Machine learning applications for electrospun nanofibers: a review

Abstract Electrospun nanofibers have gained prominence as a versatile material, with applications spanning tissue engineering, drug delivery, energy storage, filtration, sensors, and textiles. Their unique properties, including high surface area, permeability, tunable porosity, low basic weight, and mechanical flexibility, alongside adjustable fiber diameter distribution and modifiable wettability, make them highly desirable across diverse fields. However, optimizing the properties of electrospun nanofibers to meet specific requirements has proven to be a challenging endeavor. The electrospinning process is inherently complex and influenced by numerous variables, including applied voltage, polymer concentration, solution concentration, solution flow rate, molecular weight of the polymer, and needle-to-collector distance. This complexity often results in variations in the properties of electrospun nanofibers, making it difficult to achieve the desired characteristics consistently. Traditional trial-and-error approaches to parameter optimization have been time-consuming and costly, and they lack the precision necessary to address these challenges effectively. In recent years, the convergence of materials science and machine learning (ML) has offered a transformative approach to electrospinning. By harnessing the power of ML algorithms, scientists and researchers can navigate the intricate parameter space of electrospinning more efficiently, bypassing the need for extensive trial-and-error experimentation. This transformative approach holds the potential to significantly reduce the time and resources invested in producing electrospun nanofibers with specific properties for a wide range of applications. Herein, we provide an in-depth analysis of current work that leverages ML to obtain the target properties of electrospun nanofibers. By examining current work, we explore the intersection of electrospinning and ML, shedding light on advancements, challenges, and future directions. This comprehensive analysis not only highlights the potential of ML in optimizing electrospinning processes but also provides valuable insights into the evolving landscape, paving the way for innovative and precisely engineered electrospun nanofibers to meet the target properties for various applications. Graphical abstract