A 3D printer for pills will watch its own work, layer by layer, and correct mistakes as they happen. Personalised medicines—pills tailored to a patient’s exact dose, shape, and drug-release profile—could transform treatment for conditions that require precise dosing, such as cancer or heart disease. But 3D-printed pharmaceuticals are not yet reliable enough for routine use. Printing defects such as warped layers or uneven material deposition can ruin a batch, and current monitoring systems cannot catch these problems in real time. This project aims to close that gap by building a multi-sensor system that captures high-precision data from each printed layer, then uses machine learning to detect and classify defects instantly. A digital twin—a virtual replica of the printing process—would then adjust the printer’s actions to keep the build on track. If successful, the system could make pharmaceutical 3D printing robust enough for pharmacies and hospitals to produce custom pills on demand. That would shift manufacturing from centralised factories to the point of care, reducing waste and enabling truly individualised treatment. The work is applied engineering, not fundamental science, but its core challenge—real-time sensing and control of a complex process—has parallels in other precision manufacturing fields.
View original technical description
Advances in the field of pharmacogenomics have provided an impetus for the development and fabrication of personalised pharmaceutical formulations. For instance, three-dimensional (3D) printing technologies are set to revolutionise the individualisation of dosage forms at the point of dispensing or use. The main benefit of using 3D printing technology is the ability to produce small batches with carefully tailored dosages, shapes, sizes, and drug release characteristics as per physiological or therapeutic needs of patients. It also allows flavours to be incorporated into a pill without the need of a film coating, entirely masking the taste of chemical compounds. Despite continuous enhancements of 3D printing systems, the lack of process repeatability and stability still represents a serious barrier to industrial breakthroughs. Consequently, various disruptions (e.g., improper deposition of compound material, layer warping, weak infill) occur during and propagate through layer building, exerting detrimental impacts on the building process and the quality of personalised pharmaceutical products. In-situ monitoring techniques have been identified as vital factors for robust control of printing processes. However, their major limitations are: (1) measurement sensors are insufficient in meeting the requirements of the in-situ measurement of 3D printing layers; (2) data analysis methods are not robust and/or accurate enough to extract all existing printing anomalies; and (3) lack of advanced closed-loop control that could take the monitoring information to handle layer building disruptions. The main goal of this project is to create an intelligent in-situ monitoring system to enable resilient pharmaceutical 3D printing. The specific research objectives are (1) construction of in-situ multi-sensors that are capable of capturing layer-build information with high precision; (2) investigation of machine learning based data analytics that can detect, classify and quantify critical printing anomalies quickly and accurately; and (3) exploration of a digital twin based control framework that enables the precise control of the deposition of pharmaceutical compound materials.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know