Learning for Personalized Medicine: A Comprehensive Review From a Deep Learning Perspective

Abstract

With the recent advancements in analyzing high-volume, complex, and unstructured data, modern learning methods are playing an increasingly critical role in the field of personalized medicine. Personalized medicine (i.e., providing tailored medical treatment to individual patients through the identification of common features, including their genetics, inheritance, and lifestyle) has attracted the attention of many researchers in recent years. This paper provides an overview of the research progress in the application of learning methods, with a focus on deep learning in personalized medicine. In particular, three domains of applications are reviewed: drug development, disease characteristic identification, and therapeutic effect prediction. The main objective of this review is to consider the applied methods in detail and to offer insights into their pros and cons. Although having demonstrated advantages in coping with data complexity and nonlinearity and in recognizing features and associating structural data, the studied learning methods are not a panacea to all medical problems. Hence, we discuss the existing research challenges and clarify future study directions.