基于机器学习的骨科生物材料设计与优化

doi:10.6040/j.issn.1671-7554.0.2024.1274

摘要/Abstract

摘要： 随着人工智能的发展,机器学习在骨科生物材料领域的应用日益增多,具有极大潜力。本文首先介绍机器学习的基本类型、针对不同目标场景的算法选取及其评价指标;其次分析不同骨科生物材料在设计过程中的关键化学和物理参数,及其相应的机器学习训练数据集;随后详细探讨机器学习在金属生物材料、生物陶瓷材料、高分子生物材料以及生物3D打印新材料中的具体应用,通过案例展示机器学习在预测材料性能、优化制造工艺、研究生物相容性等方面的优势。骨科生物材料正在向多学科交叉融合以及智能化方向发展,机器学习作为这一发展趋势中的关键技术,将更加高效地推动材料开发及临床应用。最后,本文分析当前阻碍机器学习进一步应用在临床中的瓶颈问题,并展望机器学习在骨科生物材料优化设计领域的广阔前景。总之,机器学习为骨科生物材料的设计和优化提供新的思路和方法。

关键词: 机器学习, 骨科生物材料, 材料设计, 性能优化, 智能材料

Abstract: With the development of artificial intelligence, the application of machine learning in the field of orthopedic biomaterials is also increasing, which has great potential. This paper first introduced the basic types of machine lear-ning, algorithm selection for different target scenarios and its evaluation index. Secondly, the key chemical and physical parameters in the design process of different orthopedic biomaterials, and the training data set of machine learning were analyzed. Then, the specific applications of machine learning in metal biological materials, bioceramics implant materials, polymer biological materials and new materials for bioprinting were discussed in detail, and the advantages of machine learning in predicting material properties, optimizing manufacturing processes, and studying biocompatibility were demonstrated through cases. Orthopedic biomaterials are developing in the direction of multidisciplinary integration and intelligence, and machine learning, as a key technology in this development trend, will more efficiently promote material development and clinical application. Finally, this paper analyzed the crucial impediments that hinder the further application of machine learning in clinical research, and looksed forward to the broad prospect of machine learning in the field of orthopedic biomaterials optimization design. In conclusion, machine learning provides new ideas and methods for the design and optimization of orthopedic biomaterials.

Key words: Machine learning, Orthopedic biomaterials, Material design, Performance optimization, Smart materials

中图分类号:

R318.08

刘禹,霍娅娅,龚丞,梁婷,李斌. 基于机器学习的骨科生物材料设计与优化[J]. 山东大学学报 (医学版), 2026, 64(2): 22-33.

LIU Yu, HUO Yaya, GONG Cheng, LIANG Ting, LI Bin. Design and optimization of orthopedic biomaterials based on machine learning[J]. Journal of Shandong University (Health Sciences), 2026, 64(2): 22-33.

