Puzzle Out Machine Learning Model-Explaining Disintegration Process in ODTs

Optimizing paracetamol-ascorbic acid effervescent tablet characteristics: a quality by design approach

OBJECTIVE

This study aims to optimize paracetamol-ascorbic acid (PCM-AA) effervescent tablet characteristics through a Quality-by-Design (QbD) approach, investigating the effects of binder concentration, granulation time, and effervescent agents' ratio on hardness, disintegration, and dissolution of the tablets.

METHODS

The QbD approach was implemented by identifying the quality target product profile, critical quality attributes (CQAs), critical material attributes (CMAs), and critical process parameters for formulating PCM-AA effervescent tablets. An Ishikawa diagram identified risk factors for CQAs. A risk estimation matrix evaluated the levels of associated risks. A central composite design-based response surface methodology with 20 experimental runs, including six center points, identified key factors (binder concentration, granulation time, and effervescent agents' ratio) influencing tablet characteristics (hardness, disintegration, dissolution). The optimum formulation, determined by numerical analysis, was characterized for weight uniformity, tablet thickness and diameter, friability, and PCM and AA assay.

RESULTS

Optimized PCM (500 mg)-AA(200 mg) effervescent tablets with 2.9% PVP concentration, 15 min granulation time, and 1:1.5 (w/w) sodium bicarbonate-citric acid ratio achieved acceptable characteristics (hardness: 45 N ± 20 N, disintegration: <5 min, and both PCM and AA dissolution: <10 min). Model validation showed no significant difference (p > 0.05), indicating consistent results.

CONCLUSION

The study successfully optimized the hardness, disintegration, and dissolution rate of PCM-AA effervescent tablets via the QbD approach. Granulation time affects hardness and PCM dissolution, binder concentration influences disintegration time, and the effervescent agents' ratio impacts both disintegration time and AA dissolution. This research enhances the understanding of pharmaceutical formulation processes, risk management, and optimization in effervescent tablet development.

Comparison of Deep Learning and Traditional Machine Learning Models for Predicting Mild Cognitive Impairment Using Plasma Proteomic Biomarkers

Mild cognitive impairment (MCI) is a clinical condition characterized by a decline in cognitive ability and progression of cognitive impairment. It is often considered a transitional stage between normal aging and Alzheimer's disease (AD). This study aimed to compare deep learning (DL) and traditional machine learning (ML) methods in predicting MCI using plasma proteomic biomarkers. A total of 239 adults were selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort along with a pool of 146 plasma proteomic biomarkers. We evaluated seven traditional ML models (support vector machines (SVMs), logistic regression (LR), naïve Bayes (NB), random forest (RF), k-nearest neighbor (KNN), gradient boosting machine (GBM), and extreme gradient boosting (XGBoost)) and six variations of a deep neural network (DNN) model-the DL model in the H2O package. Least Absolute Shrinkage and Selection Operator (LASSO) selected 35 proteomic biomarkers from the pool. Based on grid search, the DNN model with an activation function of "Rectifier With Dropout" with 2 layers and 32 of 35 selected proteomic biomarkers revealed the best model with the highest accuracy of 0.995 and an F1 Score of 0.996, while among seven traditional ML methods, XGBoost was the best with an accuracy of 0.986 and an F1 Score of 0.985. Several biomarkers were correlated with the APOE-ε4 genotype, polygenic hazard score (PHS), and three clinical cerebrospinal fluid biomarkers (Aβ42, tTau, and pTau). Bioinformatics analysis using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) revealed several molecular functions and pathways associated with the selected biomarkers, including cytokine-cytokine receptor interaction, cholesterol metabolism, and regulation of lipid localization. The results showed that the DL model may represent a promising tool in the prediction of MCI. These plasma proteomic biomarkers may help with early diagnosis, prognostic risk stratification, and early treatment interventions for individuals at risk for MCI.

Deterministic Models for Performance Analysis of Lignocellulosic Biomass Torrefaction

