Machine learning applications in healthcare clinical practice and research

Machine learning(ML)is a type of artificial intelligence that assists computers in the acquisition of knowledge through data analysis,thus creating machines that can complete tasks otherwise requiring human intelligence.Among its various applications,it has proven groundbreaking in healthcare as well,both in clinical practice and research.In this editorial,we succinctly introduce ML applications and present a study,featured in the latest issue of the World Journal of Clinical Cases.The authors of this study conducted an analysis using both multiple linear regression(MLR)and ML methods to investigate the significant factors that may impact the estimated glomerular filtration rate in healthy women with and without non-alcoholic fatty liver disease(NAFLD).Their results implicated age as the most important determining factor in both groups,followed by lactic dehydrogenase,uric acid,forced expiratory volume in one second,and albumin.In addition,for the NAFLD-group,the 5th and 6th most important impact factors were thyroid-stimulating hormone and systolic blood pressure,as compared to plasma calcium and body fat for the NAFLD+group.However,the study's distinctive contribution lies in its adoption of ML methodologies,showcasing their superiority over traditional statistical approaches(herein MLR),thereby highlighting the potential of ML to represent an invaluable advanced adjunct tool in clinical practice and research.

A review on machinability of carbon fiber reinforced polymer(CFRP)and glass fiber reinforced polymer(GFRP)composite materials

Fiber reinforced polymer(FRP) composite materials are heterogeneous and anisotropic materials that do not exhibit plastic deformation. They have been used in a wide range of contemporary applications particularly in space and aviation,automotive,maritime and manufacturing of sports equipment. Carbon fiber reinforced polymer(CFRP) and glass fiber reinforced polymer(GFRP) composite materials,among other fiber reinforced materials,have been increasingly replacing conventional materials with their excellent strength and low specific weight properties. Their manufacturability in varying combinations with customized strength properties,also their high fatigue,toughness and high temperature wear and oxidation resistance capabilities render these materials an excellent choice in engineering applications.In the present review study,a literature survey was conducted on the machinability properties and related approaches for CFRP and GFRP composite materials. As in the machining of all anisotropic and heterogeneous materials,failure mechanisms were also reported in the machining of CFRP and GFRP materials with both conventional and modern manufacturing methods and the results of these studies were obtained by use of variance analysis(ANOVA),artificial neural networks(ANN) model,fuzzy inference system(FIS),harmony search(HS) algorithm,genetic algorithm(GA),Taguchi's optimization technique,multi-criteria optimization,analytical modeling,stress analysis,finite elements method(FEM),data analysis,and linear regression technique. Failure mechanisms and surface quality is discussed with the help of optical and scanning electron microscopy,and profilometry. ANOVA,GA,FEM,etc. are used to analyze and generate predictive models.

Precipitation and control of BN inclusions in 42CrMo steel and their efect on machinability

The precipitation and control of boron nitrogen （BN） inclusions in 42CrMo steel were investigated and their effect on machinability was analyzed. First, the precipitation regularity of BN in 42CrMo steel was studied by theoretical calculation. Then, the machinability of the steel was investigated through contrast cutting experiments, and the composition and cooling rate of the steel were controlled to analyze the variation laws of the size, distribution, and area ratio of BN inclusions. Finally, the results were combined with the machinability of the steel to analyze the relationship among them. It is found that the machinability of the steel is mainly influenced by the diameter and quantity of BN inclusions. Fine and dispersedly distributed BN inclusions are more beneficial for the improvement in machinability of 42CrMo steel than coarse and sparse BN inclusions.

Experiment on Machinability of Ti2AlNb Intermetallic Alloy

Ti2AlNb intermetallic alloy is a newly developed high-temperature resistant structural material due to its excellent material and mechanical properties,which also make it to be one of the most difficult-to-cut materials.In order to study the machinability of Ti2AlNb alloy,a series of turning experiments of Ti2AlNb alloy with varying cutting speed and feed rate using coated carbide tools are carried out.The results associated with cutting forces,cutting temperature and tool wear are presented and discussed.Moreover,the cutting performance of Ti2AlNb alloy is evaluated in comparison with that of most commonly used Ti6Al4 Vand Inconel 718 alloys in terms of the cutting forces and cutting temperature.The comparison results show that there is a correlation between the machinability and the mechanical properties of work material properties.Additionally,considering material removal rate and tool life,the optimized machining parameters for cutting Ti2AlNb alloys using coated carbide tools are recommended.