参考文献

[1] Guo JL, Januszyk M, Longaker MT. Machine learning in tissue engineering[J]. Tissue Eng Part A, 2023, 29(1/2): 2-19.
[2] Merz KM, Choong YS, Cournia Z, et al. Editorial: machine learning in materials science[J]. J Chem Inf Model, 2024, 64(10): 3959-3960.
[3] Suwardi A, Wang F, Xue K, et al. Machine learning-driven biomaterials evolution[J]. Adv Mater, 2022, 34(1): e2102703. doi: 10.1002/adma.202102703
[4] Xue K, Wang FK, Suwardi A, et al. Biomaterials by design: harnessing data for future development[J]. Mater Today Bio, 2021, 12: 100165. doi: 10.1016/j.mtbio.2021.100165
[5] Hughes MC, Pradier MF, Ross AS, et al. Assessment of a prediction model for antidepressant treatment stability using supervised topic models[J]. JAMA Netw Open, 2020, 3(5): e205308. doi: 10.1001/jamanetworkopen.2020.5308
[6] XU ZL, KING I. Introduction to semi-supervised learning[M]. Boca Raton: CRC Press, 2014, 250.
[7] Guo YY, Rui SS, Xu W, et al. Machine learning method for fatigue strength prediction of nickel-based superalloy with various influencing factors[J]. Materials, 2022, 16(1): 46. doi: 10.3390/ma16010046
[8] van Engelen JE, Hoos HH. A survey on semi-supervised learning[J]. Mach Learn, 2020, 109(2): 373-440.
[9] 武丽, 丁琴, 葛彩成. 融合公平因子的半监督学习医学图像分割模型[J]. 电子测量技术, 2024, 47(23): 171-180. WU Li, DING Qin, GE Caicheng. Semi-supervised lear-ning medical image segmentation model fused with equity factors[J]. Electronic Measurement Technology, 2024, 47(23): 171-180.
[10] Markovi c G, Manojlovi c V, Ru zi c J, et al. Predicting low-modulus biocompatible titanium alloys using machine learning[J]. Materials, 2023, 16(19): 6355. doi: 10.3390/ma16196355
[11] Noguchi S, Nishio M, Sakamoto R, et al. Deep learning-based algorithm improved radiologists performance in bone metastases detection on CT[J]. Eur Radiol, 2022, 32(11): 7976-7987.
[12] Rosenblum D, Joshi N, Tao W, et al. Progress and challenges towards targeted delivery of cancer therapeutics[J]. Nat Commun, 2018, 9(1): 1410. doi: 10.1038/s41467-018-03705-y
[13] Yang DP, Oo MNNL, Deen GR, et al. Nano-star-shaped polymers for drug delivery applications[J]. Macromol Rapid Commun, 2017, 38(21): 1700410. doi:10.1002/marc.201700410
[14] Koons GL, Diba M, Mikos AG. Materials design for bone-tissue engineering[J]. Nat Rev Mater, 2020, 5: 584-603. doi: 10.1038/s41578-020-0204-2
[15] Liu Q, Zheng S, Ye K, et al. Cell migration regulated by RGD nanospacing and enhanced under moderate cell adhesion on biomaterials[J]. Biomaterials, 2020, 263: 120327. doi: 10.1016/j.biomaterials.2020.120327
[16] Veiseh O, Doloff JC, Ma ML, et al. Size- and shape-dependent foreign body immune response to materials implanted in rodents and non-human Primates[J]. Nat Mater, 2015, 14(6): 643-651.
[17] Yeo T, Tan SJ, Lim CL, et al. Microfluidic enrichment for the single cell analysis of circulating tumor cells[J]. Sci Rep, 2016, 6: 22076. doi:10.1038/srep22076
[18] Charrier EE, Pogoda K, Wells RG, et al. Control of cell morphology and differentiation by substrates with independently tunable elasticity and viscous dissipation[J]. Nat Commun, 2018, 9(1): 449. doi: 10.1038/s41467-018-02906-9
[19] 闵子洋, 穆妮热·艾力, 郑耘昊, 等. 细胞外基质力学微环境与细胞间相互作用的机制与特征[J]. 中国组织工程研究, 2022, 26(25): 4034-4045. MIN Ziyang, Munire·Aili, ZHENG Yunhao, et al. Mechanism and characteristics of mechanical microenvironment of extracellular matrix and intercellular interaction[J]. Chinese Journal of Tissue Engineering Research, 2022, 26(25): 4034-4045.
[20] 欧阳思慧, 佘加, 陈先华, 等. 可降解镁基复合材料的制备及其在骨科领域的研究进展[J]. 金属学报, 2025, 61(3): 455-474. OUYANG Sihui, SHE Jia, CHEN Xianhua, et al. Pre-paration of biodegradable mg-based composites and their recent advances in orthopedic applications[J]. Acta Metallurgica Sinica, 2025, 61(3): 455-474.
