Developing a machine learning model for predicting venlafaxine active moiety concentration: a retrospective study using real-world evidence.

BACKGROUND: Venlafaxine is frequently prescribed for patients with depression. To control the concentration of venlafaxine within the therapeutic window for the best treatment effect, a model to predict venlafaxine concentration is necessary. AIM: Our objective was to develop a prediction model for venlafaxine concentration using real-world evidence based on machine learning and deep learning techniques. METHOD: Patients who underwent venlafaxine treatment between November 2019 and August 2022 were included in the study. Important variables affecting venlafaxine concentration were identified using a combination of univariate analysis, sequential forward selection, and machine learning techniques. Predictive performance of nine machine learning and deep learning algorithms were assessed, and the one with the optimal performance was selected for modeling. The final model was interpreted using SHapley Additive exPlanations. RESULTS: A total of 330 eligible patients were included. Five influential variables that affect venlafaxine concentration were venlafaxine daily dose, sex, age, hyperlipidemia, and adenosine deaminase. The venlafaxine concentration prediction model was developed using the eXtreme Gradient Boosting algorithm (R2 = 0.65, mean absolute error = 77.92, root mean square error = 93.58). In the testing cohort, the accuracy of the predicted concentration within ± 30% of the actual concentration was 73.49%. In the subgroup analysis, the prediction accuracy was 69.39% within the recommended therapeutic range of venlafaxine concentration within ± 30% of the actual value. CONCLUSION: The XGBoost model for predicting blood concentration of venlafaxine using real-world evidence was developed, guiding the adjustment of regimen in clinical practice.

Artificial intelligence-based model for dose prediction of sertraline in adolescents: a real-world study.

BACKGROUND: Variability exists in sertraline pharmacokinetic parameters in individuals, especially obvious in adolescents. We aimed to establish an individualized dosing model of sertraline for adolescents with depression based on artificial intelligence (AI) techniques. METHODS: Data were collected from 258 adolescent patients treated at the First Hospital of Hebei Medical University between December 2019 to July 2022. Nine different algorithms were used for modeling to compare the prediction abilities on sertraline daily dose, including XGBoost, LGBM, CatBoost, GBDT, SVM, ANN, TabNet, KNN, and DT. Performance of four dose subgroups (50 mg, 100 mg, 150 mg, and 200 mg) were analyzed. RESULTS: CatBoost was chosen to establish the individualized medication model with the best performance. Six important variables were found to be correlated with sertraline dose, including plasma concentration, PLT, MPV, GL, A/G, and LDH. The ROC curve and confusion matrix exhibited the good prediction performance of CatBoost model in four dose subgroups (the AUC of 50 mg, 100 mg, 150 mg, and 200 mg were 0.93, 0.81, 0.93, and 0.93, respectively). CONCLUSION: The AI-based dose prediction model of sertraline in adolescents with depression had a good prediction ability, which provides guidance for clinicians to propose the optimal medication regimen.

A machine learning model for predicting blood concentration of quetiapine in patients with schizophrenia and depression based on real-world data.

AIMS: This study aimed to establish a prediction model of quetiapine concentration in patients with schizophrenia and depression, based on real-world data via machine learning techniques to assist clinical regimen decisions. METHODS: A total of 650 cases of quetiapine therapeutic drug monitoring (TDM) data from 483 patients at the First Hospital of Hebei Medical University from 1 November 2019 to 31 August 2022 were included in the study. Univariate analysis and sequential forward selection (SFS) were implemented to screen the important variables influencing quetiapine TDM. After 10-fold cross validation, the algorithm with the optimal model performance was selected for predicting quetiapine TDM among nine models. SHapley Additive exPlanation was applied for model interpretation. RESULTS: Four variables (daily dose of quetiapine, type of mental illness, sex and CYP2D6 competitive substrates) were selected through univariate analysis (P < .05) and SFS to establish the models. The CatBoost algorithm with the best predictive ability (mean [SD] R2  = 0.63 ± 0.02, RMSE = 137.39 ± 10.56, MAE = 103.24 ± 7.23) was chosen for predicting quetiapine TDM among nine models. The mean (SD) accuracy of the predicted TDM within ±30% of the actual TDM was 49.46 ± 3.00%, and that of the recommended therapeutic range (200-750 ng mL-1 ) was 73.54 ± 8.3%. Compared with the PBPK model in a previous study, the CatBoost model shows slightly higher accuracy within ±100% of the actual value. CONCLUSIONS: This work is the first real-world study to predict the blood concentration of quetiapine in patients with schizophrenia and depression using artificial intelligent techniques, which is of significance and value for clinical medication guidance.

