Kennard-Stone method outperforms the Random Sampling in the selection of calibration samples in SNPs and NIR data

ABSTRACT: Splitting the whole dataset into training and testing subsets is a crucial part of optimizing models. This study evaluated the influence of the choice of the training subset in the construction of predictive models, as well as on their validation. For this purpose we assessed the Kennard-Stone (KS) and the Random Sampling (RS) methods in near-infrared spectroscopy data (NIR) and marker data SNPs (Single Nucleotide Polymorphisms). It is worth noting that in SNPs data, there is no knowledge of reports in the literature regarding the use of the KS method. For the construction and validation of the models, the partial least squares (PLS) estimation method and the Bayesian Lasso (BLASSO) proved to be more efficient for NIR data and for marker data SNPs, respectively. The evaluation of the predictive capacity of the models obtained after the data partition occurred through the correlation between the predicted and the observed values, and the corresponding square root of the mean squared error of prediction. For both datasets, results indicated that the results from KS and RS methods differ statistically from each other by the F test (P-value < 0.01). The KS method showed to be more efficient than RS in practically all repetitions. Also, KS method has the advantage of being easy and fast to be applied and also to select the same samples, which provides excellent benefits in the following analyses.

Being Uncertain in Chromatographic Calibration—Some Unobvious Details in Experimental Design

This is an introductory tutorial and review about the uncertainty problem in chromatographic calibration. It emphasizes some unobvious, but important details influencing errors in the calibration curve estimation, uncertainty in prediction, as well as the connections and dependences between them, all from various perspectives of uncertainty measurement. Nonuniform D-optimal designs coming from Fedorov theorem are computed and presented. As an example, all possible designs of 24 calibration samples (3–8, 4–6, 6–4, 8–3 and 12–2, both uniform and D-optimal) are compared in context of many optimality criteria. It can be concluded that there are only two independent (orthogonal, but slightly complex) trends in optimality of these designs. The conclusions are important, as the uniform designs with many concentrations are not the best choices, contrary to some intuitive perception. Nonuniform designs are visibly better alternative in most calibration cases.

Assessment of the possibility of modeling nonlinear multivariable calibration function using artificial neural networks

The use of artificial neural networks (ANNs) is considered justified when studying the problems that do not have a generally accepted solution algorithm. One of such problems in X-ray fluorescence analysis (XRF) is a control of the metal content in atmospheric air and air of the working area. Determination of the calibration characteristics is raveled by the lack of standard samples of the composition of aerosols collected on the filter. To solve this problem, synthetic calibration samples (CS) were manufactured as a thin organic film containing a powder material of the known chemical composition. The weight of the film samples varied within a range of 40 – 155 mg to simulate different aerosol loading of the filters and the content of components in them changed 20 – 200 times which corresponds to the samples of real aerosols. The possibility of modeling a nonlinear calibration multivariable function using artificial neural networks was evaluated in analysis of 38 film calibration samples (from 40 to 100 mg). The structure of the neural network, activation functions, learning algorithms have been investigated. Modeling was performed using an academic version of the BaseGroup Deductor analytical platform. It is shown that implementation of the back propagation of errors leads to much higher values of the error of analysis compared to the error of the regression calibration functions, whereas the Resilient Propagation algorithm provides the smallest values of the error of vanadium determination (Sr) in the calibration samples of aerosols. The range of low content of the elements in the training set is determined with a greater error compared to high content range, and therefore, the sigmoid activation function leads to unsatisfactory accuracy of the analysis results, and preference should be given to hyperbolic tangent (tanh).