Energy plays a key role in the socioeconomic development of society, and most of its global demand is provided by conventional resources (e.g., fossil fuels). Utilizing renewable energy is significantly growing since it can meet global energy demand while minimizing the adverse impacts of carbon emissions on climate change. Biomass is an appealing option among the emerging alternatives (e.g., wind and solar). Torrefaction is a mild pyrolysis process, and this research aims to analyze the torrefaction process of lignocellulosic biomass. The methodology proposed involves employing hybrid models of artificial neural network-particle swarm optimization (ANN-PSO), adaptive neuro-fuzzy inference system (ANFIS), and coupled simulated annealing-least-squares support vector machine (CSA-LSSVM). In addition to the machine learning algorithms, a correlation is developed using gene expression programming (GEP) to interrelate the biomass properties, including moisture content, volatile matter, fixed carbon, ash, sample size, and the contents of oxygen, carbon, hydrogen, and nitrogen along with the process operating condition encompassing residence time, temperature, and the concentration of CO2, O2, and N2 to the solid yield as the target variable. The results reveal that the CSA-LSSVM model has the highest accuracy, and the statistical metrics of the coefficient of determination (R 2), mean square error (MSE), and average absolute relative error percentage (AARE%) are 0.98, 0.00082, and 2.61%, respectively. The parametric sensitivity analysis demonstrates the residence time, temperature, and moisture content as the most influential variables, with temperature playing the most crucial role in the torrefaction process of lignocellulosic biomass. The findings and the developed models can be used to assess similar biomass torrefaction, providing the required knowledge for the modeling and optimization of the process. Hence, the bioenergy industry can be developed with optimal operating conditions, including cost and energy, and lessen the negative impacts of CO2 emission.

Evaluation of the effect of carvedilol orodispersible tablets on ischemia-reperfusion injury and flap viability in rats: An in vivo study

Flap surgery is an integral part of plastic surgery, and ischemia-reperfusion (I/R) injury significantly affects the viability of the flap. Carvedilol (CRV), a nonselective beta-blocker with alpha-1 blocking and antioxidant properties, and known for its potential in reducing I/R damage, was chosen as the active substance for our study. The aim of this study was to investigate the vasodilator and antioxidant effects of CRV on rat inferior epigastric artery skin flap using orally disintegrating tablets (ODTs). The optimized ODT formulation was subjected to in vivo experiments using Sprague-Dawley female rats (n = 24) divided into three groups: Group I (control, I/R), Group II (treatment, I/R + CRV), and Group III (treatment, I/R), I/R + CRV ODT). Reperfusion was then observed following the release of the microclamp from the pedicle, and the flap was then re-adapted to its original position. Control rats were given oral isotonic solution via gavage and were subjected to 8 h of ischemia and 12 h of reperfusion. Group II was given 2 mg/kg CRV oral tablets for 7 days before and after surgery. Group III was given 2 mg/kg/day CRV ODT for the same period. Biopsies were taken from the flap and histopathological and biochemical analyses including superoxide dismutase, glutathionenitric oxide, malondialdehyde, paraoxonase 1, total oxidant, and total antioxidant capacities were performed. This study demonstrates that CRV ODTs significantly increased flap viability by approximately 25% compared to the control group, highlighting their promising therapeutic potential.

Diagnostic performance of machine learning in systemic infection following percutaneous nephrolithotomy and identification of associated risk factors

Objective

This study aims to investigate the predictive performance of machine learning in predicting the occurrence of systemic inflammatory response syndrome (SIRS) and urosepsis after percutaneous nephrolithotomy (PCNL).

Methods

A retrospective analysis was conducted on patients who underwent PCNL treatment between January 2016 and July 2022. Machine learning techniques were employed to establish and select the best predictive model for postoperative systemic infection. The feasibility of using relevant risk factors as predictive markers was explored through interpretability with Machine Learning.

Results

A total of 1067 PCNL patients were included in this study, with 111 (10.4 %) patients developing SIRS and 49 (4.5 %) patients developing urosepsis. In the validation set, the risk model based on the GBM protocol demonstrated a predictive power of 0.871 for SIRS and 0.854 for urosepsis. Preoperative and postoperative platelet changes were identified as the most significant predictors. Both thrombocytopenia and thrombocytosis were found to be risk factors for SIRS or urosepsis after PCNL. Furthermore, it was observed that when the change in platelet count before and after PCNL surgery exceeded 30*109/L (whether an increase or decrease), the risk of developing SIRS or urosepsis significantly increased.

Conclusion

Machine learning can be effectively utilized for predicting the occurrence of SIRS or urosepsis after PCNL. The changes in platelet count before and after PCNL surgery serve as important predictors.

Comparative pharmacokinetic evaluation of glimepiride orodispersable and conventional tablets in rabbits

OBJECTIVES

Glimepiride Orodispersable Tablets (ODT) were prepared with the goal to have rapid onset of action and higher bioavailability with ease administration to individuals with swallowing difficulty to ameliorate patient compliance.

SIGNIFICANCE

Glimepiride is a contemporary hypoglycemic medication that belongs to the family of sulfonylurea derivatives. It is used in type 2 diabetes mellitus. Compliance adherence remains one of the limitations with the conventional drug delivery system especially in pediatric, geriatric, psychiatric, and traveling patients, for such population ODT provides a good alternate dosage form compared with Commercial Tablets.

METHOD

The Comparative in vivo pharmacokinetic parameters of the prepared ODT and conventional tablets (CT) were evaluated using an animal model. The plasma concentration of Glimepiride after oral administration of a single dose was determined at predetermined time intervals with HPLC. The pharmacokinetic parameters were calculated using PK Solutions 2.0 from Summit PK® software.