Atomic-scale insights on hydrogen trapping and exclusion at incoherent interfaces of nanoprecipitates in martensitic steels.

Hydrogen is well known to embrittle high-strength steels and impair their corrosion resistance. One of the most attractive methods to mitigate hydrogen embrittlement employs nanoprecipitates, which are widely used for strengthening, to trap and diffuse hydrogen from enriching at vulnerable locations within the materials. However, the atomic origin of hydrogen-trapping remains elusive, especially in incoherent nanoprecipitates. Here, by combining in-situ scanning Kelvin probe force microscopy and aberration-corrected transmission electron microscopy, we unveil distinct scenarios of hydrogen-precipitate interaction in a high-strength low-alloyed martensitic steel. It is found that not all incoherent interfaces are trapping hydrogen; some may even exclude hydrogen. Atomic-scale structural and chemical features of the very interfaces suggest that carbon/sulfur vacancies on the precipitate surface and tensile strain fields in the nearby matrix likely determine the hydrogen-trapping characteristics of the interface. These findings provide fundamental insights that may lead to a better coupling of precipitation-strengthening strategy with hydrogen-insensitive designs.

Thermodynamics Analysis of Multiple Microelements' Coupling Behavior in High Fatigue Resistance 50CrVA Spring Steel with Nanoparticles.

Solid solution and coupling precipitation behavior of multiple microelements in 50CrVA spring steel under different temperatures were analyzed based on thermodynamics. Quantitative relationships between the multiple microelements' contents and secondary phases, and their effects on fatigue life, were systematically studied in conjunction with the secondary phase microstructure characterization using scanning and transmission electron microscopy, etc. The solid solution contents of different microelements decreased as the temperature decreased, especially N and Ti, but the number of compounds gradually increased when the temperature decreased. Carbonitride constitutional liquation occurred in 50CrVA-S1# spring steel-containing microparticles, and without carbonitrides, constitutional liquation occurred in 50CrVA-S2# spring steel-containing nanoparticles. The experimental results indicate that the fatigue life reduces by about an order of magnitude when the secondary phase size changes from nanometers to microns, and the corresponding relationship among multiple microelements, microstructure of secondary phases, and fatigue life, was established in this spring steel.

One-Step in Situ Synthesis of Reduced Graphene Oxide/Zn-Al Layered Double Hydroxide Film for Enhanced Corrosion Protection of Magnesium Alloys.

As an effective and environmentally friendly material for corrosion prevention, layered double hydroxide (LDH) films have usually been degraded due to their inherent microporous structure. In this study, graphene derivatives were employed to enhance the corrosion resistance of LDH films. After ultrasonic treatment of a reaction solution mixture containing graphene oxide (GO) powder, a reduced graphene oxide/zinc-aluminum LDH (RGO/Zn-Al LDH) film was in situ synthesized on a magnesium alloy substrate by a one-step facile hydrothermal crystallization process. The characterization results demonstrated that the LDH nanosheets grew on both the GO surface and the magnesium substrate, and thus the agglomeration of graphene was effectively prevented. Furthermore, the GO plates were simultaneously reduced into RGO, which has better corrosion resistance. The as-prepared samples were individually assessed by potentiodynamic polarization measurements, and the RGO/Zn-Al LDH film showed good corrosion resistance with a lower corrosion current density (0.546 µA/cm2) than that of the bare substrate (33.2 µA/cm2) and Zn-Al LDH film (4.33 µA/cm2). The penetration resistance of the Zn-Al LDH film to a corrosive environment was significantly improved through the organic combination with graphene oxide, and this method provides a simple and facile approach to effectively enhance the corrosion protection performance of LDH materials.

Thermodynamic Analysis of Ti3O5Nanoparticles Formed in Melt and Their Effects on Ferritic Steel Microstructure.

