方永聪&杨华勇 | 人工智能如何重塑生物制造以加速临床转化

传统生物制造高度依赖专家经验与反复实验，存在周期长、成本高的瓶颈。而人工智能的融入正引发一场根本性的范式变革。通过将计算、自动化与生物学深度融合，AI催生出能够从海量数据中学习、预测复杂生物行为并自主优化生产流程的智能系统。它正成为连接实验室探索与临床精准治疗的关键桥梁，有力推动从实验室发现到临床应用的转化。

本期“AI赋能生物制造”专栏旨在汇聚前沿洞察，系统探讨人工智能如何深刻变革生物制造全链条——从材料、设计、制造到临床转化，加速其走向实际应用。目前已汇集来自清华大学、浙江大学、四川大学、东南大学及上海大学等多个顶尖团队的创新成果。

🔬 智能材料发现：从“试错作坊”到“AI计算实验室”

在材料研发领域，AI通过整合数据挖掘、机器学习与自动化实验，将传统“大海捞针”式的试错过程转变为精准的“理性设计”。在本期专栏中，清华大学熊卓团队提出了一种物理信息驱动AI框架，能够解码生物墨水与支撑浴间的复杂流变行为，推荐最优材料组合，从而实现高保真悬浮3D生物打印[1]。这项工作生动展示了AI如何架起计算与实验的桥梁，大幅加速功能生物材料的开发进程。

🧠 模型引导设计：从模仿自然到智能生成

在结构设计层面，AI能够根据临床需求生成性能最优的几何构型，使个性化、功能导向的复杂结构设计成为可能。在本期专栏中，四川大学周长春团队的案例研究展示了这一方向的临床潜力：他们利用AI辅助设计技术，为一名复杂骨包虫病患者一体化定制了替代髋关节、股骨及膝关节的3D打印假体，并取得良好治疗效果[2]。这标志着AI驱动的个性化医疗正逐步走向现实。

🦾 闭环自适应制造：赋予机器感知与决策能力

制造过程的智能化是保障结果可靠性的关键。AI通过集成实时视觉监控、过程模拟与自主决策，将生物制造从预编程执行升级为自适应调控的闭环系统。在本期专栏中，上海大学苏佳灿团队系统综述了AI如何赋能原位监控与工艺优化，引领生物打印进入5.0时代[3]。与此同时，浙江大学杨赓团队回顾了AI如何推动电子皮肤发展，实现与软体机器人间的多模态感知与智能交互[4]。这些进展共同预示生物制造设备将日益具备智能与自主特性。

🏥 临床转化：AI赋能质控与药物筛选

临床转化端，AI通过提升制造过程的预测性、标准化与可重复性，助力满足严格的监管与质控要求。例如，利用AI早期预测细胞分化结果或高效分析类器官模型，可大幅加速药物筛选与疗法评估。在本期专栏中，东南大学顾忠泽团队开发的POST精准分割算法，能够基于常规明场图像准确识别类器官并有效排除伪影干扰[5]，显著提高了高通量药物筛选的准确度与效率，是AI助力临床转化的典范。

💡 展望未来：拥抱挑战，共创智能生物制造新时代

尽管前景广阔，AI赋能生物制造仍面临生物数据标准化、生物系统复杂性、跨学科协同及监管适配等挑战。未来，通过持续的研究与合作，AI将不仅提升生物制造系统的智能化水平，更将推动其成果实现安全性、可及性与个性化，最终释放再生医学的全面潜力。

1. 人工智能驱动的悬浮打印流变学理解与可打印性预测

Zhou XH, Zhang ZR, Yu JT, et al., 2026. AI-driven interpretation and prediction of embedded printability based on rheology. Bio-Des Manuf 9(1):63–79. https://doi.org/10.1631/bdm.2500456

2. 人工智能辅助设计3D打印定制式假体用于包虫病引起的严重髋关节、股骨、膝关节一体置换

Han YL, Lei HY, Jia RZ, et al., 2026. AI-assisted design of 3D-printed prosthesis for integrated replacement of the hip, femur, and knee caused by osseous hydatidosis. Bio-Des Manuf 9(1):94–99. https://doi.org/10.1631/bdm.2500162

3. 人工智能技术赋能生物打印5.0

Bai L, Zhang Y, Wang SC, et al., 2026. Artificial intelligence-enabled Bioprinting 5.0. Bio-Des Manuf 9(1):32-62. https://doi.org/10.1631/bdm.2400354

4. 人工智能赋能的电子皮肤: 从智能设计到认知交互

Wang BC, Kong DP, Yang HY, et al., 2026. AI-powered electronic skin: from smart design to cognitive interaction. Bio-Des Manuf 9(1):4–31. https://doi.org/10.1631/bdm.2500376

