| 1. |
- Yu, Wenjin, et al.
(författare)
-
Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid
- 2022
-
Ingår i: Frontiers in Oncology. - : Frontiers Media SA. - 2234-943X. ; 12, s. 1-11
-
Tidskriftsartikel (refereegranskat)abstract
- Background: It is a critical challenge to diagnose leptomeningeal metastasis (LM), given its technical difficulty and the lack of typical symptoms. The existing gold standard of diagnosing LM is to use positive cerebrospinal fluid (CSF) cytology, which consumes significantly more time to classify cells under a microscope.Objective: This study aims to establish a deep learning model to classify cancer cells in CSF, thus facilitating doctors to achieve an accurate and fast diagnosis of LM in an early stage.Method: The cerebrospinal fluid laboratory of Xijing Hospital provides 53,255 cells from 90 LM patients in the research. We used two deep convolutional neural networks (CNN) models to classify cells in the CSF. A five-way cell classification model (CNN1) consists of lymphocytes, monocytes, neutrophils, erythrocytes, and cancer cells. A four-way cancer cell classification model (CNN2) consists of lung cancer cells, gastric cancer cells, breast cancer cells, and pancreatic cancer cells. Here, the CNN models were constructed by Resnet-inception-V2. We evaluated the performance of the proposed models on two external datasets and compared them with the results from 42 doctors of various levels of experience in the human-machine tests. Furthermore, we develop a computer-aided diagnosis (CAD) software to generate cytology diagnosis reports in the research rapidly.Results: With respect to the validation set, the mean average precision (mAP) of CNN1 is over 95% and that of CNN2 is close to 80%. Hence, the proposed deep learning model effectively classifies cells in CSF to facilitate the screening of cancer cells. In the human-machine tests, the accuracy of CNN1 is similar to the results from experts, with higher accuracy than doctors in other levels. Moreover, the overall accuracy of CNN2 is 10% higher than that of experts, with a time consumption of only one-third of that consumed by an expert. Using the CAD software saves 90% working time of cytologists.Conclusion: A deep learning method has been developed to assist the LM diagnosis with high accuracy and low time consumption effectively. Thanks to labeled data and step-by-step training, our proposed method can successfully classify cancer cells in the CSF to assist LM diagnosis early. In addition, this unique research can predict cancer’s primary source of LM, which relies on cytomorphologic features without immunohistochemistry. Our results show that deep learning can be widely used in medical images to classify cerebrospinal fluid cells. For complex cancer classification tasks, the accuracy of the proposed method is significantly higher than that of specialist doctors, and its performance is better than that of junior doctors and interns. The application of CNNs and CAD software may ultimately aid in expediting the diagnosis and overcoming the shortage of experienced cytologists, thereby facilitating earlier treatment and improving the prognosis of LM.
|
|
| 2. |
- Liu, Yu, et al.
(författare)
-
A Review of Physical and Numerical Approaches for the Study of Gas Stirring in Ladle Metallurgy
- 2019
-
Ingår i: Metallurgical and materials transactions. B, process metallurgy and materials processing science. - : SPRINGER. - 1073-5615 .- 1543-1916. ; 50:1, s. 555-577
-
Forskningsöversikt (refereegranskat)abstract
- This article presents a review of the research into gas stirring in ladle metallurgy carried out over the past few decades. Herein, the physical modeling experiments are divided into four major areas: (1) mixing and homogenization in the ladle; (2) gas bubble formation, transformation, and interactions in the plume zone; (3) inclusion behavior at the steel-slag interface and in the molten steel; and (4) open eye formation. Several industrial trials have also been carried out to optimize gas stirring and open eye formation. Approaches for selecting criteria for scaling to guarantee flow similarity between industrial trials and physical modeling experiments are discussed. To describe the bubble behavior and two-phase plume structure, four main mathematical models have been used in different research fields: (1) the quasi-single-phase model, (2) the volume of fluid (VOF) model, (3) the Eulerian multiphase (E-E) model, and (4) the Eulerian-Lagrangian (E-L) model. In recent years, the E-E model has been used to predict gas stirring conditions in the ladle, and specific models in commercial packages, as well as research codes, have been developed gradually to describe the complex physical and chemical phenomena. Furthermore, the coupling of turbulence models with multiphase models is also discussed. For physical modeling, some general empirical rules have not been analyzed sufficiently. Based on a comparison with the available experimental results, it is found that the mathematical models focusing on the mass transfer phenomenon and inclusion behaviors at the steel-slag interface, vacuum degassing at the gas-liquid interface, dissolution rate of the solid alloy at the liquid-solid interface, and the combination of fluid dynamics and thermodynamics need to be improved further. To describe industrial conditions using mathematical methods and improve numerical modeling, the results of physical modeling experiments and industrial trials must offer satisfactory validations for the improvement of numerical modeling.