Effect of inoculating addition on machinability of gray cast iron

Gray cast irons were inoculated with FeSi75＋RE and FeSi75＋Sr inoculants. The changes of apex angle of the drills before and after being used were used to evaluate machinability of gray cast irons. Effect of FeSi75＋RE and FeSi75＋Sr inoculants on mechanical properties, machinability and sensibility of gray cast iron used in cylinder block were investigated. Experimental results showed that gray cast iron treated with 60%FeSi75＋40% RE inoculants exhibited tensile strength consistently at about 295 MPa along with good hardness and improved metallurgical quality. While gray cast iron inoculated with 20%FeSi75＋80% Sr inoculants exhibited the best machinability, the lowest cross-section sensibility and the least microhardness difference. The tool flank wear of the drill increased correspondingly with the increase of the microhardness difference of the matrix, indicating the great effect of homogeneity of the matrix on the machinability of gray cast iron.

Influence of heat treatment on the machinability and corrosion behavior of AZ91 Mg alloy

In the present study,AZ91 Mg alloy was heat treated at 410℃ for 6,12 and 24 h to investigate the influence of heat treatment on machinability and corrosion behavior.The effect of soaking time on the amount and distribution of Mg 17 Al 12(β-phase)was analyzed under the optical microscope.Microhardness measurements demonstrated the increased hardness with increased heat treatment soaking time,which can be attributed to the solid solution strengthening.The influence of super saturatedα-grains on reducing the cutting force(F z)with respect to increased cutting speed was observed as prominent.The corrosion behavior of the heat treated specimens was studied by conducting electrochemical tests.Surprisingly,corrosion rate of heat treated samples was observed as increased compared with the base material.From the results,it is evident that the machinability of AZ91 Mg alloy can be improved by producing super saturatedα-grains through heat treatment but at the cost of losing corrosion resistance.

Influence of Al content on machinability of AZ series Mg alloys

Influence of A1 content on the machinability of AZ series cast Mg alloys was investigated. In order to evaluate the machinability of the alloys, measurements of the cutting forces during turning operations and surface roughness were carried out as well as considering the microstructure and tensile properties. The results show that maximum tensile properties are observed with 2% （mass fraction） A1 addition to Mg. As the A1 content of the alloy increases above 2%, the cutting forces tend to reduce along with the ductility owing to the grain boundary precipitation of intermetallic phase （fl-Mgl7All2）. Cutting forces are able to increase as the cutting speed increases for all the alloys studied, and it＇s attributed to flank built up at the tip of the cutting tool during machining.

Superior machinability of steel enhanced with BN and MnS particles

The strategy that replacing part of MnS with BN was proposed in order to decrease the sulfur content in sulfur based free-cutting steel. The effects of BN and MnS inclusions on the microstructure and machinability of the steel were systematically investigated. The results show that most of the BN and MnS inclusions exist individually in the steel and only a small amount of them are in a composite state form- ing either isolated particles or clusters of particles. In the case of multi-phased steel, the theoretical calculation predicts that the volume of large BN particles should be 0.7 times of the volume of large MnS particles. The machinability of this type of BN and MnS alloy steel over a wide range of cutting speeds ranging from a low speed appropriate for drilling to a high speed appropriate for turning is confirmed as being equal to or superior to that of an MnS reference steel, even though the sulfur content in the composite steel is only half that of the MnS steel. The aptitude for cutting effect of 240 ppm nitrogen and 115 ppm boron in the composite steel is demonstrated to be equivalent or even better than 1000 ppm sulfur in MnS free-cutting steel.

Experimental Study of Machinability in Millgrinding of SiCp/Al Composites

An attempt was made to investigate the machinability of Si Cp/Al composites based on the experimental study using mill-grinding processing method. The experiments were carried out on a high-speed CNC machining center using integrated abrasive cutting tool. The effects of combined machining parameters, e g, cutting speed（vs）, feed rate（vf）, and depth of cut（ap）, with the same change of material removal rate（MRR） on the mill-grinding force and surface roughness（Ra） were investigated. The formation mechanism of typical machined surface defects was analyzed by SEM. The experimental results reveal that with the same change of material removal rate, lower mill-grinding force values can be gained by increasing depth of cut and feed rate simultaneously at higher cutting speed. With the same change of MRR value, lower surface roughness values can be gained by increasing the feed rate at higher cutting speed, rather than just increasing the depth of cut, or increasing the feed rate and depth of cut simultaneously. The machined surface of Si Cp/Al composites reveals typical defects which can influence surface integrity.