Based on the Wagner's formalism combined with mass conservation, a thermodynamic analysis method has been developed previously. This method enables the calculation of the equilibrium matrix composition, precipitate composition and precipitate total molar fraction for TixOy(s) in molten metal, which can be determined at any appropriate temperature. In this present study, the Ti3O5 phase precipitation and the quantitative relationship between the addition of Ti, O and Ti3O5 in the molten steel were studied using the thermodynamic model. Using the combined multipoint dispersion supply method, electromagnetic stirring and well-dispersed 5-nm Ti3O5 nanoparticles were fabricated in the ferrite matrix of the as-cast high-strength steel with 0.05 wt % Ti-0.002 wt % O. The as-cast microstructure was improved by the homogeneously dispersed Ti3O5 nanoparticles through heterogeneous nucleation and grain refinement.

Aluminium-inhibited NO3- uptake is related to Al-increased H2O2 content and Al-decreased plasma membrane ATPase activity in the root tips of Al-sensitive black soybean.

In this study, Al-sensitive black soybean (Glycine max (L.) Merr.) specimens were treated in Hoagland solutions containing 50-400µM Al for 1-4 days. The measurement for NO3- uptake showed that the NO3- uptake decreased gradually as the Al concentration and treatment time increased, suggesting that Al stress significantly reduced the NO3- uptake by soybean. Under 100-µM Al stress for 4 days, the plasma membrane (PM) ATPase activity (inorganic phosphate (Pi) release), H+ pump activity, phosphorylation of PM ATPase and its interaction with 14-3-3 protein in soybean root tips were all smaller than those in the root tips of control plants. The addition of 150µM Mg2+ in Al treatment solutions significantly alleviated the Al inhibition of NO3- uptake in soybean. The presence of Mg2+ in a 100-µM Al solution pronouncedly enhanced PM ATPase activity, H+ pump activity, phosphorylation of PM ATPase and its interaction with 14-3-3 protein in soybean root tips. The application of 2mM ascorbic acid (AsA, an H2O2 scavenger) in Al treatment solutions significantly decreased Al-inhibited NO3- uptake in soybean. The cotreatment of soybeans with 2mM AsA and 100µM Al significantly reduced H2O2 accumulation and increased the PM ATPase activity, H+ pump activity, phosphorylation of PM H+-ATPase and its interaction with 14-3-3 protein in soybean root tips. The evidence suggested that Al-inhibited NO3- uptake is related to Al-increased H2O2 content and Al-decreased phosphorylation of PM ATPase and its interaction with 14-3-3 protein as well as PM ATPase activity in the root tips of soybean.

Interface and Strain Energy Revolution Texture Map To Predict Structure and Optical Properties of Sputtered PbSe Thin Films.

The preferred growth orientation of the sputtered lead selenide (PbSe) thin films on Si(100) substrates was thermodynamically simulated and calculated on the basis of the density functional theory. The results showed that the total free energy variation during the grain growth is dominated by the interface and strain energy minimization under certain conditions, indicating that the preferred growth orientation and related optical properties of the PbSe thin films can be effectively modified by these two energy variations. Thermodynamically, the PbSe[200] and PbSe[220] preferred orientations are obtained when the interface and strain energy minimization dominate the total free energy variation, respectively. A texture map related to the interface and strain energy revolution was obtained, which can be used to predict the structure and optical properties of the sputtered PbSe thin films, and its applicability was confirmed by the real X-ray diffraction and Fourier transform infrared spectroscopy experimental results of four midfrequency sputtered PbSe thin films with designed thickness and microstrain deposited on Si(100) substrates.

Development and application of metal materials in terms of vascular stents.

BACKGROUND: With life pace accelerated, poor diet habits developed, psychological burden enhanced and many other factors, the incidence of coronary heart disease, atherosclerosis and other cardiovascular cerebrovascular diseases has been increased year by year, which are serious threat to human health. OBJECTIVE: Provide relational references for the similar researchers after metal stent materials were reviewed and prospected. METHODS: This paper reviews the development and application of metal materials in terms of vascular stents, focusing on the advantages and disadvantages of 316L stainless steel, nitinol super-elastic alloys, cobalt-based alloys (Co-Cr-Ni-Mo-Mn), magnesium-based alloy, iron-based alloys and tantalum metal stents as well as in clinical practice research and application. RESULTS: Recognize the advantages and disadvantages of different metal stent materials as well as in clinical practice research and application. CONCLUSIONS: Although metal stents have been widely used in clinical practice, there are still many problems to be solved, especially to improve mechanical properties and biological activity. Strong immune rejection is also a problem. Therefore, it will be a significant direction for future material research to treat surface modification, further improve the biocompatibility, reduce the thrombosis and completely eliminate the rejection and vascular restenosis. In addition, the stent materials should be developed toward controllable degradation and special features in the future.