The Operational Inflight Radiometric Uniform Calibration of a Directional Polarimetric Camera

The directional polarimetric camera (DPC) on-board the GF-5A satellite is designed for atmospheric or water color detection, which requires high radiometric accuracy. Therefore, in-flight calibration is a prerequisite for its inversion application. For large field optical sensors, it is very challenging to ensure the consistency of radiation detection in the whole field of view in the space environment. Our work proposes a vicarious in-flight calibration method based on sea non-equipment sites (visible bands) and land non-equipment sites (all bands). Combined with environmental parameters and radiation transmission calculations, we evaluated the radiation detection accuracy of the 0° to 60° view zenith angle of the DPC in each band. Our calibration method is based on the single-day normalized radiance data measured by the DPC. Through data selection, enough calibration samples can be obtained in a single day (the number of desert samples is more than 5000, and the number of calibration samples of the ocean is more than 2.8×106). The measurements are compared with the simulation of 6SV VRT code or look-up tables. The massive amount of data averages the uncertainty of a single-point calculation. Although the uncertainty of a single sample is significant, the final fitting of the curve of the variation in the radiometric calibration coefficient with the observation angle can still keep the root mean squared error at approximately 2–3% or even lower, and for visible bands, the calibration results for both ocean sites and desert sites are in good agreement regarding the non-uniformity of the sensor.

One-Class Drift Compensation for an Electronic Nose

Drift compensation is an important issue in an electronic nose (E-nose) that hinders the development of E-nose’s model robustness and recognition stability. The model-based drift compensation is a typical and popular countermeasure solving the drift problem. However, traditional model-based drift compensation methods have faced “label dilemma” owing to high costs of obtaining kinds of prepared drift-calibration samples. In this study, we have proposed a calibration model for classification utilizing a single category of drift correction samples for more convenient and feasible operations. We constructed a multi-task learning model to achieve a calibrated classifier considering several demands. Accordingly, an associated solution process has been presented to gain a closed-form classifier representation. Moreover, two E-nose drift datasets have been introduced for method evaluation. From the experimental results, the proposed methodology reaches the highest recognition rate in most cases. On the other hand, the proposed methodology demonstrates excellent and steady performance in a wide range of adjustable parameters. Generally, the proposed method can conduct drift compensation with limited one-class calibration samples, accessing the top accuracy among all presented reference methods. It is a new choice for E-nose to counteract drift effect under cost-sensitive conditions.

Quantitative characterisation of extended-release tablets with quetiapine using NIR-chemometric methods

This study aims to develop and validate NIR-chemometric methods for quantifying the API (quetiapine) and two excipients in extended-release tablets without sample preparation. The calibration samples were prepared following an experimental design with three variables (quetiapine, HPMC and microcrystalline cellulose) and five levels (concentration 80-90-100-110-120% of API). The validation set included three concentration levels (90-100-110%). The best calibration algorithms have used the same pre-treatment method (SNV), and different factors: 7 PLS factors (R² -0,966 and RMSEP-6,84) for quetiapine, 8 PLS factors (R²-0,927 and RMSEP 6,84) for HPMC and 3 PLS factors (R²-0,983 and RMSEP-7,26) for microcrystalline cellulose. The methods were fully validated according to the ICH guidance using these calibration models. Regarding the trueness of the methods, the recovery was between 98.51 and 99.43 for quetiapine, between 98.61 and 100.85 for HPMC, and between 100.61 and 101.78 for microcrystalline cellulose. According to data obtained, the accuracy profile was ± 5 for quetiapine and HPMC, and ± 6 for microcrystalline cellulose. Linearity profile was also in establish intervals at accuracy and the R2 value was 0.983 for quetiapine, 0.948 for HPMC and 0.997 for microcrystalline cellulose. In conclusion, the developed NIR-chemometric methods have suitable reproducibility, accuracy, linearity and can be used for quantitative characterisation of extended-release tablets with quetiapine, with any sample preparation.

