A low-cost and portable near-infrared spectrometer using open-source multivariate data analysis software for rapid discriminatory quality assessment of medroxyprogesterone acetate injectables

Spectrochim Acta A Mol Biomol Spectrosc. 2021 Oct 5:259:119917. doi: 10.1016/j.saa.2021.119917. Epub 2021 May 5.

Abstract

Medroxyprogesterone acetate (MPA) injectable suspensions are used by millions of women for family planning and hormonal therapy. Falsified or substandard medications may result in a health risk for consumers. Near-infrared spectroscopy (NIR) has previously been applied as a means of non-destructive and rapid screening of product quality compliance. These methods offer advantages but can be logistically and cost prohibitive for field use in resource limited areas. Here, a handheld spectrometer (900-1700 nm) with open-sourced software is used to evaluate vials of MPA from three suppliers (N = 227 vials) and verified by a benchtop UV-VIS-NIR (350-2500 nm) with licensed software. Multivariate data analysis assesses the spectral signatures of samples and builds a discriminant classification method based on Mahalanobis distances calculated from a principal component analysis scores. The handheld device paired with open-source software resulted in a product discrimination accuracy of 100% (verified by benchtop UV-VIS-NIR and chemical testing data) as well indicating that the low-cost field portable device is suitable for rapidly assessing samples in resource limited areas for consistency of manufacturing and sourcing.

Keywords: Medroxyprogesterone acetate; Non-destructive; Open-source; Portable spectroscopy; Rapid testing.

MeSH terms

  • Data Analysis*
  • Humans
  • Medroxyprogesterone Acetate*
  • Multivariate Analysis
  • Software
  • Spectroscopy, Near-Infrared

Substances

  • Medroxyprogesterone Acetate