[2022 New Book] Deep learning biomedical application: from medical image to drug discovery

Source: Specialty

This article is introduced by books. It is recommended to read for 5 minutes

In this article, we will discuss the use of the TSFRESH package in depth. This book covers the basic method of widely used life science applications.

Biology, medicine and biochemistry have become data -centric fields, and deep learning methods are bringing breakthrough results to these fields. This "Deep Learning Biomedical", from machine learning practitioners and data scientists seeking method knowledge to solve biomedical applications.

With the contribution of internationally renowned experts, this book covers the basic methods of extensive life science applications, including electronic health record processing, diagnostic imaging, text processing, and group data processing. This book includes analysis of chemical information and biomedical interaction networks. The use of data -driven methods in life sciences also needs to carefully consider related social, ethical, legal, and transparent challenges, which is introduced in the final chapter of this book.

https://www.worldScientific.com/worldScibooks/10.1142/q0322 #t=aboutbook

The organization of this book follows the development of the intelligent information processing system to the progress of biomedical problems, and the recent recent research themes of the crossroads between modern machine learning and life sciences. Finally, we will discuss the influence of society, ethics, and laws that use deep learning technology in biological and medicine. These technologies are usually discussed under the unified term of trusted artificial intelligence.

The second chapter through the classification of general deep learning strategies considers in the literature, a comprehensive introduction to the deep learning field of medical images is introduced. The detailed analysis of brain imaging applications supplemented this extensive discussion, conducted a wide range of related work in the field, and clearly organized the relevant data sets to index. Finally, it determines the key challenge to be solved so that the applicability of deep imaging methods in clinical practice can be simplified.

The third chapter focuses on the evolution of the excavation of electronic health records in the era of deep learning, and discusses their key role in building a springboard for building a truly personalized diagnosis, treatment and nursing. Electronic Health Med low (EHR) records people's health information and accumulates in a large number of structured and non -structured data warehouses. challenge. This chapter has evolved from the origin and evolution of EHR to their current situation. Then, the main applications of deep learning were analyzed, and the supervision and unsupervised tasks of extensive categories were taken into account, including disease prediction, disease phenotypes, patient layers and clinical records.

Chapter 4 By gradually introducing the use of natural language technology in the field of biomedical, it expands the theme of understanding human language. This chapter first introduces the main concepts and methods in the field of natural language processing (NLP). Then explore the application of NLP in life science. The survey of methodology is well supplemented with accurate indexes of available resources, including language libraries, software libraries, and pre -trained language models, including general and specific areas.

Chapter 5 The vertical route is used to introduce a method to the potential space learning gene expression data of metabolic driver. This chapter discusses how to provide an effective unsupervised method to obtain a new insight into the structure of gene expression data. In particular, it focuses on how to learn the nerves through model learning that can be restrained based on the prior knowledge available in the form of metabolic models.

Chapter 6 concentrated in the deep learning of chemistry informatics and solved the long -term research field of crossroads between computer science and chemistry. It discusses how the compounds find their natural calculations as the data of the graphic structure. The atoms and their attributes are encoded by the vertex of the molecular graph, while the atomic key and their characteristics are represented. By constructing such a representation, this chapter introduces the vivid field of deep learning of structured data adaptive processing, which contains learning models that can process information in its rich structured representation. Then, it moves to two related applications in the field of chemistry informatics: designing drugs from the beginning of the nature prediction and generating deep learning model from the molecular structure.

Chapter VII focuses on the deep learning methods of network biology. In a sense, the complexity of interaction in the process of model biology by introducing a larger scale (ie, the network) is introduced. This naturally supplements Chapter 6 Discussion on structured data analysis.

Chapter 8 will focus on the challenge of application -driven to a people -centered perspective, detailing the needs of deep learning that can be explained in medical and medical health.

Chapter Nine summarizes the critical analysis of the morality, social and legal issues in the deep learning of medical care. This chapter not only praises the importance of artificial intelligence ethics, but also reviews this theme from a practical perspective, analyzes the meaning of the ethics and legal guidelines of deep learning in the medical field. Special attention to Europe's guidelines on credible AI and the realization of the life cycle of related AI applications. In the end, this chapter carried out technical in -depth discussions on bias, fairness and privacy in deep learning.

- END -