News
Research and Innovation

HKUST Researchers Set to Transform Food Production with AI-enhanced 3D Food Printing Solution

(This article was originally published on EurekAlert! on March 31, 2025)

03 Good Health & Well-Being
09 Industry, Innovation and Infrastructure
12 Responsible Consumption and Production
The research team from the Division of Integrative Systems and Design (ISD) at HKUST comprised Prof. Mitch LI Guijun, ISD Assistant Professor (back row, first from left), and Connie LEE Kong-Wai (first row, right), the paper’s first author and a PhD student in ISD.

The research team from the Division of Integrative Systems and Design (ISD) at HKUST comprised Prof. Mitch LI Guijun, ISD Assistant Professor (back row, first from left), and Connie LEE Kong-Wai (first row, right), the paper’s first author and a PhD student in ISD.

The 3D food printer includes the real-time print and cook system.

The 3D food printer includes the real-time print and cook system.

The integrative 3D food printer and AI digital food design platform

The integrative 3D food printer and AI digital food design platform

Produced by the 3D food printer, the starch-based food items retain their intended shape and quality.

Produced by the 3D food printer, the starch-based food items retain their intended shape and quality.

a) Schematic diagram and scanning electron microscopy image of the LIG heating film. b) Optical microscope images of the LIG heating film. c) Raman spectra of the LIG heating film. d) Fourier transform infrared spectroscopy of the LIG heating film. e) Absorption spectrum of the printed dough sample.

a) Schematic diagram and scanning electron microscopy image of the LIG heating film. b) Optical microscope images of the LIG heating film. c) Raman spectra of the LIG heating film. d) Fourier transform infrared spectroscopy of the LIG heating film. e) Absorption spectrum of the printed dough sample.

a–d) SEM images at ×100 magnification of starch-based dough samples processed with air-frying, oven-baking, IR cooking, and laser cooking (from left to right). e–h) SEM images at ×2000 magnification of starch-based dough samples processed with air-frying, oven-baking, IR cooking, and laser cooking (from left to right). i–k) Reconstructed micro-computed tomography 2D cross-section images of the printed dough with different cooking methods. l) Porosity rate (%) of the printed dough with different cooking meth

a–d) SEM images at ×100 magnification of starch-based dough samples processed with air-frying, oven-baking, IR cooking, and laser cooking (from left to right). e–h) SEM images at ×2000 magnification of starch-based dough samples processed with air-frying, oven-baking, IR cooking, and laser cooking (from left to right). i–k) Reconstructed micro-computed tomography 2D cross-section images of the printed dough with different cooking methods. l) Porosity rate (%) of the printed dough with different cooking methods. m) Printability test using a cylindric shape of 30 × 30 × 30 mm as the testing model. n) Shape fidelity value (%) of untreated and IR-cooked samples in terms of length, width, and height. o) 3D printing process of the starch-based dough without heat treatment for 30 min. p) 3D printing process of the starch-based dough with LIG treatment for 30 min.

a–d) SEM images at ×100 magnification of starch-based dough samples processed with air-frying, oven-baking, IR cooking, and laser cooking (from left to right). e–h) SEM images at ×2000 magnification of starch-based dough samples processed with air-frying, oven-baking, IR cooking, and laser cooking (from left to right). i–k) Reconstructed micro-computed tomography 2D cross-section images of the printed dough with different cooking methods. l) Porosity rate (%) of the printed dough with different cooking meth

a) Food shape design process and workflow using AI diffusion models and Python scripts, compared to the conventional preparation method. b) Final 3D-printed food samples based on generated 3D files utilizing AI text prompts and Python script.

Researchers at the Hong Kong University of Science and Technology (HKUST) have developed a cutting-edge AI-assisted 3D food printing solution that combines printing with infrared cooking, paving the way for safer, more efficient, and visually appealing food production.

Traditional 3D food printing methods often require additional postprocessing steps, which can have unappealing food ingredients, imperfect shapes, and even potential microbial contamination.

To address these challenges, the team from the Division of Integrative Systems and Design (ISD) at HKUST has developed an AI-enhanced system that combines extrusion-based printing with simultaneous infrared heating for on-the-fly cooking of intricate starch-based foods. Using graphene heaters for cooking, they precisely controlled the cooking process, ensuring that starch-based food items retain their intended shape and quality.

The system is supported by AI-assisted design, which employs generative algorithms and Python scripts to craft intricate food patterns. By leveraging AI, the design process became accessible to even computer novices.

The research tackles issues such as shape retention and contamination risks and opens up exciting prospects for tailored nutrition—benefiting individuals with special dietary needs, including patients with dysphagia. From enhancing meal customization in elderly care centers and central kitchens to meeting special dietary needs and enabling creative culinary experiences in restaurants, the technology offers a versatile solution for various sectors.

“With its potential to streamline food production processes, improve food quality, and cater to individual preferences, this innovation can transform how food is prepared and served in diverse settings, paving the way for a future where personalized and visually appealing food creations are more accessible than ever before,” said Prof. Mitch LI Guijun, Assistant Professor of ISD, who led the research team.

“We’re excited about the potential of this technology to deliver customized, safe, and delicious food with a process that is both efficient and accessible. It represents a significant step forward in how we think about food creation,” Prof. Li added.

Connie LEE Kong-Wai, the paper’s first author and a PhD student at HKUST, said: “We’ve reimagined what 3D food printing can achieve by merging technology and culinary creativity. Our cutting-edge integrative 3D food printing technology can potentially revolutionize personalized food creation.”

The research represents a collaborative effort spanning user-centric design, mechanical engineering, food science, chemistry, and AI. This cross-disciplinary approach brought together diverse expertise to tackle the complex challenges associated with 3D food printing.

Looking ahead, the team plans to refine the technology by examining the preservation of heat-sensitive vitamins and optimizing starch digestibility. Future studies will also focus on consumer acceptance through sensory evaluations involving target users such as children or hospital caretakers, ensuring that the system is ready for real-world applications.

The study, “Advanced 3D Food Printing with Simultaneous Cooking and Generative AI Design”, was recently published in the top-tier international journal Advanced Materials.

Sign Up for Our Latest News

Stay informed with the latest updates, events, and breakthroughs from HKUST.