Machinability of Hastelloy C-276 Using Hot-pressed Sintered Ti(C_7N_3)-based Cermet Cutting Tools

C-276 nickel-based alloy is a difficult-to-cut material. In high-speed machining of Hastelloy C-276, notching is a prominent failure mode due to high mechanical properties of work piece, which results in the short tool life and low productivity. In this paper, a newly developed Ti（C7N3）-based cermet insert manufactured by a hot-pressing method is used to machine the C-276 nickel-based alloy, and its cutting performances are studied. Based on orthogonal experiment method, the influence of cutting parameters on tool life, material removal rates and surface roughness are investigated. Experimental research results indicate that the optimal cutting condition is a cutting speed of 50 m/min, depth of cut of 0.4 mm and feed rate of 0.15 mm/r if the tool life and material removal rates are considered comprehensively. In this case, the tool life is 32 min and material removal rates are 3000 mm^3/min, which is appropriate to the rough machining. If the tool life and surface roughness are considered, the better cutting condition is a cutting speed of 75 m/min, depth of cut of 0.6 mm and feed rate of 0.1 mm/r. In this case, the surface roughness is 0.59μm. Notch wear, flank wear, chipping at the tool nose, built-up edge（BUE） and micro-cracks are found when Ti（C7N3）-based cermet insert turned Hastelloy C-276. Oxidation, adhesive, abrasive and diffusion are the wear mechanisms, which can be investigated by the observations of scanning electron microscope and energy-dispersive spectroscopy. This research will help to guide studies on the evaluation of machining parameters to further advance the productivity of nickel based alloy Hastelloy C-276 machining.

Forgeability and Machinability of Stir Cast Aluminum Alloy Metal Matrix Composites

In the present paper, the aluminum alloy i.e. LM6 based composites reinforced with different weight fraction of SiC particles was produced by stir cast technique and the effect of reinforced ratios on the forgeability and the machinability was examined. The test results show that the increment in weight fraction of reinforcement particles in the matrix metal produced better mechanical property like hardness but the forgeability of the cast MMCs decreases. The forgeability of the as cast MMCs were also varied with the change in thickness of the casting. The results show that the forgeability of cast metal matrix composites at the mid section of the casting is minimum compared to both end section of a three-step casting. The effect of machining parameters, e.g. cutting speed and depth of cut on the surface roughness and cutting forces at constant feed rate was investigated during experimentation. The results show that higher weight percentage of SiCp reinforcement produced a higher surface roughness and needs higher cutting forces during machining operation. It has also observed that the depth of cut and the cutting speed at constant feed rate affected the surface roughness and the cutting forces. This practical research analysis and test results on the forgeability and machinability of Al/SiC-MMC will provide useful guidelines to the present day manufacturing engineers.

An Interpretable Denoising Layer for Neural Networks Based on Reproducing Kernel Hilbert Space and its Application in Machine Fault Diagnosis

Deep learning algorithms based on neural networks make remarkable achievements in machine fault diagnosis,while the noise mixed in measured signals harms the prediction accuracy of networks.Existing denoising methods in neural networks,such as using complex network architectures and introducing sparse techniques,always suffer from the difficulty of estimating hyperparameters and the lack of physical interpretability.To address this issue,this paper proposes a novel interpretable denoising layer based on reproducing kernel Hilbert space(RKHS)as the first layer for standard neural networks,with the aim to combine the advantages of both traditional signal processing technology with physical interpretation and network modeling strategy with parameter adaption.By investigating the influencing mechanism of parameters on the regularization procedure in RKHS,the key parameter that dynamically controls the signal smoothness with low computational cost is selected as the only trainable parameter of the proposed layer.Besides,the forward and backward propagation algorithms of the designed layer are formulated to ensure that the selected parameter can be automatically updated together with other parameters in the neural network.Moreover,exponential and piecewise functions are introduced in the weight updating process to keep the trainable weight within a reasonable range and avoid the ill-conditioned problem.Experiment studies verify the effectiveness and compatibility of the proposed layer design method in intelligent fault diagnosis of machinery in noisy environments.

Scratching and Turning Experiments on Machinability of Optical Glass SF6 in Diamond Cutting Process

To improve the machinability of optical glass and achieve optical parts with satisfied surface quality and dimensional accuracy, scratching experiments with increasing cutting depth were conducted on glass SF6 to evaluate the influence of cutting fluid properties on the machinability of glass. The sodium carbonate solution of 10.5% concentration was chosen as cutting fluid. Then the critical depths in scratching experiments with and without cutting fluid were examined. Based on this, turning experiments were carried out, and the surface quality of SF6 was assessed. Compared with the process of dry cutting, the main indexes of surface roughness decrease by over 70% totally. Experimental results indicated that the machinability of glass SF6 can be improved by using the sodium carbonate solution as cutting fluid.

