Quantification and handling of nonlinearity in Raman micro-spectrometry of pharmaceuticals

Deep learning-based Raman spectroscopy qualitative analysis algorithm: A convolutional neural network and transformer approach

Raman spectroscopy is a general and non-destructive detection technique that can obtain detailed information of the chemical structure of materials. In the past, when using chemometric algorithms to analyze the Raman spectra of mixtures, the challenges of complex spectral overlap and noise often limited the accurate identification of components. The emergence of deep learning has introduced a novel approach to qualitative analysis of mixed Raman spectra. In this paper, we propose a deep learning-based Raman spectroscopy qualitative analysis algorithm (RST) by borrowing the ideas of convolutional neural network and Transformer. By transforming the Raman spectrum into 64 word vectors, the contribution weights of each word vector to the components are obtained. For the 75 spectral data used for validation, the positive identification rate can reach 100.00 %, the recall rate can reach 99.3 %, the average identification score can reach 9.51, and it is applicable to the fields of Raman and surface-enhanced Raman spectroscopy. Furthermore, compared with traditional CNN models, RST has excellent accuracy and robustness in identifying components in complex mixtures. The model's interpretability has been enhanced, aiding in a deeper understanding of spectroscopic learning patterns for future analysis of more complex mixtures.

Quantitative analysis of low-content impurity crystal forms in canagliflozin tablets by NIR solid-state analysis technique

Canagliflozin (CFZ) tablets was a commercially new class of anti-diabetic drug, CFZ had various anhydrate crystal forms and two hydrate crystal forms (Canagliflozin hemihydrate (Hemi-CFZ) and Canagliflozin monohydrate (Mono-CFZ) crystal form). The active pharmaceutical ingredients (APIs) of commercially available CFZ tablets were Hemi-CFZ, was easily convert to CFZ or Mono-CFZ under the influence of temperature, pressure, humidity and other factors in tablets processing, storage, and transportation, thus affected bioavailability and efficacy of tablets. Therefore, quantitative analysis of low-content CFZ and Mono-CFZ in tablets was essential to control tablets' quality. The main objective of this study was to explore the feasibility and in-depth explain its quantitative analysis mechanism of NIR for quantitative analysis of low-content CFZ/Mono-CFZ in CFZ tablets. PLSR models for low-content CFZ/Mono-CFZ were established by NIR solid-state analysis technique in different resolutions with different wavenumber regions combined with various pretreatments methods (such as Multiplicative Scatter Correction (MSC), Standard Normal Variate (SNV), Savitzky-Golay First Derivative (SG1st), Savitzky-Golay Second Derivative (SG2nd) and Wavelet Transform (WT)), and the PLSR models were verified. The feasibility of NIR spectroscopy for quantitative analysis of low-content CFZ and Mono-CFZ in CFZ tablets was discussed and analyzed from multiple perspectives, which included the distribution of effective information on the spectrum, the influence of resolution on PLSR models performance, the variance contribution/cumulative variance contribution of PLSR model principal components (PCs), the relation of PCI loadings, scores of the spectra and CFZ/Mono-CFZ content, and the mechanism of quantitative analysis was in-depth explained simultaneously. Eventually the most suitable PLSR models in 0.0000-10.0000 % w/w % obtained. That can provide theoretical support for quantitative analysis of low-content impurity crystal during the production, storage and transportation of CFZ tablets, thus provide reference methods for quality control of CFZ tablets and a reliable reference method for quantitative analysis of impurity crystal forms and quality control of similar drugs.

Quantitative analysis of low content polymorphic impurities in canagliflozin tablets by PXRD, NIR, ATR-FITR and Raman solid-state analysis techniques combined with stoichiometry

Canagliflozin (CFZ) was a commercially new class of anti-diabetic drug, which had various anhydrate crystal forms and two hydrate crystal forms (Canagliflozin hemihydrate (Hemi-CFZ) and Canagliflozin monohydrate (Mono-CFZ) crystal form). Commercially available CFZ tablets' active pharmaceutical ingredient (API) was Hemi-CFZ, which was easy conversion to CFZ or Mono-CFZ under the influence of temperature, pressure, humidity and other factors in tablets processing, storage, and transportation, thus affected bioavailability and efficacy of tablets. Therefore, quantitative analysis low content of CFZ and Mono-CFZ in tablets was essential to control tablets' quality. The main objective of this study was to examine the feasibility of Powder X-ray Diffraction (PXRD), Near Infrared Spectroscopy (NIR), Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (ATR-FTIR) and Raman for quantitative analysis the low content of CFZ or Mono-CFZ in ternary mixtures. PLSR calibration models for low content of CFZ and Mono-CFZ were established by the solid analysis techniques of PXRD, NIR, ATR-FTIR and Raman combined with various pretreatments (such as Multiplicative Scatter Correction (MSC), Standard Normal Variate (SNV), Savitzky-Golay First Derivative (SG^1st), Savitzky-Golay Second Derivative (SG^2nd) and Wavelet Transform (WT)), and the correction models were verified. However, compared with PXRD, ATR-FTIR and Raman, NIR due to its water sensitivity was the most suitable for the quantitative analysis low content of CFZ or Mono-CFZ in tablets. Partial Least Squares Regression (PLSR) model for quantitative analysis low content of CFZ in tablets was as follow: Y = 0.0480 + 0.9928 X, R² = 0.9986, LOD = 0.1596 %, LOQ = 0.4838 %, SG^1st + WT pretreated. And that of Mono-CFZ were Y = 0.0050 + 0.9996 X, R² = 0.9996, LOD = 0.0164 %, LOQ = 0.0498 %, MSC + WT pretreated and Y = 0.0051 + 0.9996 X, R² = 0.9996, LOD = 0.0167 %, LOQ = 0.0505 %, SNV + WT pretreated, respectively. That can be used for quantitative analysis of impurity crystal content in drug production to ensure drug quality.

