A prediction model based on artificial intelligence techniques for disintegration time and hardness of fast disintegrating tablets in pre-formulation tests.

BACKGROUND: The pharmaceutical industry is continually striving to innovate drug development and formulation processes. Orally disintegrating tablets (ODTs) have gained popularity due to their quick release and patient-friendly characteristics. The choice of excipients in tablet formulations plays a critical role in ensuring product quality, highlighting its importance in tablet creation. The traditional trial-and-error approach to this process is both expensive and time-intensive. To tackle these obstacles, we introduce a fresh approach leveraging machine learning and deep learning methods to automate and enhance pre-formulation drug design. METHODS: We collected a comprehensive dataset of 1983 formulations, including excipient names, quantities, active ingredient details, and various physicochemical attributes. Our study focused on predicting two critical control test parameters: tablet disintegration time and hardness. We compared a range of models like deep learning, artificial neural networks, support vector machines, decision trees, multiple linear regression, and random forests. RESULTS: A 12-layer deep neural network, as a form of deep learning, surpassed alternative techniques by achieving 73% accuracy for disintegration time and 99% for tablet hardness. This success underscores its efficacy in predicting complex pharmaceutical factors. Such an approach streamlines the drug formulation process, reducing iterations and material consumption. CONCLUSIONS: Our findings highlight the deep learning potential in pharmaceutical formulations, particularly for tablet hardness prediction. Future work should focus on enlarging the dataset to improve model effectiveness and extend its application in pharmaceutical product development and assessment.

Application of the Non-dominated Sorting Genetic Algorithm II in Multi-objective Optimization of Orally Disintegrating Tablet Formulation.

In the context of increasing application of modelling methods in the field of pharmaceutics, this study aims to reduce the weight of sildenafil orally disintegrating tablets (ODTs) and optimize their formulation through modelling methods. To achieve the goal, the back-propagation neural network (BPNN)-based non-dominated sorting genetic algorithm II (NSGA-II) was introduced to establish the models and to optimize the percentage of magnesium stearate (MgSt), crospovidone (PVPP), and croscarmellose sodium (CCNa) to obtain satisfactory candidate ODTs. Ultimately, the bioequivalence trial was conducted to verify the effectiveness of the formulation. With the support of the neural network, the model showed satisfactory results in the prediction of hardness and disintegration time of ODTs, and the pareto front obtained by the NSGA-II suggested that there was a strong "competition" between disintegration time and hardness. Since disintegration time should be given the priority, the optimal formulation was determined as 1% MgSt, 6% CCNa, and 2.6% PVPP. The bioequivalence trial results indicated a bioequivalence between the test and the reference formulations of sildenafil, and better medication experience for the test formulation. A bioequivalent formulation with better medication experience is successfully prepared using the NSGA-II. It proves that the NSGA-II is applicable to multi-objective optimization of the drug formulation.

Development and optimization of levodopa and benzylhydrazine orally disintegrating tablets by direct compression and response surface methodology.

The number of Parkinson's disease (PD) patients with the advanced phase and motor fluctuations is increasing. The objective of this study is developing levodopa/benzylhydrazine orally disintegrating tablets (L/B ODTs), which would provide greater convenience and ease of use than conventional tablets for these patients. In the present study, the L/B ODTs were developed successfully with an optimized formulation using response surface methodology (RSM). The direct compression technology was employed for the preparation of L/B ODTs. Considerably shorter disintegration time and faster dissolution profile were obtained under the optimum formulation with microcrystalline cellulose 25.7%, cross-polyvinylpyrrolidone 6.22% and Sodium carboxymethyl starch 5.36%. The content uniformity (%) of levodopa and benzylhydrazine was 50 ± 1.4% and 14.25 ± 0.6%, respectively. Thickness, friability, hardness and wetting time were 2.8 ± 0.05 mm, 0.46 ± 0.21%, 5.42 ± 1.1 kp and 31.2 ± 2.1 s, respectively, and all of data well comply with the General Principles of the Chinese Pharmacopeia. Mannitol of 22% in formulation could bring a pleasant taste: sweet, cool and refreshing. Almost all the volunteers felt that the ODTs had good taste, no roughness, and no gritty feeling, indicating that the ODTs prepared had good palatability, so patients will have good compliance when taking medicine.

Optimized furosemide taste masked orally disintegrating tablets.

Optimized orally disintegrating tablets (ODTs) containing furosemide (FUR) were prepared by direct compression method. Two factors, three levels (32) full factorial design was used to optimize the effect of taste masking agent (Eudragit E100; X1) and superdisintegarant; croscarmellose sodium (CCS; X2) on tablet properties. A composite was prepared by mixing ethanolic solution of FUR and Eudragit E100 with mannitol prior to mixing with other tablet ingredients. The prepared ODTs were characterized for their FUR content, hardness, friability and wetting time. The optimized ODT formulation (F1) was evaluated in term of palatability parameters and the in vivo disintegration. The manufactured ODTs were complying with the pharmacopeia guidelines regarding hardness, friability, weight variation and content. Eudragit E100 had a very slightly enhancing effect on tablets disintegration. However, the effects of both Eudragit E100 (X1) and CCS (X2) on ODTs disintegration time (Y1) were insignificant (p > 0.05). Moreover, X1 exhibited antagonistic effect on the dissolution after 5 and 30 min (D5 and D30, respectively), but only its effect on D30 is significant (p = 0.0004). Furthermore, the optimized ODTs formula showed good to acceptable taste in term of palatability, and in vivo disintegration time of this formula was about 10 s.

Selective Laser Sintering 3D Printing of Orally Disintegrating Printlets Containing Ondansetron.

The aim of this work was to explore the feasibility of using selective laser sintering (SLS) 3D printing (3DP) to fabricate orodispersable printlets (ODPs) containing ondansetron. Ondansetron was first incorporated into drug-cyclodextrin complexes and then combined with the filler mannitol. Two 3D printed formulations with different levels of mannitol were prepared and tested, and a commercial ondansetron orally disintegrating tablet (ODT) product (Vonau® Flash) was also investigated for comparison. Both 3D printed formulations disintegrated at ~15 s and released more than 90% of the drug within 5 min independent of the mannitol content; these results were comparable to those obtained with the commercial product. This work demonstrates the potential of SLS 3DP to fabricate orodispersible printlets with characteristics similar to a commercial ODT, but with the added benefit of using a manufacturing technology able to prepare medicines individualized to the patient.