The development of automation and artificial intelligence (AI) software has progressed rapidly in radiation therapy over the years. These technologies have the ability to significantly improve the efficiency, quality, and consistency of the entire radiation therapy process, as well as to better understand patient outcomes (1–3). To realize the full potential of AI and automation, overcoming the barrier of moving these advancements from “benchtop to bedside” is essential. Integrating AI presents challenges similar to those faced with other technologies in the broader radiation therapy community, including stakeholder buy-in, practitioner training, and managing process change. However, AI also raises additional concerns, including the interpretability of results and the potential for bias to be incorporated into models (4). An important intermediate step to increasing the adoption of AI and automation-based software is to demonstrate validity in a prospective setting and ensure the quality of retrospective models that are reproducible and interpretable by clinicians (5, 6). Prospective implementation of these tools within multiple institutions is essential to familiarize clinical staff with their operation and demonstrate real-world application.
In the field of radiation oncology, there have been some successful examples, including treatment planning, brachytherapy, image analysis, and predictive modeling of patient outcomes (7-9). These examples are limited compared to the extensive studies that have performed retrospective model building applied to internal datasets or holdout sets. The community is approaching a critical point of development where novelty will be demonstrated by moving from initial development to clinical integration. This Frontiers in Oncology Research Topic will highlight research specifically focused on demonstrating the predictive use of automation and AI in clinical radiation oncology.
In this Research Topic, the articles are diverse in their specific clinical applications of automation or AI investigations, but unified in the development and validation of tools to improve the quality, consistency, and efficiency of radiation therapy workflows. New commercial techniques such as adaptive planning workflows, AI-based contouring, and automated quality assurance are available and require clinical validation before their impact can be fully realized. Three separate studies in this Research Topic, authored by Kehayias et al., Galand et al., and Doolan et al., provide the community with a framework for validation and specific clinical implementations that can be emulated in future studies. Apart from that, development of tools for diagnostic decision support, improving treatment quality, and better understanding patient outcomes is reported by Wang et al., Kowalchuk et al., Gan et al., and Schröder et al. These studies demonstrate this principle in a variety of applications throughout the treatment process, from pretreatment lesion detection to posttreatment survival prediction.
The highlights in these articles are intended to motivate the community to further explore the transition steps of automation and AI tools that need to be validated to maximize their impact on clinical workflow and outcomes for patients undergoing radiation therapy. We hope that readers will find these articles informative, motivating, and thought-provoking.
Author contributions
MR: Conceptualization, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing. XJ: Conceptualization, Investigation, Writing – original draft, Writing – review & editing. EK: Investigation, Writing – original draft, Writing – review & editing. SQ: Conceptualization, Investigation, Writing – original draft, Writing – review & editing, Methodology, Project administration, Visualization.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher's note
All claims made in this article are those of the authors themselves and do not necessarily represent those of their institutions, publishers, editors, or reviewers. Products evaluated in this article, or claims made by their manufacturers, are not endorsed or approved by the publishers.
References
1. Ciunkiewicz P, Roumeliotis M, Stenhouse K, McGeachy P, Quirk S, Grendarova P, et al. “Assessing tissue toxicity risk in breast radiotherapy using Bayesian networks.” Medical Physics(2022) 49:3585–96. doi: 10.1002/mp.15651
PubMed Abstract | CrossRef Full Text | Google Scholar
2. Morin O, Vallières M, Braunstein S, Ginart JB, Upadhaya T, Woodruff HC, et al. An artificial intelligence framework integrating longitudinal electronic health records and real-world data enables continuous pan-cancer prognosis. Nat Cancer(2021) 2:709–22. doi: 10.1038/s43018-021-00236-2
PubMed Abstract | CrossRef Full Text | Google Scholar
3. Quirk S, Lovis J, Stenhouse K, Van Dyke L, Roumeliotis M, Thind K. Technical note: a standardized automated framework for monitoring institutional radiation therapy protocol compliance. Medical Physics(2021) 48:2661–6. doi: 10.1002/mp.14797
PubMed Abstract | CrossRef Full Text | Google Scholar
5. Claessens M, Oria CS, Brouwer CL, Ziemer BP, Schooley JE, Lin H, et al. Quality assurance of AI-based applications in radiation therapy. Seminar: Radiation Oncology(2022) 32:421–31. doi: 10.1016/j.semradonc.2022.06.011
PubMed Abstract | CrossRef Full Text | Google Scholar
6. Ong Ly C, Unnikrishnan B, Tadic T, Patel T, Duhamel J, Kandel S, et al. “Shortcut learning in medical AI hinders generalization: How to estimate the generalization of AI models without external data.” NPJ Digit Med(2024) 7:124. doi: 10.1038/s41746-024-01118-4
PubMed Abstract | CrossRef Full Text | Google Scholar
7. McIntosh C, Conroy L, Tjong MC, Craig T, Bayley A, Catton C, et al. Clinical integration of machine learning in curative-intent radiation therapy for patients with prostate cancer. National Med(2021) 27:999–1005. doi: 10.1038/s41591-021-01359-w
PubMed Abstract | CrossRef Full Text | Google Scholar
8. Jung H, Shen C, Gonzalez Y, Albuquerque K, Jia X. Deep learning-assisted automated digitization of intrastitial needles in 3D CT image-based high-dose-rate brachytherapy for gynecological cancers. Physics, Medicine, Biology(2019) 64:215003. doi: 10.1088/1361-6560/ab3fcb
PubMed Abstract | CrossRef Full Text | Google Scholar
9. Stenhouse K, Roumeliotis M, Ciunkiewicz P, Martell K, Quirk S, Banerjee R, et al. Prospective validation of machine learning models for applicator and hybrid interstitial needle selection in high dose rate (HDR) cervical brachytherapy. Brachytherapy(2024) 23:368–76. doi: 10.1016/j.brachy.2024.02.008
PubMed Abstract | CrossRef Full Text | Google Scholar
keyword: Artificial intelligence, potential, implementation, automation, outcomes
Quote: Roumeliotis M, Jia X, Kim E, Quirk S (2024) Editorial: Future uses and clinical applications of artificial intelligence and data-driven automation in radiation therapy. Front desk. Oncol. 14:1445048. doi: 10.3389/fonc.2024.1445048
received: June 6, 2024; approved: June 17, 2024;
release date: July 18, 2024.
Copyright © 2024 Roumeliotis, Jia, Kim, Quirk. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY). Use, distribution, or reproduction in other forums is permitted, provided the original author and copyright holder are credited and the original publication in this journal is cited in accordance with accepted scholarly practice. Any use, distribution, or reproduction not in accordance with these terms is not permitted.
*correspondence: Michael Roumeliotis, mroumel1@jhu.edu; Sarah Quirk, squirk2@bwh.harvard.edu
