Review on machine learning application in tissue engineering: What has been done so far? Application areas, challenges, and perspectives

Abstract Artificial intelligence and machine learning (ML) approaches have recently been getting much of researchers’ attention. The growing interest in these methods results from the fast development of machine learning algorithms in the last few years, especially artificial neural networks. In this review, we pay attention to the need and benefits that ML approaches can bring to tissue engineering (TE). We critically evaluate the possibilities of using the ML approaches in the tissue engineering field. We consider various paths of its utility in the TE, such as scaffold design, predicting the biological response to the scaffold, optimizing drug delivery approaches, supporting image analysis, and modeling scaffold in vivo performance. The current status of ML implementation is presented and supported by many study examples. On the other hand, we analyze the present difficulties and challenges in implementing ML approaches to tissue engineering, including the quality of published data, databases and repositories availability, the need for experiment and results publishing standardization, and ethical issues. Additionally, we assess the available natural language processing tools that could support TE research. Graphical abstract