Journal of Zhejiang University

Bio-Design and Manufacturing 2026 Vol.9 No.1 P.63 - 79

AI-driven interpretation and prediction of embedded printability based on rheology

Author(s): Xianhao Zhou,Zhenrui Zhang,Jintian Yu,Lixi Ma,Sicheng Ma,Bingyan Wu,Zixuan Wang,Ting Zhang,Yongcong Fang,Zhuo Xiong
Affiliation(s): 1. Biomanufacturing Center, Department of Mechanical Engineering, Tsinghua University, Beijing, 100084, China more
Corresponding email(s): fangyc@tsinghua.edu.cn, fangyc@tsinghua.edu.cn
Key Words: Embedded bioprinting, Printability, Rheology, Machine learning, Neural network

Xianhao Zhou, Zhenrui Zhang, Jintian Yu, Lixi Ma, Sicheng Ma, Bingyan Wu, Zixuan Wang, Ting Zhang, Yongcong Fang, Zhuo Xiong. AI-driven interpretation and prediction of embedded printability based on rheology[J]. Journal of Zhejiang University Science D, 2026, 9(1): 63 - 79.

@article{title="AI-driven interpretation and prediction of embedded printability based on rheology",
author="Xianhao Zhou, Zhenrui Zhang, Jintian Yu, Lixi Ma, Sicheng Ma, Bingyan Wu, Zixuan Wang, Ting Zhang, Yongcong Fang, Zhuo Xiong",
journal="Journal of Zhejiang University Science D",
volume="9",
number="1",
pages="63 - 79",
year="2026",
publisher="Zhejiang University Press & Springer",
doi="10.1631/bdm.2500456"
}

%0 Journal Article
%T AI-driven interpretation and prediction of embedded printability based on rheology
%A Xianhao Zhou
%A Zhenrui Zhang
%A Jintian Yu
%A Lixi Ma
%A Sicheng Ma
%A Bingyan Wu
%A Zixuan Wang
%A Ting Zhang
%A Yongcong Fang
%A Zhuo Xiong
%J Journal of Zhejiang University SCIENCE D
%V 9
%N 1
%P 63 - 79
%@ 1869-1951
%D 2026
%I Zhejiang University Press & Springer
%DOI 10.1631/bdm.2500456

TY - JOUR
T1 - AI-driven interpretation and prediction of embedded printability based on rheology
A1 - Xianhao Zhou
A1 - Zhenrui Zhang
A1 - Jintian Yu
A1 - Lixi Ma
A1 - Sicheng Ma
A1 - Bingyan Wu
A1 - Zixuan Wang
A1 - Ting Zhang
A1 - Yongcong Fang
A1 - Zhuo Xiong
J0 - Journal of Zhejiang University Science D
VL - 9
IS - 1
SP - 63
EP - 79
%@ 1869-1951
Y1 - 2026
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/bdm.2500456

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: Embedded printing is a highly promising approach for creating complex structures within a yield-stress support bath. However, the accurate prediction and control of printability remain fundamental challenges due to the complex interactions between inks and support baths. Here, we present an artificial intelligence (AI)-driven framework that interprets and predicts embedded printability using rheological data. Using a standardized workflow, we extracted 21 rheological descriptors and established 12 indicators to evaluate structural continuity and geometric fidelity. Interpretable machine learning models revealed that direction-dependent defects are governed by the synergistic interplay among ink yield stress, support bath zero-shear viscosity, flow behavior index, and time constant. To enable the prediction of printability in a generalizable manner, we further developed a cascaded neural network, which achieved mean relative prediction errors below 15% across all indicators. Experimental validation using three-dimensional (3D)-printed constructs and micro-computed tomography (μCT) reconstructions confirmed a strong correlation between predicted and actual fidelity. This work establishes a physics-informed, data-driven paradigm for decoding and optimizing embedded printing, offering broad applicability and providing a robust tool for the rapid pairing of suitable printable ink-support bath combinations.

AI-driven interpretation and prediction of embedded printability based on rheology

悬浮打印是一种利用屈服应力支撑浴中构建复杂三维结构的先进方法, 具有广阔的应用前景。然而, 由于打印墨水与支撑浴之间存在复杂的相互作用, 准确预测与控制其可打印性仍面临巨大挑战。为此, 本文提出一种基于人工智能的框架, 通过流变学数据实现悬浮打印材料体系可打印性的解释与预测。研究中采用标准化工作流程提取了21项流变学特征参数, 并建立了12项评价指标, 用于量化打印结构的连续性与几何保真度。可解释机器学习模型分析表明, 方向依赖性缺陷受墨水屈服应力、支撑浴零剪切粘度、流动行为指数及时间常数等多参数协同调控。为构建普适化的可打印性预测模型, 我们进一步开发了LSTM-MLP级联神经网络, 该模型在所有评估指标上的平均相对预测误差均低于15%。此外, 通过三维打印实验结合微计算机断层扫描 (μCT) 重建的进一步验证, 证实了预测保真度与实际保真度具有强相关性。本研究建立了一种物理信息驱动的数据范式, 可用于解码与优化悬浮打印工艺, 该方法具备良好的普适性, 为快速匹配适宜的墨水–支撑浴组合提供了可靠工具。
Embedded bioprinting; Printability; Rheology; Machine learning; Neural network