[21] 马静, 苏秀云, 唐斌, 等. 骨修复中可吸收材料降解行为的研究进展[J]. 中国生物医学工程学报, 2023, 42(5): 626-635. MA Jing, SU Xiuyun, TANG Bin, et al. Research progress of degradable behavior of absorbable biomaterials in bone repair[J]. Chinese Journal of Biomedical Engineering, 2023, 42(5): 626-635.
[22] Li L, Yu F, Zheng LM, et al. Natural hydrogels for cartilage regeneration: modification, preparation and application[J]. J Orthop Transl, 2019, 17: 26-41. doi: 10.1016/j.jot.2018.09.003
[23] 潘雨崧, 彭旭龙, 陈子光. 金属材料腐蚀疲劳的近场动力学模拟[J]. 固体力学学报, 2024, 45(3): 326-340. PAN Yusong, PENG Xulong, CHEN Ziguang. Peridynamie modeling of corrosion fatigue in metallic materials[J]. China Industrial Economics, 2024: 45(3): 326-340.
[24] Liu FL, Chen Y, He C, et al. Tensile and very high cycle fatigue behaviors of a compressor blade titanium alloy at room and high temperatures[J]. Mater Sci Eng A, 2021, 811: 141049. doi: 10.1016/j.msea.2021.141049
[25] Mitragotri S, Anderson DG, Chen XY, et al. Accelerating the translation of nanomaterials in biomedicine[J]. ACS Nano, 2015, 9(7): 6644-6654.
[26] Dai Q, Wilhelm S, Ding D, et al. Quantifying the ligand-coated nanoparticle delivery to cancer cells in solid tumors[J]. ACS Nano, 2018, 12(8): 8423-8435.
[27] Vegas AJ, Veiseh O, Doloff JC, et al. Combinatorial hydrogel library enables identification of materials that mitigate the foreign body response in Primates[J]. Nat Biotechnol, 2016, 34(3): 345-352.
[28] Salinas AJ, Esbrit P, Vallet-Regí M. A tissue enginee-ring approach based on the use of bioceramics for bone repair[J]. Biomater Sci, 2013, 1(1): 40-51.
[29] Barber CC, Burnham M, Ojameruaye O, et al. A systematic review of the use of titanium versus stainless steel implants for fracture fixation[J]. OTA Int, 2021, 4(3): e138. doi: 10.1097/OI9.0000000000000138
[30] Xing YH, Zhong XY, Chen ZF, et al. Optimized osteogenesis of biological hydroxyapatite-based bone grafting materials by ion doping and osteoimmunomodulation[J]. Biomed Mater Eng, 2023, 34(3): 195-213.
[31] Chen N. Embedded 3D printing and pressurized thermo-curing of PMMA for medical implants[J]. J Mech Behav Biomed Mater, 2023, 146: 106083. doi: 10.1016/j.jmbbm.2023.106083
[32] Koakowska A, Kobuk D, Chwojnowski A, et al. Chitosan-based high-intensity modification of the biodegradable substitutes for cancellous bone[J]. J Funct Biomater, 2023, 14(8): 410. doi: 10.3390/jfb14080410
[33] Meyer M. Processing of collagen based biomaterials and the resulting materials properties[J]. Biomed Eng Online, 2019, 18(1): 24. doi: 10.1186/s12938-019-0647-0
[34] Prasad K, Bazaka O, Chua M, et al. Metallic biomate-rials: current challenges and opportunities[J]. Materials, 2017, 10(8): 884. doi: 10.3390/ma10080884
[35] Kordijazi A, Zhao T, Zhang J, et al. A review of application of machine learning in design, synthesis, and characterization of metal matrix composites: current status and emerging applications[J]. JOM, 2021, 73(7): 2060-2074.
[36] Eliaz N. Corrosion of metallic biomaterials: a review[J]. Materials, 2019, 12(3): 407. doi:10.3390/ma12030407
[37] Su YC, Luo C, Zhang ZH, et al. Bioinspired surface functionalization of metallic biomaterials[J]. J Mech Behav Biomed Mater, 2018, 77: 90-105. doi: 10.1016/j.jmbbm.2017.08.035
[38] Tepla TL, Izonin IV, Duriagina ZA, et al. Alloys selection based on the supervisedlearning technique for design ofbiocompatible medical materials[J]. Arch Mater Sci Eng, 2018, 1(93): 32-40.
[39] Cao XY, Zhang YB, Li JH, et al. Composition design of 7XXX aluminum alloys optimizing stress corrosion cracking resistance using machine learning[J]. Mater Res Express, 2020, 7(4): 046506. doi: 10.1088/2053-1591/ab8492
[40] Pourrahimi S, Hakimian S, Bouzid AH, et al. On the use of machine learning algorithms to predict the corrosion behavior of stainless steels in lactic acid[J]. Metals, 2023, 13(8): 1459. doi: 10.3390/met13081459
