The paper showed how a machine learning (ML) model can predict the inkjet printability of pharmaceutical formulations with high accuracy.
A paper has been published showing that an optimized machine learning (ML) model predicted the printability of formulations with 97.22% accuracy using inkjet printing. International Journal of Pharmacy: X showed.
Inkjet printing takes time to optimize formulations and printing parameters. This is especially true for additive manufacturing and the manufacturing of unique dosage forms and personalized medicines. Advantages of the latter include low cost and versatility. For example, piezoelectric inkjet printing is one printing method used for personalized pharmaceuticals.
ML has been used to predict the printing results and dissolution behavior of Fused Deposition Modeling (FDM™) printed dosage forms and Digital Light Processing (DLP) printed tablets, the paper states.
This study evaluated how ML can analyze nuanced differences and provide more reliable predictions compared to traditional guidance on jettability based on printhead settings Z-value. bottom.
The researchers theorized that a predictive model of inkjet printing results could be developed. This study therefore aimed to develop and evaluate the performance of ML models for predicting the printability of inkjet printing and the total drug dose of the final printed dosage form.
The ML models used in the study to predict printability are:
- random forest
- multilayer perceptron
- support vector machine.
How well did the optimized machine learning model predict printability?
In addition to predicting the printability of formulations with high accuracy, the optimized ML model also predicted print quality with 97.14% accuracy. Current guidance states that only inks with Z values less than 10 are printable. By comparison, following this guidance yields an accuracy of 64.39 percent.
Benefits of machine learning and current guidelines for predicting printability
Formulation development and optimization is a time and resource intensive process that can be greatly accelerated with the guidance of predictive in silico tools.
Researchers believe that with predictive tools to better determine the printability of inks prior to actual formulation and testing, pharmaceutical researchers can find more unique agents to solve unmet clinical challenges. He said that he would be able to concentrate on designing the shape.
Analysis of a dataset consisting of 687 formulations revealed that positive print results were overwhelmingly more frequently published than negative results. Despite the imbalanced dataset, the ML model optimized for predicting printability performed significantly better than the traditional guidance.
Application of Inkjet Printing in Pharmaceuticals
For drug formulations, inkjet printing has been used to load drugs into orodispersible films, bioadhesive films for cervical administration, and transdermal microneedles, the authors noted. . Inkjet printing has also been used to dispense drug-laden microparticles and nanoparticles dispersed in an ink solution.
Inkjet printing can also be combined with other additive manufacturing techniques that are not feasible with traditional manufacturing techniques.
Collaboration in the production of 3D screen-printed pharmaceuticals
For example, one study featured in the paper combined inkjet printing with FDM™ 3D printing to produce drug-filled tablets with printed quick response (QR) codes. These QR codes encode patient-related information that can be read using a smartphone and are designed to act as an anti-counterfeiting strategy.
Another study also applied the fabrication of an orodispersible matrix and capsules with a printed QR code. Another case study presented in the paper uses inkjet printing to manufacture an entire 3D drug-filled tablet.
Carousenra other. The study concludes that machine learning models “demonstrate that they can feasibly provide predictive insight into the outcome of inkjet printing prior to formulation preparation, saving resources and time.”