Machinability Investigation and Optimization of Process Parameters in Cryogenic Assisted Sustainable Turning of AISI‑L6 Tool Steel

The application of cutting fluid is significantly increased in the machining sector to improve productivity.However,the inherent characteristics of cutting fluids on ecology,environment,and society shift the interest of researchers to work on environmentally friendly cooling conditions such as cryogenic cooling.Here,the effect of cutting speed and feed rate on the machining performance of the AISI‑L6 tool steel is investigated under cryogenic cooling conditions.Then,the L9 Taguchi based grey relational analysis(GRA)is conducted to investigate the essential machining indices such as cutting energy,surface roughness,tool wear,and material removal rate(MRR).The results indicate that the cutting speed of 160 m/min and feed rate of 0.16 mm/r are the optimum parameters that significantly improves the machining performance of AISI‑L6 tool steel.

Milling Machinability of TiC Particle and TiB Whisker Hybrid Reinforced Titanium Matrix Composites

The milling machinabilities of titanium matrix composites were comprehensively evaluated to provide a theoretical basis for cutting parameter determination. Polycrystalline diamond （PCD） tools with different grain sizes and geometries, and carbide tools with and without coatings were used in the experiments. Milling forces, milling temperatures, tool lifetimes, tool wear, and machined surface integrities were investigated. The PCD tool required a primary cutting force 15 % smaller than that of the carbide tool, while the uncoated carbide tool required a primary cutting force 10% higher than that of the TiA1N-eoated tool. A cutting force of 300 N per millimeter of the cutting edge （300 N/mm） was measured. This caused excessive tool chipping. The cutting temperature of the PCD tool was 20%-30% lower than that of the carbide tool, while that of the TiA1N-coated tool was 12% lower than that of the uncoated carbide tool. The cutting temperatures produced when using water-based cooling and minimal quantity lubrication （MQL） were reduced by 100 ~C and 200 ~C, compared with those recorded with dry cutting, respectively. In general, the PCD tool lifetimes were 2--3 times longer than the carbide tool lifetimes. The roughness Ra of the machined surface was less than 0.6μm, and the depth of the machined surface hardened layer was in the range of 0.15-0.25 mm for all of the PCD tools before a flank wear land of 0.2 mm was reached. The PCD tool with a 0.8 mm tool nose radius, 0% rake angle, 10% flank angle, and grain size of （30＋2） μm exhibited the best cutting performance. For this specific tool, a lifetime of 16 rain can be expected.

A Principle of Producing Multi-ply Paper Using a Wire in an Inclined Wire Machine

In this study,we improved the dispersibility of the stocks in the headbox of an inclined wire machine to produce a distinct paper,and analyzed some factors affecting paper formation in the production of multiply paper.We used FLUENT6.3 to analyze the flow of the stocks in the headbox and select the structure of the diffusion part required for improving the dispersibility of fibers.Moreover,based on a simulation experiment,the optimal rational angle of the diffusion part(g)was found to be approximately 8°~10°,and it improved the paper formation in the case of usage of two plates.Using the equation for the formation of paper layers in the headbox of an inclined wire machine,we obtained a paper with the given basic weight by controlling the inclined angle of the wire(a),initial height of water(H),and concentration of the stocks.We considered the effect of a and H of the stocks in the headbox on the fiber distribution,and according to the results,a should be set as approximately 20°~30°and H should be maximally high.When producing multi-ply paper by a wire,the line pressure of the couch roll should be maintained at 1.8~2.0 kN/m to avoid the damage to the paper sheets.In addition,we found the optimal structure parameter of the dehydrated roll was as follows:hole ratio of approximately 30%of the dehydrated roll surface area,width of 1.5~2.0 mm,slot pitch of 5~6 mm,slot depth of 2~3 mm,and inclined angle of diffusion part(b)of 5°.

