摘要：Orthopedic bone plates are most commonly used f.r bone f.acture f.xation f.r more than 100 *** bone plate design had evolved over time overcoming many challenges such as insuf.icient strength and excessive plate–bone contact af.ecting the blood ***,it is only made of.two materials,either stainless steel(AISI 316L)or titanium(Ti–6Al–4V).There are two main limitations of.metallic bone implants,namely stress shielding and the problem of.malocclusion caused by the displacement of.the f.acture site during *** overcome the two problems,a new bone plate design with the incorporation of.auxetic structures is proposed in this *** study aims to use auxetic structure section in the bone plate that would decrease the stif.ness of.the region,thereby mitigating the stress-shielding ef.ect and at the same time act as a def.rmable section to enable intra-operative bending f.r ef.ective alignment while having enough bending strength and *** dif.erent auxetic structures namely re-entrant honeycomb and missing rib structures were *** auxetic structure incorporated bone plates were designed,f.nite element analysis was done,f.bricated using direct metal laser sintering technique,and *** results indicate that the re-entrant honeycomb structure incorporated bone plates serve as an ef.ective bone design compared to the conventional bone plate design,in terms of.stress shielding and intra-operative bending while of.ering similar mechanical and bending strength.

摘要：Drop-on-demand (DOD) bioprinting has been widely used in tissue engineering due to its highthroughput ef.iciency and cost ef.ectiveness. However, this type of.bioprinting involves challenges such as satellite generation, too-large droplet generation, and too-low droplet speed. These challenges reduce the stability and precision of.DOD printing, disorder cell arrays, and hence generate f.rther structural errors. In this paper, a multi-objective optimization (MOO) design method f.r DOD printing parameters through f.lly connected neural networks (f.NNs) is proposed in order to solve these challenges. The MOO problem comprises two objective f.nctions: to develop the satellite f.rmation model with f.NNs;and to decrease droplet diameter and increase droplet speed. A hybrid multi-subgradient descent bundle method with an adaptive learning rate algorithm (HMSGDBA), which combines the multisubgradient descent bundle (MSGDB) method with Adam algorithm, is introduced in order to search f.r the Pareto-optimal set f.r the MOO problem. The superiority of.HMSGDBA is demonstrated through comparative studies with the MSGDB method. The experimental results show that a single droplet can be printed stably and the droplet speed can be increased f.om 0.88 to 2.08 m·s^-1 af.er optimization with the proposed method. The proposed method can improve both printing precision and stability, and is usef.l in realizing precise cell arrays and complex biological f.nctions. f.rthermore, it can be used to obtain guidelines f.r the setup of.cell-printing experimental platf.rms.