|
|
| 3. |
- Liu, Yu, et al.
(författare)
-
An Experimental and Mathematical Work on Single Bubble Behavior under Reduced Pressure
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- During secondary steelmaking, argon bubbles are often passed through molten steel to ensure a clean and homogeneous product. The behavior of the bubbles and the capacity of the bubbles to stir the melt and remove impurities depends on their size, shape, and velocity. These factors depend on the ambient pressure of the melt, the temperature and flow rate of the gas and the geometry of the gas nozzles. There have been many studies that investigate the behavior of bubbles when the melt is under atmospheric pressure, but few when the melt is held under vacuum. This makes it difficult to optimize the argon blowing process. The current study addresses this lack of knowledge by studying bubble behavior when the melt is under vacuum. Physical modeling was used to analyze the effects of the reduced pressure and nozzle diameter on the bubbles’ initial diameter and ascent behavior in a molten steel. Moreover, a multiphase fluid dynamics solver for compressible fluids called ‘compressibleInterFoam’ was validated and used. Increasing the flow rate leads to larger initial bubble diameters and more frequent bubble formation, and increasing the nozzle diameter leads to larger initial bubble diameters and less frequent bubble formation. Decreasing the subjected pressure causes the bubble diameter to increase substantially but bubbles to form less frequently. For flow rates in the range of 5.0(mL·min-1)STP to 2000(mL·min-1)STP, the bubble diameter ranges from 6.0mm to 20.0mm. The frequency of bubble generation initially increases with flow rate before reaching a constant value. During the ascent, a bubble will shed several small bubbles at the bottom to reach a constant shape. In the steel-argon system, under laminar flow conditions, the maximum bubble width under a pressure of 0.2bar is 65mm and is 58mm under a pressure of 2.0bar. As the surrounding pressure increases, the maximum size of the bubble under the steady condition will decrease. These findings can be used to determine the bubble behaviors and to optimize the conditions of argon blowing to produce steel that is sufficiently clean, while minimizing argon usage.
|
|
| 4. |
- Liu, Yu, et al.
(författare)
-
Comparison of Euler-Euler Approach and Euler–Lagrange Approach to Model Gas Injection in a Ladle
- 2019
-
Ingår i: Steel Research International. - : Wiley-VCH Verlagsgesellschaft. - 1611-3683 .- 1869-344X. ; 90:5
-
Tidskriftsartikel (refereegranskat)abstract
- The gas injection in a ladle using a porous plug is simulated using both the Euler-Euler and Euler-Lagrange approaches. The effects of various forces, bubble sizes, and bubble injection frequencies on the flow pattern are modeled. For predicting axial velocity and turbulent kinetic energy, the Euler-Lagrange approach fits better than Euler-Euler approach with the measured data. In the Euler-Euler approach, differences in axial velocities and turbulent kinetic energies for various bubble sizes mainly appears in the plume zone. In the Euler-Lagrange approach, different bubble sizes with the same injection frequency have a small impact on the turbulence dissipation. Furthermore, the turbulent dispersion from the gas phase to the liquid phase has an important effect on the plume structure and spout eye formation. For both modeling, the smaller the bubble diameter is, the larger the axial velocity and turbulent kinetic dissipation are in the central zone. For the bubble coalescence and breakup, according to the comparison of two modeling approaches, the Euler-Lagrange approach is more accurate in predicting the flow pattern for gas injection with a porous plug in the ladle.
|
|
| 5. |
- Liu, Yu, et al.
