Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Machine Learning in Oncology: Methods, Applications, and Challenges

Machine Learning in Oncology: Methods, Applications, and Challenges review articles SPECIAL SERIES: CANCER CLASSIFICATION SYSTEMS Machine Learning in Oncology: Methods, Applications, and Challenges 1,2 2 Dimitris Bertsimas, PhD ; and Holly Wiberg, BS INTRODUCTION a tumor, length of survival, or treatment response. Unsupervised learning identifies patterns and sub- Machine learning (ML) has the potential to transform 1 groups within data where there is no clear outcome to oncology and, more broadly, medicine. The in- predict. It is often used for more exploratory analysis. troduction of ML in health care has been enabled by the Reinforcement learning is yet a third class of ML used digitization of patient data, including the adoption of for sequential decision making where a strategy must electronic medical records (EMRs). This transition be learned from data; this has natural applications in provides an unprecedented opportunity to derive clin- determining optimal treatment regimens for patients ical insights from large-scale analysis of patient data. 7,8 with cancer. This review focuses on supervised and Clinical decisions have traditionally been guided by unsupervised learning settings. medical guidelines and accumulated experience. ML Supervised Learning methods add rigor to this process; algorithms can generate individualized predictions by synthesizing In this section, we introduce several common super- data across broad http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JCO Clinical Cancer Informatics Wolters Kluwer Health

Machine Learning in Oncology: Methods, Applications, and Challenges

Loading next page...
 
/lp/wolters-kluwer-health/machine-learning-in-oncology-methods-applications-and-challenges-luqvoeZYWD
Publisher
Wolters Kluwer Health
Copyright
(C) 2020 American Society of Clinical Oncology
ISSN
2473-4276
DOI
10.1200/CCI.20.00072
Publisher site
See Article on Publisher Site

Abstract

review articles SPECIAL SERIES: CANCER CLASSIFICATION SYSTEMS Machine Learning in Oncology: Methods, Applications, and Challenges 1,2 2 Dimitris Bertsimas, PhD ; and Holly Wiberg, BS INTRODUCTION a tumor, length of survival, or treatment response. Unsupervised learning identifies patterns and sub- Machine learning (ML) has the potential to transform 1 groups within data where there is no clear outcome to oncology and, more broadly, medicine. The in- predict. It is often used for more exploratory analysis. troduction of ML in health care has been enabled by the Reinforcement learning is yet a third class of ML used digitization of patient data, including the adoption of for sequential decision making where a strategy must electronic medical records (EMRs). This transition be learned from data; this has natural applications in provides an unprecedented opportunity to derive clin- determining optimal treatment regimens for patients ical insights from large-scale analysis of patient data. 7,8 with cancer. This review focuses on supervised and Clinical decisions have traditionally been guided by unsupervised learning settings. medical guidelines and accumulated experience. ML Supervised Learning methods add rigor to this process; algorithms can generate individualized predictions by synthesizing In this section, we introduce several common super- data across broad

Journal

JCO Clinical Cancer InformaticsWolters Kluwer Health

Published: Oct 15, 2020

References