Optimization of polydopamine coating process for poly lactic acid‐based 3D printed bone plates using machine learning approaches

Abstract The three‐dimensional (3D) printed poly lactic acid (PLA) bone plates lack mechanical strength, resulting in premature failure. Coating these plates with polydopamine (PDM) forms covalent bonds with the PLA molecular structure, enhancing their mechanical properties. The mechanical strength of the coated bone plates is influenced by infill density, submersion time, shaker speed, and coating solution concentration. However, conducting experiments for each parameter value to achieve maximum biomechanical tensile strength (BTS) and biomechanical flexural strength (BFS) is time‐consuming and costly. Overall, the combination of response surface methodology (RSM) and machine learning (ML) enables determination of the best printing parameters, leading to reduced material waste, personalized bone plates tailored to individual anatomy, improved implant fit, and functionality. Moreover, this approach has the potential to reduce the need for additional surgeries and overall costs. To optimize coating parameters, this study employs RSM and ML techniques, including genetic algorithm (GA), particle swarm optimization (PSO), random search optimization (RSO), and differential evolution (DE). Experimental validation of the optimized process parameters and their corresponding fitness values is carried out using both RSM and ML approaches. The results demonstrate that GA has the closest relationship between experimental and fitness values, followed by DE, RSM, PSO, and RSO. Highlights Direct immersion coating of polydopamine on 3D printed PLA bone plates. Evaluating mechanical strength for bone plates coated at varying parameters. Statistical modeling and optimization of mechanical strength using RSM. Mechanical strength optimization and convergence properties for ML models. Experimental validation of RSM and ML‐based optimization algorithms.