(författare)
-
Influence of Plug Design on Bubble Injection Characteristics in Ladle Metallurgy
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- The bubble characteristic created during bottom injection in a ladle has been studied using physical modeling and mathematical modeling. The width of central plume, statistical number of bubbles, and the periodic behavior of bubble injection have been compared using a slot plug and three porous plugs with various porosities. Furthermore, the effect of the plug’s permeability on the plume structure and bubble size distribution has been predicted using the Eulerian Multiphase Model and Population Balance Model. The results show that, for the bottom blowing using a slot plug, an increased flow rate will not change the gas-liquid plume width significantly, but lead to an increased periodic behavior of bubble’s statistic amount. For the bottom blowing using porous plug, an increased flow rate results in a wider central plume, but no obvious change of the periodic characteristic of bubbles. The effect of porosity on the bubble behavior was also studied. As the porosity increases, the volume fraction of gas and the average size of bubbles at the central plume zone increases, and the width of two-phase plume decreases. Moreover, the results predicted by the mathematical modeling are consistent with those from the physical modeling. The difference between a slot plug and a porous plug in industrial applications has been analyzed theoretically. The results show that it is more beneficial to use the slot plug to create strong stirring and to reach beneficial desulfurization condition nearby the open eye, and to use the porous plug for the inclusion removal.
|
|
| 6. |
- Liu, Yu, et al.
(författare)
-
Physical and Numerical Modelling on the Mixing Condition in a 50 t Ladle
- 2019
-
Ingår i: Metals. - : MDPI. - 2075-4701. ; 9:11
-
Tidskriftsartikel (refereegranskat)abstract
- The bubbly flow and mixing conditions for gas stirring in a 50t ladle were investigated by using physical modelling and mathematical modelling. In the physical modelling, the effect of the porous plugs' configurations on the tracer homogenization was studied by using a saturated NaCl solution to predict the mixing time and a color dye to show the mixing pattern. In the mathematical modelling, the Euler-Lagrange model and species transport model were used to predict the flow pattern and tracer homogenization, respectively. The results show that, for a +/- 5% homogenization degree, the mixing time with dual plugs using a radial angle of 180 degrees is shortest. In addition, the mixing time using a radial angle of 135 degrees decreases the most with an increased flow rate. The flow pattern and mixing conditions predicted by mathematical modelling agree well with the result of the physical modelling. For a +/- 1% homogenization degree, the influence of the tracer's natural convection on its homogenization pattern cannot be neglected. This is especially true for a 'soft bubbling' case using a low gas flow rate. Overall, it is recommended that large radial angles in the range of 135 degrees 180 degrees are chosen for gas stirring in the present study when using dual porous plugs.
|
|
| 7. |
- Yuan, Wenping, et al.
(författare)
-
Redefinition and global estimation of basal ecosystem respiration rate
- 2011
-
Ingår i: Global Biogeochemical Cycles. - 0886-6236. ; 25
-
Tidskriftsartikel (refereegranskat)abstract
- Basal ecosystem respiration rate (BR), the ecosystem respiration rate at a given temperature, is a common and important parameter in empirical models for quantifying ecosystem respiration (ER) globally. Numerous studies have indicated that BR varies in space. However, many empirical ER models still use a global constant BR largely due to the lack of a functional description for BR. In this study, we redefined BR to be ecosystem respiration rate at the mean annual temperature. To test the validity of this concept, we conducted a synthesis analysis using 276 site-years of eddy covariance data, from 79 research sites located at latitudes ranging from similar to 3 degrees S to similar to 70 degrees N. Results showed that mean annual ER rate closely matches ER rate at mean annual temperature. Incorporation of site-specific BR into global ER model substantially improved simulated ER compared to an invariant BR at all sites. These results confirm that ER at the mean annual temperature can be considered as BR in empirical models. A strong correlation was found between the mean annual ER and mean annual gross primary production (GPP). Consequently, GPP, which is typically more accurately modeled, can be used to estimate BR. A light use efficiency GPP model (i.e., EC-LUE) was applied to estimate global GPP, BR and ER with input data from MERRA (Modern Era Retrospective-Analysis for Research and Applications) and MODIS (Moderate resolution Imaging Spectroradiometer). The global ER was 103 Pg C yr (-1), with the highest respiration rate over tropical forests and the lowest value in dry and high-latitude areas.
