What is it about?
In the modern advanced pharmaceutical sector, tablets, which constitute a significant portion of orally administered drugs, can still deviate from the manufacturer's established standards. Typical quality control methods often focus on selected specific parameters, overlooking other potential variations arising during manufacturing. Improperly manufactured drugs may contain incorrect active pharmaceutical ingredient (API) dosages, hazardous contaminants from defective ingredient batches, or be poorly formulated, all of which can reduce API bioavailability and pose serious health risks to patients. No single technique can directly detect all these anomalies. However, near-infrared hyperspectral imaging (NIR-HSI) systems can indirectly identify defective tablets by monitoring chemical and physical changes on their surface. Deviations in measures representing surface chemical or physical heterogeneity most often indicate substandard quality. This research was supported by the National Science Centre, Poland (research grant no. 2018/29/N/ST4/01547).
Featured Image
Photo by Mika Baumeister on Unsplash
Why is it important?
This study explores the potential of hyperspectral imaging (HSI) for monitoring pharmaceutical tablet integrity. We propose an advanced data analysis strategy that integrates HSI image processing techniques already verified in food analysis studies – hyperspectrogram – with one-class modeling, a recommended approach for such applications. This method enables an easy-to-apply yet comprehensive description of three-component mixture samples, capturing both physical and chemical heterogeneity variations of tablets and indicating diverse production anomalies. To the best of our knowledge, this is the first proposed application of the NIR-HSI-based modeling approach for universal use in tablet pharmaceutical quality control, capable of detecting various anomalies of unknown origin. Our study confirms that the developed expert system can identify even minor tablet defects while ensuring high model robustness.
Perspectives
In the future, similar systems could be directly implemented in pharmaceutical manufacturing for real-time drug monitoring. The study also discusses key model training challenges and bottlenecks of applying similar expert systems in the pharmaceutical sector, including developing robust chemometric models for detecting tablet defects within process analytical technology (PAT) frameworks.
Dr. Łukasz Pieszczek
University of Silesia in Katowice
Read the Original
This page is a summary of: Integrating hyperspectrograms with class modeling techniques for the construction of an effective expert system: Quality control of pharmaceutical tablets based on near-infrared hyperspectral imaging, Journal of Pharmaceutical and Biomedical Analysis, April 2025, Elsevier,
DOI: 10.1016/j.jpba.2025.116697.
You can read the full text:
Contributors
The following have contributed to this page