5. 精准类器官分割技术 (POST): 针对挑战性明场图像的高精度类器官分割

Du X, Li YC, Song JP, et al., 2026. Precision organoid segmentation technique (POST): accurate organoid segmentation in challenging bright-field images. Bio-Des Manuf 9(1):80–93. https://doi.org/10.1631/bdm.2500119

上下滑动以阅览

1. Gil D, Moler KA (2025) Accelerating science with AI. Science 390(6777):965. https://doi.org/10.1126/science.aee0605

2. Wu C, Entezari A, Zheng K et al (2021) A machine learningbased multiscale model to predict bone formation in scaffolds. Nat Comput Sci 1:532–541. https://doi.org/10.1038/s43588-021-00115-x

3. Yuan XM, Kong WC, Shao YJ et al (2025) AI-feedback bioelectronics promote infectious wound healing. Cell Biomater (early access). https://doi.org/10.1016/j.celbio.2025.100191

4. Burger B, Maffettone PM, Gusev VV et al (2020) A mobile robotic chemist. Nature 583:237–241. https://doi.org/10.1038/s41586-020-2442-2

5. Steiner S, Wolf J, Glatzel S et al (2019) Organic synthesis in a modular robotic system driven by a chemical programming language. Science 363(6423):eaav2211. https://doi.org/10.1126/science.aav2211

6. Szymanski NJ, Rendy B, Fei Y et al (2023) An autonomous laboratory for the accelerated synthesis of novel materials. Nature 624:86–91. https://doi.org/10.1038/s41586-023-06734-w

7. Liao HG, Hu S, Yang H et al (2025) Data-driven de novo design of super-adhesive hydrogels. Nature 644:89–95. https://doi.org/10.1038/s41586-025-09269-4

8. Zhou XH, Zhang ZR, Yu JT et al (2026) AI-driven interpretation and prediction of embedded printability based on rheology. Bio-Des Manuf 9(1):63–79. https://doi.org/10.1631/bdm.2500456

9. Sexton ZA, Rutsche D, Herrmann JE et al (2025) Rapid model-guided design of organ-scale synthetic vasculature for biomanufacturing. Science 388(6752):1198–1204. https://doi.org/10.1126/science.adj6152

10. Peng B, Wei Y, Qin Y et al (2023) Machine learning-enabled constrained multi-objective design of architected materials. Nat Commun 14:6630. https://doi.org/10.1038/s41467-023-42415-y

11. Han YL, Lei HY, Jia RZ et al (2026) AI-assisted design of 3D-printed prosthesis for integrated replacement of the hip, femur, and knee caused by osseous hydatidosis. Bio-Des Manuf 9(1): 94–99. https://doi.org/10.1631/bdm.2500162

12. Mo XW, Zhang YM, Wang ZX et al (2024) Satellite-based on-orbit printing of 3D tumor models. Adv Mater 36(34):2309618. https://doi.org/10.1002/adma.202309618

13. Buchner TJK, Rogler S, Weirich S et al (2023) Vision-controlled jetting for composite systems and robots. Nature 623: 522–530. https://doi.org/10.1038/s41586-023-06684-3

14. Florczak S, Größbacher G, Ribezzi D et al (2025) Adaptive and context-aware volumetric printing. Nature 645:108–114. https://doi.org/10.1038/s41586-025-09436-7

15. Bai L, Zhang Y, Wang SC et al (2026) Artificial intelligence-enabled Bioprinting 5.0. Bio-Des Manuf 9(1):32–62. https://doi.org/10.1631/bdm.2400354

16. Wang BC, Kong DP, Yang HY et al (2026) AI-powered electronic skin: from smart design to cognitive interaction. Bio-Des Manuf 9(1):4–31. https://doi.org/10.1631/bdm.2500376

17. Zhu YJ, Huang RQ, Wu ZR et al (2021) Deep learning-based predictive identification of neural stem cell differentiation. Nat Commun 12:2614. https://doi.org/10.1038/s41467-021-22758-0

18. Diosdi A, Toth T, Harmati M et al (2025) HCS-3DX, a nextgeneration AI-driven automated 3D-oid high-content screening system. Nat Commun 16:8897. https://doi.org/10.1038/s41467-025-63955-5

19. Du X, Li YC, Song JP et al (2026) Precision organoid segmentation technique (POST): accurate organoid segmentation in challenging bright-field images. Bio-Des Manuf 9(1):80–93. https://doi.org/10.1631/bdm.2500119

20. Tebon PJ, Wang B, Markowitz AL et al (2023) Drug screening at single-organoid resolution via bioprinting and interferometry. Nat Commun 14:3168. https://doi.org/10.1038/s41467-023-38832-8

21. Rivenson Y, Wang H, Wei Z et al (2019) Virtual histological staining of unlabelled tissue-autofluorescence images via deep learning. Nat Biomed Eng 3:466–477. https://doi.org/10.1038/s41551-019-0362-y

客座编辑

BDM期刊主编