AI improves tissue printing
Print
Machine learning is being used to speed up a new style of medical 3D-printing.
Biomedical engineers at Queensland University of Technology (QUT) have created an automated system that transforms melt electrowriting (MEW), a precise 3D printing technology.
By cutting down on trial-and-error experimentation and ensuring better results, the system promises to make high-resolution 3D printing faster, easier, and more accessible to industries worldwide.
Researcher Dr Pawel Mieszczanek says the new system solves some of MEW’s biggest challenges.
“MEW is a multifaceted 3D printing technology that also has applications in bioengineering, biomaterials science, and soft robotics,” he said.
“However, it has faced many challenges from its early stages more than 10 years ago to its current stage, hampered by long experimentation times, low printing speeds, poor consistency in results, and dependence on the user for printer operation.
“To address these problems, we used machine learning (ML) to create a closed-loop process control system for MEW.”
The research team used machine learning (ML) to build a system that monitors and adjusts the MEW process in real time.
This ensures higher precision and more reliable results.
“The novel MEW system design is effective because it monitors the fibre-flight pass, allowing us to use real-time imaging for continuous analysis,” Dr Mieszczanek explained.
The automated system is a potential game-changer for MEW.
The advanced automation cuts experimental time from weeks to hours.
“We use a feedforward neural network, optimisation techniques, and a feedback loop to ensure that printed parts are consistently reproducible,” said Distinguished Professor Dietmar W Hutmacher, who led the project.
Their system measures key details, such as the thickness of the fibres and the angle of the material as it flows during printing.
Advanced camera systems and algorithms process these details quickly, allowing the system to make adjustments on the fly.
Unlike traditional MEW, which relies on trial and error, this new system collects and analyses large amounts of data in real time. The closed-loop control ensures the final product matches exact specifications, improving both speed and quality.
MEW is known for its ability to create delicate, detailed structures needed in areas like tissue engineering and soft robotics. However, its complexity has limited its use outside research settings.
By simplifying operations and improving reliability, the new system could bring MEW into industries such as healthcare and manufacturing.
The researchers showed that the machine learning-based system achieved precise control over fibre thickness, jet angle, and material shape during printing. This means high-quality results can be produced even by less experienced users.
The study also demonstrated that the system can maintain accuracy to within 5 per cent of target specifications, a significant improvement over traditional methods.
The research team believes their work is an important step in making MEW ready for industrial use.
The next goals include real-time quality control and faster testing of new materials. The project also suggests the system could support other types of 3D printing technologies in the future.
More details are accessible here.