A Review of Pharmaceutical Robot based on Hyperspectral Technology

The quality and safety of medicinal products are related to patients’ lives and health. Therefore, quality inspection takes a key role in the pharmaceutical industry. Most of the previous solutions are based on machine vision, however, their performance is limited by the RGB sensor. The pharmaceutical visual inspection robot combined with hyperspectral imaging technology is becoming a new trend in the high-end medical quality inspection process since the hyperspectral data can provide spectral information with spatial knowledge. Yet, there is no comprehensive review about hyperspectral imaging-based medicinal products inspection. This paper focuses on the pivotal pharmaceutical applications, including counterfeit drugs detection, active component analysis of tables, and quality testing of herbal medicines and other medical materials. We discuss the technology and hardware of Raman spectroscopy and hyperspectral imaging, firstly. Furthermore, we review these technologies in pharmaceutical scenarios. Finally, the development tendency and prospect of hyperspectral imaging technology-based robots in the field of pharmaceutical quality inspection is summarized.

Potential of Raman spectroscopy in facilitating pharmaceutical formulations development - An AI perspective

Drug development is time-consuming and inherently possesses a high failure rate. Pharmaceutical formulation development is the bridge that links a new chemical entity (NCE) to pre-clinical and clinical trials, and has a high impact on the efficacy and safety of the final drug product. Further, the time required for this process is escalating as formulation techniques are becoming more complicated due to the rising demands for drug products with better efficacy and patient compliance, as well as the inherent difficulties of addressing the unfavorable properties of NCEs such as low water solubility. The advent of artificial intelligence (AI) provides possibilities to accelerate the drug development process. In this review, we first examine applications of AI methods in different types of pharmaceutical formulations and formulation techniques. Moreover, as availability of data is the engine for the advancement of AI, we then suggest a potential way (i.e. applying Raman spectroscopy) for faster high-quality data gathering from formulations. Raman techniques have the capability of analyzing the composition and distribution of components and the physicochemical properties thereof within formulations, which are prominent factors governing drug dissolution profiles and subsequently bioavailability. Thus, useful information can be obtained bridging formulation development to the final product quality.

A Raman imaging-based technique to assess HPMC substituent contents and their effects on the drug release of commercial extended-release tablets

In this study, we tried to assess the substitute contents of HPMC used in commercial extended-release tablets directly by an innovative Raman imaging-based analysis technique and find their effects on the in vitro performance of these pharmaceuticals. Twenty-seven batches of metformin hydrochloride extended-release tablets from various sources were collected in the Chinese mainland market. While Raman imaging was used to qualitatively analyze the composition of the tablets, the MeO and HPO contents of HPMC were quantitatively assessed by a newly proposed calculation method based on the Raman intensity of corresponding characteristic band. Additionally, the dissolution test was performed to evaluate the relationship between HPMC substitution pattern and in vitro behavior. In sum, our findings indicate that the drug release rate can be downregulated by increasing the MeO content of HPMC, while the high HPO content would largely eliminate the variation of drug release profiles among batches.

In-line Raman spectroscopic monitoring and feedback control of a continuous twin-screw pharmaceutical powder blending and tableting process

The integration of Process Analytical Technology (PAT) initiative into the continuous production of pharmaceuticals is indispensable for reliable production. The present paper reports the implementation of in-line Raman spectroscopy in a continuous blending and tableting process of a three-component model pharmaceutical system, containing caffeine as model active pharmaceutical ingredient (API), glucose as model excipient and magnesium stearate as lubricant. The real-time analysis of API content, blend homogeneity, and tablet content uniformity was performed using a Partial Least Squares (PLS) quantitative method. The in-line Raman spectroscopic monitoring showed that the continuous blender was capable of producing blends with high homogeneity, and technological malfunctions can be detected by the proposed PAT method. The Raman spectroscopy-based feedback control of the API feeder was also established, creating a 'Process Analytically Controlled Technology' (PACT), which guarantees the required API content in the produced blend. This is, to the best of the authors' knowledge, the first ever application of Raman-spectroscopy in continuous blending and the first Raman-based feedback control in the formulation technology of solid pharmaceuticals.