Si photoanode protected by a metal modified ITO layer with ultrathin NiO(x) for solar water oxidation.

We report an ultrathin NiOx catalyzed Si np(+) junction photoanode for a stable and efficient solar driven oxygen evolution reaction (OER) in water. A stable semi-transparent ITO/Au/ITO hole conducting oxide layer, sandwiched between the OER catalyst and the Si photoanode, is used to protect the Si from corrosion in an alkaline working environment, enhance the hole transportation, and provide a pre-activation contact to the NiOx catalyst. The NiOx catalyzed Si photoanode generates a photocurrent of 1.98 mA cm(-2) at the equilibrium water oxidation potential (EOER = 0.415 V vs. NHE in 1 M NaOH solution). A thermodynamic solar-to-oxygen conversion efficiency (SOCE) of 0.07% under 0.51-sun illumination is observed. The successful development of a low cost, highly efficient, and stable photoelectrochemical electrode based on earth abundant elements is essential for the realization of a large-scale practical solar fuel conversion.

Metal oxide composite enabled nanotextured Si photoanode for efficient solar driven water oxidation.

We present a study of a transition metal oxide composite modified n-Si photoanode for efficient and stable water oxidation. This sputter-coated composite functions as a protective coating to prevent Si from photodecomposition, a Schottky heterojunction, a hole conducting layer for efficient charge separation and transportation, and an electrocatalyst to reduce the reaction overpotential. The formation of mixed-valence oxides composed of Ni and Ru effectively modifies the optical, electrical, and catalytic properties of the coating material, as well as the interfaces with Si. The successful application of this oxide composite on nanotextured Si demonstrates improved conversion efficiency due to enhanced catalytic activity, minimized reflection, and increased surface reaction sites. Although the coated nanotextured Si shows a noticeable degradation from 500 cycles of operation, the oxide composite provides a simple method to enable unstable photoanode materials for solar fuel conversion.

[Study on genotoxicity of aldicarb and methomyl].

Genotoxicity of aldicarb and methomyl was explored. The aldicarb and methomyl were diluted by the deionized water respectively, and then five concentrations of aldicarb were generated as 0.002, 0.02, 0.2, 2, 20 microg/L, methomyl as 0.02, 0.2, 2, 20, 200 microg/L. The micronucleus of carp erythrocyte was counted by micronucleus test. The mutation of bacteria was assessed by Ames test. The DNA damage of human lymphocytes was tested by comet assay. The genotoxicity of aldicarb and methomyl was estimated by the three toxicology tests mentioned above. The results showed that, in the micronucleus test, both any concentration of two pesticides were not able to induce higher frequency of micronucleus in carp erythrocyte (p > 0.05). Under condition of metabolic inactivation, although the number of colony with back mutation in any concentration of two pesticides did not exceed the double number of those with spontaneous mutation, the revertants of TA97 strains in the aldicarb 2-20 microg/L and the methomyl 20-200 microg/L were (129.17 +/- 17.00), (129.50 +/- 18.28), (109.83 +/- 10.80) and (114.17 +/- 9.37) entries/plate, respectively, they were significantly greater than those in spontaneous mutation (p < 0.05, p < 0.01). In the methomyl 200 microg/L group, the revertants of TA100 and TA102 strains were (147.83 +/- 23.29) and (275.83 +/- 20.63) entries/plate, respectively, they are significantly higher than that of the control group under condition of metabolic activation (p < 0.05). In comet assay, both the high concentration groups of aldicarb and methomyl resulted in different degrees of DNA damage of human peripheral blood lymphocytes. Compared with deionized water group, all of three indexes of comet assay in the aldicarb 20 microg/L groups and the methomyl 200 microg/L groups were significantly higher (p < 0.01). Despite that both aldicarb and methomyl did not results in damaging chromosome carp erythrocyte and producing apparent mutagenicity, the effect of mutagenicity and DNA damage in human lymphocytes were observed in high concentration groups of both aldicarb and methomyl. Water polluted by aldicarb and methomyl may have the potential adverse effects on the environment and human health.

