Artificial intelligence in radiotherapy: a technological review
- PDF / 2,309,493 Bytes
- 19 Pages / 595.276 x 785.197 pts Page_size
- 2 Downloads / 208 Views
REVIEW
Artificial intelligence in radiotherapy: a technological review Ke Sheng (
✉)
Department of Radiation Oncology, University of California, Los Angeles, CA 90095, USA
© Higher Education Press 2020
Abstract Radiation therapy (RT) is widely used to treat cancer. Technological advances in RT have occurred in the past 30 years. These advances, such as three-dimensional image guidance, intensity modulation, and robotics, created challenges and opportunities for the next breakthrough, in which artificial intelligence (AI) will possibly play important roles. AI will replace certain repetitive and labor-intensive tasks and improve the accuracy and consistency of others, particularly those with increased complexity because of technological advances. The improvement in efficiency and consistency is important to manage the increasing cancer patient burden to the society. Furthermore, AI may provide new functionalities that facilitate satisfactory RT. The functionalities include superior images for real-time intervention and adaptive and personalized RT. AI may effectively synthesize and analyze big data for such purposes. This review describes the RT workflow and identifies areas, including imaging, treatment planning, quality assurance, and outcome prediction, that benefit from AI. This review primarily focuses on deep-learning techniques, although conventional machine-learning techniques are also mentioned. Keywords
artificial intelligence; radiation therapy; medical imaging; treatment planning; quality assurance; outcome prediction
Introduction Radiation therapy (RT) is used to treat over 60% of cancer patients in the US and 30% of cancer patients in China. The oncological use of radiation to treat malignancies started immediately after the discovery of radioactive isotopes. In the first half-century of RT technological development, most of the research and development effort has been allocated to making high-energy and highly penetrating radiation sources available for the treatment of deep tumors. In the past 40 years, engineering has played an important role in RT technologies. The change has been largely driven by the availability of three-dimensional (3D) images for treatment planning. 3D images from modalities, such as computed tomography (CT) or magnetic resonance (MR), provide quantitative delineation of the tumor target and organs at risk (OARs). Such images also support accurate 3D dose calculation, which in combination with the organ delineation provides a wealth of statistical information to correlate with the tumor control probability
Received August 10, 2019; accepted February 14, 2020 Correspondence: Ke Sheng, [email protected]
and normal organ toxicity—this knowledge quantumleaped modern RT to be a quantitative science. With decades of technological evolution, modern RT workflow can be simplified into a flowchart (Fig. 1). The 3D images are first acquired for a RT patient. Then, the gross tumor volume (GTV) and OARs are delineated on 3D images. GTV describes visible tumors based on medical images. Th
Data Loading...
