Related Experiment Video
Updated: Jan 9, 2026
Viability of Bioprinted Cellular Constructs Using a Three Dispenser Cartesian Printer
Published on: September 22, 2015
AI-powered printability evaluation framework for 3D bioprinting using Hausdorff distance metrics
Colin Zhang1, Kelum Chamara Manoj Lakmal Elvitigala1, Shinji Sakai1
1Division of Chemical Engineering, Department of Materials Engineering Science, Graduate School of Engineering Science, The University of Osaka, 1-3 Machikaneyama-cho, Toyonaka, Osaka 560-8531, Japan.
This study introduces a novel Hausdorff distance (HD) metric and AI models to optimize 3D bioprinting bioinks. The approach significantly improves the prediction of printability and cell viability, reducing optimization time for tissue engineering applications.
Area of Science:
- Biotechnology
- Materials Science
- Tissue Engineering
Background:
- 3D bioprinting is crucial for fabricating complex biological structures.
- Optimizing bioink formulation is challenging due to intricate material property, printability, and cell viability relationships.
- Existing printability metrics like perimeter ratio (Pr) lack comprehensive geometric fidelity assessment.
Purpose of the Study:
- To develop a novel Hausdorff distance (HD) metric for accurate printability quantification in 3D bioprinting.
- To apply machine learning (ML) models for predicting bioink printability and cell viability.
- To optimize alginate-hyaluronic acid composite inks for enhanced tissue engineering applications.
Main Methods:
- Utilized Hausdorff distance (HD) to measure geometric deviation between designed and printed structures.
- Employed machine learning algorithms including Support Vector Regression (SVR), Multi-layer Perceptron (MLP), and Convolutional Neural Networks (CNN).
- Assessed printability and PC12 cell viability in alginate-hyaluronic acid composite inks.
Main Results:
- SVR accurately characterized rheological parameters (R² ≥ 0.974).
- MLP models predicted HD values (R² = 0.932) and cell viability (R² = 0.945).
- CNN achieved high accuracy (R² = 0.986) in predicting HD from images, enabling optimal bioink formulation with ≥95% cell viability and HD ≤ 0.20.
Conclusions:
- The AI-integrated approach significantly reduces bioink optimization time.
- The novel HD metric provides a more comprehensive assessment of printability.
- Optimized bioinks demonstrate high printability and excellent long-term cell viability and proliferation potential for tissue engineering.
