Deep Learning and Its Applications in Biomedicine.

Cao C1, Liu F2, Tan H3, Song D3, Shu W2, Li W4, Zhou Y5, Bo X6, Xie Z7.

Author information

1: CapitalBio Corporation, Beijing 102206, China.
2: Department of Biotechnology, Beijing Institute of Radiation Medicine, Beijing 100850, China.
3: State Key Lab of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou 500040, China.
4: Zhongshan School of Medicine, Sun Yat-sen University, Guangzhou 500040, China.
5: CapitalBio Corporation, Beijing 102206, China; Department of Biomedical Engineering, Medical Systems Biology Research Center, Tsinghua University School of Medicine, Beijing 100084, China. Electronic address: yimingzhou@capitalbio.com.
6: Department of Biotechnology, Beijing Institute of Radiation Medicine, Beijing 100850, China. Electronic address: boxc@bmi.ac.cn.
7: State Key Lab of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou 500040, China. Electronic address: xiezh8@sysu.edu.cn.

Abstract

Advances in biological and medical technologies have been providing us explosive volumes of biological and physiological data, such as medical images, electroencephalography, genomic and protein sequences. Learning from these data facilitates the understanding of human health and disease. Developed from artificial neural networks, deep learning-based algorithms show great promise in extracting features and learning patterns from complex data. The aim of this paper is to provide an overview of deep learning techniques and some of the state-of-the-art applications in the biomedical field. We first introduce the development of artificial neural network and deep learning. We then describe two main components of deep learning, i.e., deep learning architectures and model optimization. Subsequently, some examples are demonstrated for deep learning applications, including medical image classification, genomic sequence analysis, as well as protein structure classification and prediction. Finally, we offer our perspectives for the future directions in the field of deep learning.

KEYWORDS:

Big data; Bioinformatics; Biomedical informatics; Deep learning; High-throughput sequencing; Medical image

PMID:

29522900

PMCID:

PMC6000200

DOI:

10.1016/j.gpb.2017.07.003

Figure 4. Popularity of deep learning frameworks in Github .

The distributions of stars in Github of deep learning frameworks written in C++, Lua, Python, Matlab, Julia, and Java are shown in the pie chart. More stars in Github indicate higher popularity. Font size of the frameworks in the pie chart reflects the number of stars.