 × 10-3; 1.068 ± 0.108 × 10-3; and 0.422 ± 0.049 × 10-3 mm/s) compared to benign lesions (1.418 ± 0.274 × 10-3; 0.555 ± 0.113 × 10-3; 1.216 ± 0.149 × 10-3; and 0.490 ± 0.044 × 10-3 mm/s; p<0.05). The ADCSkewness and DSkewness were significantly different between lung cancer (2.35 ± 0.72; 2.58 ± 1.14) and benign lesions (1.85 ± 0.54; 1.59 ± 1.47; p<0.05). D10th was robust in differentiating lung cancer from benign lesions. Using 0.453 × 10-3 mm/s as the optimal threshold, the sensitivity, specificity, and accuracy of D10th were 78.12%, 82.35%, and 79.6%, respectively.Conclusion: Whole-lesion histogram analysis of ADC values derived by mono-exponential and bi-exponential DWI using 3 T magnetic resonance imaging helps distinguish lung cancer from benign pulmonary lesions. [ABSTRACT FROM AUTHOR]