Developing an Instrument to Measure Maternal Knowledge and Attitude of Oral Health on Children Under 3 Years

Objective Parental knowledge, belief, and attitude about oral health affect children’s dental cleaning behavior. Further research on maternal knowledge and attitude about early-age children’s oral health has been suggested to identify factors related to pediatric dental and oral health. For that purpose, a measurement instrument is needed. The research aimed to develop an instrument to measure maternal knowledge and attitude about under 3-year-old children’s oral health. Materials and Methods Using a validity-based approach, we held a series of basic consultation (workshops and interviews) to identify the conceptually different domains. Instrument items were derived from relevant theories. Cognitive interviews were conducted to ensure that the items were properly understood. The items were first tested among the population calibration samples (n = 150). All instrument items were analyzed for reliability and validity. Results In total, 15 items were derived from Bloom’s theory of learning and were developed for the knowledge instrument, while 10 others were developed for the attitude instrument. The reliability analysis yielded Cronbach’s α scores of 0.620 for the knowledge instrument and 0.565 for the attitude instrument. All items were considered valid based on Pearson’s correlation test results. Conclusion The instruments on maternal knowledge and attitude about under 3-year-old children’s oral health consisted of three dimensions, respectively. Both instruments have been tested and analyzed and therefore are applicable for use.

Determination of Mitotane (DDD) and Principal Metabolite by a Simple HPLC-UV Method and Its Validation in Human Plasma Samples

Mitotane (DDD) is prescribed in adrenocortical renal carcinoma. Its principal metabolite, dichlorodiphenylethene (DDE), can accumulate in fat tissues and from a toxicological point of view, is probably more interesting than the other metabolite dichlorodiphenylacetate (DDA). Therapeutic Drug Monitoring (TDM) of DDD plasma concentrations is required to combine therapeutic efficacy with acceptable toxicity. Therefore, we developed a simple and fast HPLC-UV method to monitor plasma concentrations after a liquid–liquid extraction of plasma calibration samples, quality controls, and anonymous plasma samples with unknown DDD and DDE concentrations. Samples were injected into an HPLC instrument and peaks of mitotane (DDD), DDE and aldrin (internal standard, IS) were resolved by a stationary phase C18 column (250 mm × 4.6 mm, 5 μm), maintained at 35 °C. Mobile phase, made by water/acetonitrile (10/90, v/v), was pumped at a flow of 1.0 mL/min, and absorbance was monitored at a wavelength of 226 nm. Average recovery was 95% for all analytes, and the method was linear for both DDD (r2 = 0.9988, range 1–50 mg/L) and DDE (r2 = 0.9964, range 1–40 mg/L). The values of limit of detection and quantitation were 0.102 and 0.310 mg/L for DDD and 0.036 and 0.108 mg/L for DDE, respectively. The retention time values of DDD, DDE and IS were 7.06, 9.42 and 12.60 min, respectively. The method was successfully validated according to FDA guidelines and finally adopted for routine TDM.

Quantitative Detection of Chromium Pollution in Biochar Based on Matrix Effect Classification Regression Model

Returning biochar to farmland has become one of the nationally promoted technologies for soil remediation and improvement in China. Rapid detection of heavy metals in biochar derived from varied materials can provide a guarantee for contaminated soil, avoiding secondary pollution. This work aims first to apply laser-induced breakdown spectroscopy (LIBS) for the quantitative detection of Cr in biochar. Learning from the principles of traditional matrix effect correction methods, calibration samples were divided into 1–3 classifications by an unsupervised hierarchical clustering method based on the main elemental LIBS data in biochar. The prediction samples were then divided into diverse classifications of calibration samples by a supervised K-nearest neighbor (KNN) algorithm. By comparing the effects of multiple partial least squares regression (PLSR) models, the results show that larger numbered classifications have a lower averaged relative standard deviations of cross-validation (ARSDCV) value, signifying a better calibration performance. Therefore, the 3 classification regression model was employed in this study, which had a better prediction performance with a lower averaged relative standard deviations of prediction (ARSDP) value of 8.13%, in comparison with our previous research and related literature results. The LIBS technology combined with matrix effect classification regression model can weaken the influence of the complex matrix effect of biochar and achieve accurate quantification of contaminated metal Cr in biochar.