Case Study 1: Quantitative Microscopy

NIR Based Sample Prescreening

NIR-PCA Based Trend Analysis (If no trend derivable based on repeatability or reproducibility, Sample Lot is Homogeneous)

Sampling Strategy: Based on NIR-PCA

If homogeneous: 1 batch, 1 reproducible, 5 repeatable and 100 particles (1x1x5x100=~500 particles) would be sufficient
If heterogeneous: 3 batch, 3 reproducible, 5 repeatable and 100 particles (3x3x5x100=~4500 particles) need to be studied

Automated Image Analysis by Confocal Microscopy

Data Quality Metrics

Approach

Test for Multicollinearity

Test for Normality

Test for Heteroscedasticity

Test for Precision

Test for Population

*Understand whether PSD requires Parametric or Non- parametric Statistics

Bootstrap model – Statistical tool

*Sample Variability Can Be Understood by Calculating Interval Estimate

Percentile Based Control Chart

*For Monitoring Particle Size Changes over Time/Batch/Lot/Vendor etc

Delivery Impact

Non-compendial Method Development

CREDIBILITY FROM EDITOR-IN-CHIEF : Manuscript Accepted as Most Significant and Most Original Work

Muthudoss et al., 2021. Quantitative Microscopy: Particle Size/Shape Characterization-Addressing Common Errors in Microscopy using the ‘Analytics Continuum’ approach, Journal of Pharmaceutical Sciences, 110 (2), 833-849. (“Virtual Issue: Most Original and Most Significant Scientific Findings”) https://doi.org/10.1016/j.xphs.2020.09.022