|
|
| 8. |
- Golub, Malgorzata, et al.
(författare)
-
Diel, seasonal, and inter-annual variation in carbon dioxide effluxes from lakes and reservoirs
- 2023
-
Ingår i: Environmental Research Letters. - : IOP Publishing. - 1748-9326. ; 18:3
-
Tidskriftsartikel (refereegranskat)abstract
- Accounting for temporal changes in carbon dioxide (CO2) effluxes from freshwaters remains a challenge for global and regional carbon budgets. Here, we synthesize 171 site-months of flux measurements of CO2 based on the eddy covariance method from 13 lakes and reservoirs in the Northern Hemisphere, and quantify dynamics at multiple temporal scales. We found pronounced sub-annual variability in CO2 flux at all sites. By accounting for diel variation, only 11% of site-months were net daily sinks of CO2. Annual CO2 emissions had an average of 25% (range 3%-58%) interannual variation. Similar to studies on streams, nighttime emissions regularly exceeded daytime emissions. Biophysical regulations of CO2 flux variability were delineated through mutual information analysis. Sample analysis of CO2 fluxes indicate the importance of continuous measurements. Better characterization of short- and long-term variability is necessary to understand and improve detection of temporal changes of CO2 fluxes in response to natural and anthropogenic drivers. Our results indicate that existing global lake carbon budgets relying primarily on daytime measurements yield underestimates of net emissions.
|
|
| 9. |
- Liu, Heping, et al.
(författare)
-
Realization of quaternionic discrete series on the unit ball in H^d
- 2012
-
Ingår i: Journal of Functional Analysis. - : Elsevier BV. - 0022-1236 .- 1096-0783. ; 262:7, s. 2979-3005
-
Tidskriftsartikel (refereegranskat)abstract
- We give a realization for the quaternionic discrete series for the group Sp(1, d) as the kernel of certainelementary Cauchy–Riemann–Fueter operator, and we compute the corresponding reproducing kernels.
|
|
| 10. |
- Yi, Chuixiang, et al.
(författare)
-
Climate control of terrestrial carbon exchange across biomes and continents
- 2010
-
Ingår i: Environmental Research Letters. - : IOP Publishing. - 1748-9326. ; 5:3
-
Tidskriftsartikel (refereegranskat)abstract
- Understanding the relationships between climate and carbon exchange by terrestrial ecosystems is critical to predict future levels of atmospheric carbon dioxide because of the potential accelerating effects of positive climate-carbon cycle feedbacks. However, directly observed relationships between climate and terrestrial CO2 exchange with the atmosphere across biomes and continents are lacking. Here we present data describing the relationships between net ecosystem exchange of carbon (NEE) and climate factors as measured using the eddy covariance method at 125 unique sites in various ecosystems over six continents with a total of 559 site-years. We find that NEE observed at eddy covariance sites is (1) a strong function of mean annual temperature at mid-and high-latitudes, (2) a strong function of dryness at mid-and low-latitudes, and (3) a function of both temperature and dryness around the mid-latitudinal belt (45 degrees N). The sensitivity of NEE to mean annual temperature breaks down at similar to 16 degrees C (a threshold value of mean annual temperature), above which no further increase of CO2 uptake with temperature was observed and dryness influence overrules temperature influence.
|
|