An ultra-performance liquid chromatography-tandem mass spectrometry method for determination of microcystins occurrence in surface water in Zhejiang Province, China.

A novel method had been developed for determination of microcystin in surface water by ultra-performance liquid chromatography combined with tandem mass spectrometry (UPLC-MS/MS). Water samples after being concentrated and cleaned with solid phase extraction (SPE) with Oasis HLB cartridge were separated on AcQuity UPLC BEH C18 column (1.7 microm, 1.0 x 50 mm) with a gradient elution and a mobile phase consisting of 0.1% formic acid in water and methanol, and then detected with an electrospray ionization tandem mass spectrometry in positive ion mode with multiple reaction monitoring. Compared with traditional liquid chromatography, it took much less time for UPLC to analyze microcystin. Additionally, the UPLC-MS/MS method produced satisfactory reliability, sensitivity, and accuracy. Recoveries of the four microcystins were 91.7-111% with relative standard deviation being 7.9-12%. The calibration curves for microcystins were linear with correlation coefficient being larger than 0.99. Limit of quantification of the four microcystins (microcystin-LR, RR, LW, and LF) after 1000-fold SPE pre-concentration, were 2.5, 6.0, 2.5, and 1.3 ng/L. The presented method had been successfully used to detect microcystin in surface water from the Qiantang River, the West Lake, and three drinking water reservoirs, which represented water bodies of river, lake, and drinking water reservoir, respectively. It was shown that microcystin pollution in drinking water reservoirs was the heaviest, and microcystin-LR and RR were the predominant kinds, with concentrations ranging 0.045-2.73 and 0.021-1.36 microg/L, respectively. The concentrations of microcystins in river and lake were all under detection limit.

[Determination of trace barium in soil and sediment by Zeeman graphite AAS with coated graphite tube].

The sample was decomposed by HNO3-HF-HClO4. Using a tungsten-coated graphite tube, trace barium in soil and sediment was determined by Zeeman graphite AAS. To avoid producing carbide, the graphite tube was coated with tungsten. Tungsten and carbon in the surface layer of graphite tube became tungsten carbide. Tungsten carbide prevented barium and carbon from forming barium carbide, which in turn not only led to a long service life for the tube, but also increased greatly the sensitivity and precision of the determination Ba. Tungsten carbide belongs to internal filled type and can give reduction environment. To some extent, the common interfering elements co-existing in the soil and sediment had little chance to form oxides to interfere the determination of Ba in the atomization period. The method was easy and sensitive. The detection limit of Ba was 4.2 x 10(-10) g x g(-1). The relative standard deviation (RSD) was in the range of 2.0%-6.5% (n = 4). The relative deviations from the certificated values of standard soils were under 5%.

[Determination of trace microcystins in water by ultra performance liquid chromatography/tandem mass spectrometry].

An analytical method for the analysis of trace microcystins (MCYST) in water was developed using solid phase extraction (SPE) for enrichment and ultra performance liquid chromatography/tandem mass spectrometry (UPLC/MS/MS) for detection. One litre of water was passed through SPE columns, the extracted sample was rinsed off by 10 mL methanol, then evaporated to 1.0 mL before being analyzed with UPLC/MS/MS. The effect of formic acid concentration in the mobile phase on the sensitivity was studied and the results showed that 0.1% was the optimum concentration. Four microcystins, MCYST-LR, RR, LW, LF, can be separated and detected in 5 min, which is much shorter than that by the conventional liquid chromatography. The detection limits were 1.3 - 6.0 ng/L, and the recoveries were 91.1% - 111%. The calibration curves showed good linearity in the range of 1.0 microg/L - 1.0 mg/L with correlation coefficients larger than 0.99. The method was also applied to determine MCYST in real water samples from three reservoirs in Zhejiang Province, and the results showed that the concentrations of LR and RR were 0.044 7 - 2.73 microg/L and 0.020 8 - 1.36 microg/L respectively, and LW and LF were not detected.