ROECSG report: Real-world implementation patterns and perceptions of artificial intelligence in radiation oncology.
6 affiliations
- 1Department of Radiation Medicine, University of Wisconsin, Madison, Wisconsin, USA.
- 2Department of Radiation Oncology, University of Michigan, Ann Arbor, Michigan, USA.
- 3Department of Radiation Oncology, University of Miami Miller School of Medicine, Miami, Florida, USA.
- 4Department of Radiation Oncology, University of Alabama at Birmingham, Birmingham, Alabama, USA.
- 5Department of Radiation Oncology, Medical College of Wisconsin, Milwaukee, Wisconsin, USA.
- 6Department of Therapeutic Radiology, Yale University, New Haven, Connecticut, USA.
Abstract
Purpose
While international surveys have examined radiation oncology professionals' attitudes toward artificial intelligence (AI) and documented generally low early utilization, actual implementation patterns and the depth of integration among engaged practitioners remain unclear. This multi-institutional survey aims to characterize current AI implementation, identify critical barriers to optimization, and evaluate the psychological factors driving clinical trust.Methods and materials
A de-identified survey covering six domains was developed by a multi-institutional working group and distributed via a widely subscribed medical physics listserv and at the 2025 American Association of Physicists in Medicine (AAPM) Annual Meeting. The instrument assessed adoption rates, specific workflow applications, and user perceptions (trust and understanding) using 6-point Likert scales to reduce central tendency bias. Quantitative data were analyzed using Pearson correlation coefficients and Welch's two-sample t-tests, while qualitative responses underwent inductive thematic analysis (κ = 0.78).Results
In this convenience sample of 80 respondents (46.3% radiation oncologists, 37.5% medical physicists), 85.0% (68/80) reported current AI use. Adoption is heavily concentrated in auto-contouring (88.2% of AI users), whereas diagnosis/prognostication (14.7%) and treatment planning (23.5%) remain limited. Most users (60.3%) employ AI for task automation without influencing clinical judgment. Trust in AI (3.55 ± 1.02) correlates positively with perceived time savings (r = 0.57), clarity of use (r = 0.54), and technical understanding (r = 0.51, all p < 0.001), with no significant difference among these correlations (Williams test, both p > 0.59). Notably, self-rated understanding averaged near the scale midpoint (3.44 ± 1.22), yet 88.8% welcomed greater AI integration. Qualitative analysis identified transparency (38%) and validation protocols (31%) as the primary barriers to deeper implementation.Conclusions
AI use was common within this sample (85.0%), although self-selection likely inflates this figure relative to the broader field. However, implementation remains "broad but shallow," focused on workflow automation rather than clinical intelligence. The gap between enthusiasm and understanding necessitates a shift toward active institutional governance and standardized AI-literacy education.
Related
- Perceptions of artificial intelligence for radiographic image accuracy: a cross-sectional survey of radiology personnel
- Early insights into nurses' perceptions, opportunities, and concerns regarding Ambient artificial intelligence in nursing practice
- Artificial intelligence-enabled mental health applications for adolescents: a scoping review of safety, effectiveness and clinical implementation
- Digital, Mobile and Artificial Intelligence‐Supported Interventions for Pressure Injury Prevention and Management: A Systematic Review of Clinical, Workflow and Implementation Outcomes
- Bridging the Artificial Intelligence Translation Gap in Perioperative Pulmonary and Respiratory Complications: A Practical Framework for Institution-Level Implementation
BibTeX
@article{sohn2026roecsg,
title = {{ROECSG report: Real-world implementation patterns and perceptions of artificial intelligence in radiation oncology.}},
author = {James J Sohn and Cory Knill and Elizabeth Bossart and Samuel Chen and Rohan K Patel and Kevin Du and Leah M Katz},
journal = {Journal of applied clinical medical physics},
year = {2026},
month = oct,
doi = {10.1002/acm2.70871},
url = {https://europepmc.org/article/MED/42905968},
note = {CC BY},
}
RIS
TY - JOUR
TI - ROECSG report: Real-world implementation patterns and perceptions of artificial intelligence in radiation oncology.
AU - James J Sohn
AU - Cory Knill
AU - Elizabeth Bossart
AU - Samuel Chen
AU - Rohan K Patel
AU - Kevin Du
AU - Leah M Katz
PY - 2026
DA - 2026/10/30
JO - Journal of applied clinical medical physics
DO - 10.1002/acm2.70871
UR - https://europepmc.org/article/MED/42905968
AB - Purpose While international surveys have examined radiation oncology professionals' attitudes toward artificial intelligence (AI) and documented generally low early utilization, actual implementation patterns and the depth of integration among engaged practitioners remain unclear. This multi-institutional survey aims to characterize current AI implementation, identify critical barriers to optimization, and evaluate the psychological factors driving clinical trust.Methods and materials A de-identified survey covering six domains was developed by a multi-institutional working group and distributed via a widely subscribed medical physics listserv and at the 2025 American Association of Physicists in Medicine (AAPM) Annual Meeting. The instrument assessed adoption rates, specific workflow applications, and user perceptions (trust and understanding) using 6-point Likert scales to reduce central tendency bias. Quantitative data were analyzed using Pearson correlation coefficients and Welch's two-sample t-tests, while qualitative responses underwent inductive thematic analysis (κ = 0.78).Results In this convenience sample of 80 respondents (46.3% radiation oncologists, 37.5% medical physicists), 85.0% (68/80) reported current AI use. Adoption is heavily concentrated in auto-contouring (88.2% of AI users), whereas diagnosis/prognostication (14.7%) and treatment planning (23.5%) remain limited. Most users (60.3%) employ AI for task automation without influencing clinical judgment. Trust in AI (3.55 ± 1.02) correlates positively with perceived time savings (r = 0.57), clarity of use (r = 0.54), and technical understanding (r = 0.51, all p < 0.001), with no significant difference among these correlations (Williams test, both p > 0.59). Notably, self-rated understanding averaged near the scale midpoint (3.44 ± 1.22), yet 88.8% welcomed greater AI integration. Qualitative analysis identified transparency (38%) and validation protocols (31%) as the primary barriers to deeper implementation.Conclusions AI use was common within this sample (85.0%), although self-selection likely inflates this figure relative to the broader field. However, implementation remains "broad but shallow," focused on workflow automation rather than clinical intelligence. The gap between enthusiasm and understanding necessitates a shift toward active institutional governance and standardized AI-literacy education.
SN - 1526-9914
ER -