Quantitative Microscopy: Particle Size/Shape Characterization, Addressing Common Errors Using ‘Analytics Continuum’ Approach

Abstract

Particle size/shape characterization of active pharmaceutical ingredient (API) is integral to successful product development. It is more of a correlative property than a decision-making measure. Though microscopy is the only technique that provides a direct measure of particle properties, it is neglected for reasons like non-repeatability and non-reproducibility which is often attributed to a) fundamental error, b) segregation error, c) human error, d) sample randomness, e) sample representativeness etc. Using the "Sucrose" as model sample, we propose "analytics continuum" approach that integrates optical microscope PSD measurements complimented by NIR spectroscopy-based trending analysis as a prescreening tool to demonstrate sample randomness and representativeness. Furthermore, plethora of statistical tests are utilized to infer population statistics. Subsequently, an attribute-based control chart and bootstrap-based confidence interval was developed to monitor product performance. A flowchart to serve as an elementary guideline is developed, which is then extended to handle more complex situations involving API crystallized from two different solvent systems. The results show that the developed methodology can be utilized as a quantitative procedure to assess the suitability of API/excipients from different batches or from alternate vendors and can significantly help in understanding the differences between material even on a minor scale.

Water Effective Diffusion Coefficient in Dairy Powder Calculated by Digital Image Processing and through Machine Learning Algorithms of CLSM Micrographs

“…Also, the estimation of particle diameter is not precise, which results in high errors and deviations. The number of investigated particles is significantly lower in comparison with other methods, which makes data statistically poor [ 30 ]. As an advantage of manual measurement is no need for special software and D eff may be simply estimated by various available photo redactors.…”

Machine Learning-Enabled NIR Spectroscopy in Assessing Powder Blend Uniformity: Clear-Up Disparities and Biases Induced by Physical Artefacts

“…The results indicated that the data is non-normal, non-linear, inhomogeneity in variance/ heteroscedasticity. That is NIR BU data were found to have violations of parametric assumptions [43,44]. On the other hand, no outliers were identified.…”

“…When k = 1, it is leave one out cross-validation (LOOCV). Bootstrapping is a statistical process that creates several simulated samples by resampling a single dataset with replacement [43,53]. Those instances that were not resampled are used in the test set.…”

“…Understanding the reliability of data through the application of diverse data quality metrics is a prerequisite; otherwise, poor quality might have adverse effects on model prediction [43,44]. In this regard, the BU data was subjected to rigorous tests to examine for normality, linearity, multicollinearity, homogeneity in variance/homoscedasticity, presence of outliers, sample adequacy, and clustering tendency.…”

“…For NIR-based BU or CU, current pre-processing approaches involving normalization, second derivative, MSC, EMSC, and SNV might not be pertinent if any of the following conditions exist: (1) data demonstrating heteroscedasticity, non-normality, multicollinearity which are the prerequisites for the parametric models [43][44][45]; (2) also, if physical properties are of simultaneous importance; (3) the types of physical fluctuations are uncontrollable or may not be representative of future samples; (4) the artefacts are non-linear/non-parametric. To the best of our knowledge, this is the first work to provide an intuitive understanding of the gaps in existing NIR-based BU analysis.…”