Metal 3D printing of biometals for prostheses and implants: a review

Опубликована: Май 13, 2025

Metal 3D printing has revolutionized the fabrication of biometallic prostheses and implants, offering unprecedented design flexibility, patient-specific customization, enhanced biomechanical performance. This review explores current advancements in metal additive manufacturing (AM) techniques, including selective laser melting (SLM), electron beam (EBM), fused deposition modeling (FDM), directed energy (DED), sheet lamination, stereolithography (SLA), binder jetting, for processing biocompatible metals such as titanium, cobalt-chromium, stainless steel. The article discusses major benefits, osseointegration, complex lattice architectures weight saving, optimized mechanical properties. challenges residual stresses, surface finish, regulatory issues are also discussed. concludes by defining future research avenues material design, process development, clinical translation to increase efficacy reliability 3D-printed biometal implants.

Язык: Английский

---

Explainable machine learning for predicting tensile properties of aluminum alloys in the laser powder bed fusion process

Materials Science and Engineering A, Год журнала: 2025, Номер unknown, С. 148482 - 148482

Опубликована: Май 1, 2025

Язык: Английский

---

Development of an Algorithm Based on an Artificial Neural Network to Predict the Thermal Profile of Weld Joints of Dissimilar High Strength Steels Using GMAW Process

International journal of engineering research in Africa, Год журнала: 2025, Номер 74, С. 17 - 32

Опубликована: Июнь 2, 2025

In developing an accurate modelling technique of thermal profile parameters when welding High-strength steel, algorithm based on artificial neural network (ANN) for predicting cooling time using Gas Metal Arc Welding (GMAW) was set up. The developed has a 4-20-1 architecture with the input voltage station ( U ), current intensity I speed V and heat Q ) output parameter Cooling ∆t8/5 ). A protocol been MATLAB R2020a software containing three networks. goal to determine that lowest root mean square error (MSE). results showed first system produced MSE 1.295 × 10 −3 regression R = 0.995 Relative 0 8 initial 14 data. second 4.278 0.978 11/15 showing 0. Finally, third system, consisting associated experimental data analytical 2.506 10−3 0.972 slight difference between predicted all 29 points. obtained by two systems are satisfactory networks can be found reliable times welded joints steel high-strength.

Язык: Английский

---

Evaluation of Machine Learning Models for Enhancing Sustainability in Additive Manufacturing

Technologies, Год журнала: 2025, Номер 13(6), С. 228 - 228

Опубликована: Июнь 3, 2025

Additive manufacturing (AM) presents significant opportunities for advancing sustainability through optimized process control and material utilization. This research investigates the application of machine learning (ML) models to directly associate AM parameters with metrics, which is often a challenge by experimental methods alone. Initially, data are generated systematically varying key parameters, layer height, infill density, pattern, build orientation, number shells. Subsequently, four ML models, Linear Regression, Decision Trees, Random Forest, Gradient Boosting, trained evaluated. Hyperparameter tuning conducted using Limited-memory Broyden–Fletcher–Goldfarb–Shanno Box constraints (L-BFGS-B) algorithm, demonstrates superior computational efficiency compared traditional approaches such as grid random search. Among Forest yields highest predictive accuracy lowest mean squared error across all target indicators: energy consumption, part weight, scrap production time. The results confirm efficacy in predicting outcomes when supported robust data. offers scalable computationally efficient approach enhancing processes contributes data-driven decision-making sustainable manufacturing.

Язык: Английский

---

Predicting the Relative Density of Stainless Steel and Aluminum Alloys Manufactured by L-PBF Using Machine Learning

Journal of Manufacturing and Materials Processing, Год журнала: 2025, Номер 9(6), С. 185 - 185

Опубликована: Июнь 3, 2025

Metal additive manufacturing is a disruptive technology that changing how various alloys are processed. Although this has several advantages over conventional manufacturing, it still necessary to standardize its properties, which dependent on the relative density (RD). In addition, since experimental designs costly, one solution using machine learning algorithms allow effects of variations in processing parameters resulting additively manufactured components be anticipated. This work assembled database based data from 673 observations and 10 predictors forecast 316L stainless steel AlSi10Mg produced by laser powder bed fusion (L-PBF). LazyPredict was employed select algorithm best models variability inherent data. Ensemble boosting regressors offer higher accuracy, providing hyperparameter fitting optimization advantages. The predictions’ precision for aluminum obtained an R2 value greater than 0.86 0.83, respectively. results SHAP values indicated power energy have greatest impact predictability Al-Si10-Mg SS materials processed L-PBF. study presents compendium fabrication alloys, offering researchers guide understanding influence RD.

Язык: Английский

---