[41] Bao H, Wu SC, Wu ZK, et al. A machine-learning fatigue life prediction approach of additively manufactured metals[J]. Eng Fract Mech, 2021, 242: 107508. doi: 10.1016/j.engfracmech.2020.107508
[42] Barbosa JF, Correia JAFO, Júnior RCSF, et al. Fatigue life prediction of metallic materials considering mean stress effects by means of an artificial neural network[J]. Int J Fatigue, 2020, 135: 105527. doi: 10.1016/j.ijfatigue.2020.105527
[43] Zou CX, Li JS, Wang WY, et al. Integrating data mi-ning and machine learning to discover high-strength ductile titanium alloys[J]. Acta Mater, 2021, 202: 211-221. doi: 10.1016/j.actamat.2020.10.056
[44] Yang F, Li Z, Wang Q, et al. Cluster-formula-embedded machine learning for design of multicomponent β-Ti alloys with low youngs modulus[J]. NPJ Comput Mater, 2020, 6: 101. doi: 10.1038/s41524-020-00372-w
[45] Marimuthu KP, Lee J, Han G, et al. Machine learning based dual flat-spherical indentation approach for rough metallic surfaces[J]. Eng Appl Artif Intell, 2023, 125: 106724. doi: 10.1016/j.engappai.2023.106724
[46] Valipoorsalimi P, Sari YA, Pekguleryuz M. Mechanical property design of bio-compatible mg alloys using machine-learning algorithms[J]. ArXiv, 2023, abs/2305.12060. doi: 10.48550/arXiv.2305.12060
[47] Punj S, Singh J, Singh K. Ceramic biomaterials: properties, state of the art and future prospectives[J]. Ceram Int, 2021, 47(20): 28059-28074.
[48] Vaiani L, Boccaccio A, Uva AE, et al. Ceramic materials for biomedical applications: an overview on properties and fabrication processes[J]. J Funct Biomater, 2023, 14(3): 146. doi: 10.3390/jfb14030146
[49] Piconi C. Bioinert ceramics: state-of-the-art[J]. Key Eng Mater, 2017, 758: 3-13. doi: 10.4028/www.scientific.net/kem.758.3
[50] Montazerian M, Dutra Zanotto E. History and trends of bioactive glass-ceramics[J]. J Biomed Mater Res A, 2016, 104(5): 1231-1249.
[51] Qi L, Zhao T, Yan JG, et al. Advances in magnesium-containing bioceramics for bone repair[J]. Biomater Transl, 2024, 5(1): 3-20.
[52] Behbahani R, Yazdani Sarvestani H, Fatehi E, et al. Machine learning-driven process of alumina ceramics laser machining[J]. Phys Scr, 2023, 98(1): 015834. doi: 10.1088/1402-4896/aca3da
[53] Arif S, Alam MT, Ansari AH, et al. Analysis of tribological behaviour of zirconia reinforced Al-SiC hybrid composites using statistical and artificial neural network technique[J]. Mater Res Express, 2018, 5(5): 056506. doi: 10.1088/2053-1591/aabec8
[54] Gopinath KGS, Pal S, Tambe P. Prediction of weight percentage alumina and pore volume fraction in bio-ceramics using Gaussian process regression and minimax probability machine regression[J]. Mater Today Proc, 2018, 5(5): 12233-12239.
[55] Boone K, Tjokro N, Chu KN, et al. Machine learning enabled design features of antimicrobial peptides selectively targeting peri-implant disease progression[J]. Front Dent Med, 2024, 5: 1372534. doi: 10.3389/fdmed.2024.1372534
[56] Koh YG, Park KM, Lee JA, et al. Total knee arthroplasty application of polyetheretherketone and carbon-fiber-reinforced polyetheretherketone: a review[J]. Mater Sci Eng C Mater Biol Appl, 2019, 100: 70-81. doi: 10.1016/j.msec.2019.02.082
[57] Gupta A, Prasad A, Mulchandani N, et al. Multifunctional nanohydroxyapatite-promoted toughened high-molecular-weight stereo complex poly(lactic acid)-based bionanocomposite for both 3D-printed orthopedic implants and high-temperature engineering applications[J]. ACS Omega, 2017, 2(7): 4039-4052.
[58] Xu XW, Song J. Segmental long bone regeneration guided by degradable synthetic polymeric scaffolds[J]. Biomater Transl, 2020, 1(1): 33-45.