RESULTS

The Cmax obtained with ODT (22.08 µg/ml) was significantly (p = 0.006) high, a lower tmax of 3.0 hr was achieved with the orodispersable formulation of the drug. The ODT showed 104.34% relative bioavailability as compared to CT and left shift of tmax as well.

CONCLUSION

As per findings of the in vivo investigation, the Glimepiride ODT would be beneficial in terms of patient compliance, quick onset of action, and increased bioavailability.

A Review on the Recent Advancements and Artificial Intelligence in Tablet Technology

BACKGROUND

Tablet formulation could be revolutionized by the integration of modern technology and established pharmaceutical sciences. The pharmaceutical sector can develop tablet formulations that are not only more efficient and stable but also patient-friendly by utilizing artificial intelligence (AI), machine learning (ML), and materials science.

OBJECTIVES

The primary objective of this review is to explore the advancements in tablet technology, focusing on the integration of modern technologies like artificial intelligence (AI), machine learning (ML), and materials science to enhance the efficiency, cost-effectiveness, and quality of tablet formulation processes.

METHODS

This review delves into the utilization of AI and ML techniques within pharmaceutical research and development. The review also discusses various ML methodologies employed, including artificial neural networks, an ensemble of regression trees, support vector machines, and multivariate data analysis techniques.

RESULTS

Recent studies showcased in this review demonstrate the feasibility and effectiveness of ML approaches in pharmaceutical research. The application of AI and ML in pharmaceutical research has shown promising results, offering a potential avenue for significant improvements in the product development process.

CONCLUSION

The integration of nanotechnology, AI, ML, and materials science with traditional pharmaceutical sciences presents a remarkable opportunity for enhancing tablet formulation processes. This review collectively underscores the transformative role that AI and ML can play in advancing pharmaceutical research and development, ultimately leading to more efficient, reliable and patient-centric tablet formulations.

Emerging Artificial Intelligence (AI) Technologies Used in the Development of Solid Dosage Forms

Artificial Intelligence (AI)-based formulation development is a promising approach for facilitating the drug product development process. AI is a versatile tool that contains multiple algorithms that can be applied in various circumstances. Solid dosage forms, represented by tablets, capsules, powder, granules, etc., are among the most widely used administration methods. During the product development process, multiple factors including critical material attributes (CMAs) and processing parameters can affect product properties, such as dissolution rates, physical and chemical stabilities, particle size distribution, and the aerosol performance of the dry powder. However, the conventional trial-and-error approach for product development is inefficient, laborious, and time-consuming. AI has been recently recognized as an emerging and cutting-edge tool for pharmaceutical formulation development which has gained much attention. This review provides the following insights: (1) a general introduction of AI in the pharmaceutical sciences and principal guidance from the regulatory agencies, (2) approaches to generating a database for solid dosage formulations, (3) insight on data preparation and processing, (4) a brief introduction to and comparisons of AI algorithms, and (5) information on applications and case studies of AI as applied to solid dosage forms. In addition, the powerful technique known as deep learning-based image analytics will be discussed along with its pharmaceutical applications. By applying emerging AI technology, scientists and researchers can better understand and predict the properties of drug formulations to facilitate more efficient drug product development processes.

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

NIR spectroscopy is a non-destructive characterization tool for the blend uniformity (BU) assessment. However, NIR spectra of powder blends often contain overlapping physical and chemical information of the samples. Deconvoluting the information related to chemical properties from that associated with the physical effects is one of the major objectives of this work. We achieve this aim in two ways. Firstly, we identified various sources of variability that might affect the BU results. Secondly, we leverage the machine learning-based sophisticated data analytics processes. To accomplish the aforementioned objectives, calibration samples of amlodipine as an active pharmaceutical ingredient (API) with the concentrations ranging between 67 and 133% w/w (dose ~ 3.6% w/w), in powder blends containing excipients, were prepared using a gravimetric approach and assessed using NIR spectroscopic analysis, followed by HPLC measurements. The bias in NIR results was investigated by employing data quality metrics (DQM) and bias-variance decomposition (BVD). To overcome the bias, the clustered regression (non-parametric and linear) was applied. We assessed the model's performance by employing the hold-out and k-fold internal cross-validation (CV). NIR-based blend homogeneity with low mean absolute error and an interval estimates of 0.674 (mean) ± 0.218 (standard deviation) w/w was established. Additionally, bootstrapping-based CV was leveraged as part of the NIR method lifecycle management that demonstrated the mean absolute error (MAE) of BU ± 3.5% w/w and BU ± 1.5% w/w for model generalizability and model transferability, respectively. A workflow integrating machine learning to NIR spectral analysis was established and implemented. Impact of various data learning approaches on NIR spectral data.