Significant risk factors for intensive care unit-acquired weakness:A processing strategy based on repeated machine learning

BACKGROUND Intensive care unit-acquired weakness(ICU-AW)is a common complication that significantly impacts the patient's recovery process,even leading to adverse outcomes.Currently,there is a lack of effective preventive measures.AIM To identify significant risk factors for ICU-AW through iterative machine learning techniques and offer recommendations for its prevention and treatment.METHODS Patients were categorized into ICU-AW and non-ICU-AW groups on the 14th day post-ICU admission.Relevant data from the initial 14 d of ICU stay,such as age,comorbidities,sedative dosage,vasopressor dosage,duration of mechanical ventilation,length of ICU stay,and rehabilitation therapy,were gathered.The relationships between these variables and ICU-AW were examined.Utilizing iterative machine learning techniques,a multilayer perceptron neural network model was developed,and its predictive performance for ICU-AW was assessed using the receiver operating characteristic curve.RESULTS Within the ICU-AW group,age,duration of mechanical ventilation,lorazepam dosage,adrenaline dosage,and length of ICU stay were significantly higher than in the non-ICU-AW group.Additionally,sepsis,multiple organ dysfunction syndrome,hypoalbuminemia,acute heart failure,respiratory failure,acute kidney injury,anemia,stress-related gastrointestinal bleeding,shock,hypertension,coronary artery disease,malignant tumors,and rehabilitation therapy ratios were significantly higher in the ICU-AW group,demonstrating statistical significance.The most influential factors contributing to ICU-AW were identified as the length of ICU stay(100.0%)and the duration of mechanical ventilation(54.9%).The neural network model predicted ICU-AW with an area under the curve of 0.941,sensitivity of 92.2%,and specificity of 82.7%.CONCLUSION The main factors influencing ICU-AW are the length of ICU stay and the duration of mechanical ventilation.A primary preventive strategy,when feasible,involves minimizing both ICU stay and mechanical ventilation duration.

Identifi cation of paralytic shellfi sh toxin-producing microalgae using machine learning and deep learning methods

Paralytic shellfi sh poisoning(PSP)microalgae,as one of the harmful algal blooms,causes great damage to the of fshore fi shery,marine culture,and marine ecological environment.At present,there is no technique for real-time accurate identifi cation of toxic microalgae,by combining three-dimensional fluorescence with machine learning(ML)and deep learning(DL),we developed methods to classify the PSP and non-PSP microalgae.The average classifi cation accuracies of these two methods for microalgae are above 90%,and the accuracies for discriminating 12 microalgae species in PSP and non-PSP microalgae are above 94%.When the emission wavelength is 650-690 nm,the fl uorescence characteristics bands(excitation wavelength)occur dif ferently at 410-480 nm and 500-560 nm for PSP and non-PSP microalgae,respectively.The identification accuracies of ML models(support vector machine(SVM),and k-nearest neighbor rule(k-NN)),and DL model(convolutional neural network(CNN))to PSP microalgae are 96.25%,96.36%,and 95.88%respectively,indicating that ML and DL are suitable for the classifi cation of toxic microalgae.

Sintering and Machinability of Monazite-Type CePO_4 Ceramics

The sintering and machinability of monazite-type CePO_4 ceramics were investigated. Relative density ≥98% and apparent porosity <2% were achieved when the monazite-type CePO_4 were sintered at 1500 ℃/1 h in air,and the maximal bending strength value (184 MPa) was achieved at this temperature. CePO_4 ceramics has a multilayer structure and an exciting 'ductility',so it can be drilled and cut with WC cutter with a small machining damage.

Machine learning for predicting the outcome of terminal ballistics events

Machine learning(ML) is well suited for the prediction of high-complexity,high-dimensional problems such as those encountered in terminal ballistics.We evaluate the performance of four popular ML-based regression models,extreme gradient boosting(XGBoost),artificial neural network(ANN),support vector regression(SVR),and Gaussian process regression(GP),on two common terminal ballistics’ problems:(a)predicting the V50ballistic limit of monolithic metallic armour impacted by small and medium calibre projectiles and fragments,and(b) predicting the depth to which a projectile will penetrate a target of semi-infinite thickness.To achieve this we utilise two datasets,each consisting of approximately 1000samples,collated from public release sources.We demonstrate that all four model types provide similarly excellent agreement when interpolating within the training data and diverge when extrapolating outside this range.Although extrapolation is not advisable for ML-based regression models,for applications such as lethality/survivability analysis,such capability is required.To circumvent this,we implement expert knowledge and physics-based models via enforced monotonicity,as a Gaussian prior mean,and through a modified loss function.The physics-informed models demonstrate improved performance over both classical physics-based models and the basic ML regression models,providing an ability to accurately fit experimental data when it is available and then revert to the physics-based model when not.The resulting models demonstrate high levels of predictive accuracy over a very wide range of projectile types,target materials and thicknesses,and impact conditions significantly more diverse than that achievable from any existing analytical approach.Compared with numerical analysis tools such as finite element solvers the ML models run orders of magnitude faster.We provide some general guidelines throughout for the development,application,and reporting of ML models in terminal ballistics problems.