[59] Elkaseer A, Mueller T, Azcarate S, et al. Replication of overmolded orthopedic implants with a functionalized thin layer of biodegradable polymer[J]. Polymers, 2018, 10(7): 707. doi: 10.3390/polym10070707
[60] Özden G, Öteyaka MÖ, Cabrera FM. Modeling of cutting parameters in turning of PEEK composite using artificial neural networks and adaptive-neural fuzzy inference systems[J]. J Thermoplast Compos Mater, 2023, 36(2): 493-509.
[61] Li F, Han JS, Cao T, et al. Design of self-assembly dipeptide hydrogels and machine learning via their che-mical features[J]. Proc Natl Acad Sci U S A, 2019, 116(23): 11259-11264.
[62] Moumen A, Lakhdar A, Laabid Z, et al. Towards smart modeling of mechanical properties of a bio composite based on a machine learning[J]. Int J Electr Comput Eng IJECE, 2022, 12(3): 3138. doi: 10.11591/ijece.v12i3.pp3138-3145
[63] Shokrollahi Y, Dong PF, Gamage PT, et al. Finite element-based machine learning model for predicting the mechanical properties of composite hydrogels[J]. Appl Sci, 2022, 12(21): 10835. doi: 10.3390/app122110835
[64] Collins MN, Ren G, Young K, et al. Scaffold fabrication technologies and structure/function properties in bone tissue engineering[J]. Adv Funct Materials, 2021, 31(21): 2010609. doi: 10.1002/adfm.202010609
[65] Roseti L, Parisi V, Petretta M, et al. Scaffolds for bone tissue engineering: state of the art and new perspectives[J]. Mater Sci Eng C Mater Biol Appl, 2017, 78: 1246-1262. doi: 10.1016/j.msec.2017.05.017
[66] Meng M, Wang JZ, Huang HG, et al. 3D printing metal implants in orthopedic surgery: methods, applications and future prospects[J]. J Orthop Translat, 2023, 42: 94-112. doi: 10.1016/j.jot.2023.08.004
[67] Tan J, Chen ZC, Xu Z, et al. A 3D-printed scaffold composed of Alg/HA/SIS for the treatment of diabetic bone defects[J]. J Orthop Translat, 2024, 48: 25-38. doi: 10.1016/j.jot.2024.07.006
[68] Wang C, Huang W, Zhou Y, et al. 3D printing of bone tissue engineering scaffolds[J]. Bioact Mater, 2020, 5(1): 82-91.
[69] Zhang YH, He FP, Zhang Q, et al. 3D-printed flat-bone-mimetic bioceramic scaffolds for cranial restoration[J]. Research, 2023, 6: 0255. doi: 10.34133/research.0255
[70] Sun J, Yao K, An J, et al. Machine learning and 3D bioprinting[J]. Int J Bioprint, 2023, 9(4): 717. doi: 10.18063/ijb.717
[71] Chen HY, Liu YC, Balabani S, et al. Machine learning in predicting printable biomaterial formulations for direct ink writing[J]. Research, 2023, 6: 0197. doi: 10.34133/research.0197
[72] Gu GX, Chen CT, Richmond DJ, et al. Bioinspired hierarchical composite design using machine learning: simulation, additive manufacturing, and experiment[J]. Mater Horiz, 2018, 5(5): 939-945.
[73] Peng B, Wei Y, Qin Y, et al. Machine learning-enabled constrained multi-objective design of architected materials[J]. Nat Commun, 2023, 14(1): 6630. doi: 10.1038/s41467-023-42415-y
[74] Ratnavel R, Viswanath S, Subramanian J, et al. Predicting the optimal input parameters for the desired print quality using machine learning[J]. Micromachines, 2022, 13(12): 2231. doi: 10.3390/mi13122231
[75] Horikawa S, Suzuki K, Motojima K, et al. Material design of porous hydroxyapatite ceramics via inverse analysis of an estimation model for bone-forming ability based on machine learning and experimental validation of biological hard tissue responses[J]. Materials, 2024, 17(3): 571. doi: 10.3390/ma17030571
[76] Wu C, Entezari A, Zheng KK, et al. A machine learning-based multiscale model to predict bone formation in scaffolds[J]. Nat Comput Sci, 2021, 1(8): 532-541.
[77] Ryu B, Wang LQ, Pu HH, et al. Understanding, discovery, and synthesis of 2D materials enabled by machine learning[J]. Chem Soc Rev, 2022, 51(6): 1899-1925.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed