Research Article

Recent Trends of Artificial Intelligence in Radiation Oncology: A Narrative Review of Prospective Studies

By Jas Virk, MD1; Simeng Zhu, MD2; Austin J. Sim, MD, JD2; Sung Jun Ma, MD2*
Published Date: August 1, 2025

Affiliations

  1. 1 Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, The State University of New York, Buffalo, NY
  2. 2 Department of Radiation Oncology, The Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, The Ohio State University Comprehensive Cancer Center, Columbus, OH

In the past several decades, the delivery of radiation therapy has become increasingly intricate and precise. Such advancements were observed in conjunction with abundant multimodal data available for analysis; these include sophisticated diagnostic imaging, electronic health records, and digital pathology. The impact of artificial intelligence (AI) has become more prominent as numerous prior and ongoing prospective studies aim to integrate it into clinical care in radiation oncology. This review article provides an overview of such prospective studies and examines the role of AI in radiation therapy. By providing an understanding of recent trends in AI, we hope to contribute to improved patient outcomes and precision medicine in radiation oncology.

Keywords: AI, machine learning, deep learning, radiomics, large language model, multimodal

Introduction

Radiation therapy has progressed significantly over the past decades through such advances as stereotactic body radiation therapy (SBRT) for lung cancer1 - 3 and oligometastatic cancer,4 - 6 proton therapy for leptomeningeal metastasis,7 magnetic resonance imaging (MRI)-guided SBRT for prostate cancer,8 MRI-guided adaptive radiation therapy for pancreatic cancer,9 and adaptive radiation therapy for head and neck cancer.10 In addition, precision medicine has evolved to improve patient selection for various treatment approaches, including prostate-specific membrane antigen (PSMA) positron emission tomography (PET) for prostate cancer,11, 12 F-fluoromisonidazole PET for head and neck cancer,13, 14 gallium DOTATATE PET for meningioma,15, 16 21-gene recurrence scores for breast cancer,17, 18 osimertinib after definitive chemoradiation for stage III epidermal growth factor receptor (EGFR)-mutant non-small cell lung cancer,19 and chimeric antigen receptor T-cell therapy.12

With such advancements in precision medicine, cancer genetics, and imaging modalities leading to abundant multimodal data available for health care professionals to interpret, artificial intelligence (AI) has emerged to leverage such data.20 For example, AI-based algorithms have greatly improved early diagnosis of breast cancer,21 pancreatic cancer,22 lung cancer,23 and skin cancer.24 Furthermore, generative AI has been shown to answer questions with more empathy than humans25 and to assist with medical documentation.26 In radiation oncology, several AI-related studies have emerged to minimize unplanned hospitalization27 and detect extranodal extension (ENE) in head and neck cancer.28, 29 Since then, numerous reviews have summarized the role of AI in radiation oncology.30 - 33 However, none have focused on prospective studies incorporating AI into practice. In this review, we aimed to highlight the overview of recent trends in the application of AI in radiation oncology based on prior and ongoing prospective studies.

ADVERTISEMENT

Methods

To identify relevant prospective studies on AI trends in radiation oncology, a literature search was conducted of the following electronic databases: PubMed, Medline, and Google Scholar. The following keywords were used: “radiation,” “radiation oncology,” and “artificial intelligence.” The search was limited to publications ranging from January 2002 to December 2024 and excluded retrospective studies, systematic reviews, case reports, conference abstracts, and expert opinion articles. Additional filters included utilizing only English language-written articles. Article titles and abstracts were then reviewed after initial screening, followed by full-text review prior to finalizing study inclusion.

Current clinical trials were searched utilizing the ClinicalTrials.gov website with the following keywords: “cancer,” “artificial intelligence,” and “radiation.” Studies that were completed or active (recruiting or not) were included, while those that were suspended or withdrawn were excluded. Trials were further categorized based on type, with only interventional studies included with no specific date range. When evaluating prospective studies or clinical trials, two reviewers determined the eligibility of such studies for inclusion.

ADVERTISEMENT

Results

Of 4469 articles found through our literature search, 234 were initially identified as prospective studies. After reviewing abstracts and full texts to confirm their eligibility, 30 studies met our criteria, as shown in Table 1 .

Table 1.
ADVERTISEMENT

Prior prospective studies

AUTHORS YEAR DISEASE SITE PROSPECTIVE DATA DATA TYPES MAIN FINDINGS
Zeleznik et al.34 2021 Breast Not available CT scan With deep learning assistance, heart segmentation time was significantly reduced. Expert accuracy was comparable with deep learning-only segmentations.
Ma et al.35 2023 Breast ClinicalTrials.gov ID: NCT05609058 CT scan Deep learning model identified the lead wire markers in the CT scan images, and the organ feature based on such markers was correlated with ipsilateral lung V20.
Dembrower et al.14 2023 Breast ScreenTrustCAD Mammogram Replacing one radiologist with AI for independent assessment of screening mammograms was non-inferior for cancer detection compared with reading by two radiologists.
Preetha et al.36 2021 CNS CORE, CENTRIC, EORTC 26101 MRI scan Synthetic postcontrast MRI scan based on pre-contrast MRI scanning using deep learning was feasible with no statistically significant difference in the contrast-enhancing tumor burden when compared to postcontrast MRI scanning.
Tsang et al.37 2024 CNS Not available CT scan 94% of ML plans and 93% of manual plans were deemed to be clinically acceptable. ML plans were able to give 1 Gy less radiation to the normal brain than the manual plan. ML plans required 45 fewer minutes on average to create compared to manual plans.
George et al.38 2024 CNS ClinicalTrials.gov ID: NCT02336165 MRI scan First on-treatment MRI features were correlated with overall and progression-free survival, while baseline MRI features were not.
Hong et al.27 2020 General SHIELD-RT Clinical variables AI-based algorithm based on routine electronic health record data triaged patients and reduced acute care visits during treatments.
Friesner et al.39 2022 General NCT02649569, NCT03102229, NCT03115398 Daily step counts Daily step counts using an ML model were correlated with hospitalizations.
Kehayias et al.40 2024 General Not available CT scan The integration of Deep Learning On-Demand Assistant, an automated clinical platform to help with auto-segmentations and QA reporting using AI, into radiation oncology clinic workflow was feasible.
Natesan et al.41 2024 General SHIELD-RT Clinical variables High-risk patients identified by the AI-based algorithm experienced lower total medical costs from twice-weekly evaluations.
Wang et al.42 2022 GI RTOG 0822 CT scan AI-based algorithm using clinical variables, DVH, and radiomic features predicted pCR.
Wesdorp et al.43 2023 GI CAIRO5 CT scan A DL autosegmentation model accurately segmented the liver and metastatic lesions.
Fremond et al.44 2023 GYN PORTEC-1, PORTEC-2, PORTEC-3, TransPORTEC Whole-slide images of H&E slides A DL model predicted molecular classification.
Walker et al.45 2014 Head/Neck Not available CT scan Autosegmentation of organs at risk reduced the amount of time needed for segmentation, but expert oversight is still required for accuracy.
Men et al.46 2019 Head/Neck RTOG 0522 CT scan AI-based algorithm predicted the incidence of late xerostomia.
Sher et al.47 2021 Head/Neck Not available Radiation plans AI-based decision support tool improved the dose metrics for organs at risk.
Osapoetra et al.48 2021 Head/Neck ClinicalTrials.gov ID: NCT03908684 Quantitative ultrasound AI-based algorithm predicted treatment response of involved lymph nodes.
Mashayekhi et al.49 2023 Head/Neck Not available Radiation plans AI-based decision support tool improved uniformity of practice.
Kann et al.29 2023 Head/Neck ECOG/ACRIN 3311 CT scan AI-based algorithm predicted extranodal extension more effectively than did radiologists.
Sher et al.50 2023 Head/Neck INRT-AIR CT scan AI-based algorithm identified involved or suspicious lymph nodes, and there was no solitary elective nodal recurrence at 2 years without elective nodal irradiation.
Nicolae et al.51 2020 Prostate Not available Ultrasound AI-based radiation treatment planning reduced the time required for planning and was considered clinically acceptable.
McIntosh et al.52 2021 Prostate Not available Radiation plans AI-based radiation treatment planning reduced the time required for planning and was considered clinically acceptable.
Sanders et al.53 2022 Prostate Not available MRI scan Autosegmentation of prostate and organs at risk was considered clinically feasible.
Thomas et al.54 2022 Prostate ClinicalTrials.gov ID: NCT03238170 Radiation plans AI-based algorithm predicted those who would benefit from rectal spacer placement.
Johnsson et al.55 2022 Prostate OSPREY PSMA PET/CT AI-based algorithm identified potential lesions and autosegmented organs.
Esteva et al.56 2022 Prostate NRG/RTOG 9202, 9413, 9910, 0126 Whole slide images of H&E slides AI-based algorithm risk stratified and identified patients with poor prognoses.
Spratt et al.57 2023 Prostate NRG/RTOG 9202, 9413, 9910, 0126, 9408 Whole slide images of H&E slides AI-based algorithm predicted patients who would benefit from androgen deprivation therapy.
Ross et al.58 2024 Prostate NRG/RTOG 9902 Whole slide images of H&E slides AI-based algorithm risk stratified and identified patients with poor prognoses.
Spratt et al.59 2024 Prostate NRG/RTOG 9202, 9408, 9413, 9910, 9902, 0521 Whole slide images of H&E slides AI-based algorithm risk stratified and identified patients with poor prognoses.
Wong et al.60 2020 Prostate/Head Neck/CNS Not available CT scan AI-based algorithm reduced the time required for contouring and autosegmented at-risk organs and target volumes.

Abbreviations: AI, artificial intelligence; CT, computed tomography; CNS, central nervous system; DVH, dose volume histogram; H&E, hematoxylin and eosin; GI; gastrointestinal, GYN, gynaecological; MRI, magnetic resonance imaging; ML, machine learning; PSMA, prostate specific membrane antigen; pCR, pathologic complete response; QA, quality assurance.

AI in Prostate Cancer

AI has been investigated extensively to improve outcomes of patients with prostate cancer. In earlier years, because of substantial interobserver disagreements in Gleason grade among pathologists,61, 62 AI-assisted digital pathology algorithms based on whole-slide images of hematoxylin and eosin-stained tissues were developed to improve reproducibility in determining Gleason grade,63 which were recognized by Food and Drug Administration and other regulatory agencies.63

Beyond assessment of Gleason grades, the role of digital pathology has been investigated in radiation oncology. Esteva et al. initially leveraged five NRG Oncology phase III randomized clinical trials (NRG/RTOG 9202, 9413, 9910, 0126, and 9408) that included patients with localized prostate cancer who received radiation with or without androgen-deprivation therapy (ADT).56 Self-supervised, prognostic, and multimodal AI architecture was developed based on clinical variables (age, Gleason primary and secondary grades, T stage, and baseline PSA) from over 5600 patients and imaging features from over 16 000 histopathology slides.56 Across all endpoints, AI outperformed the National Comprehensive Cancer Network (NCCN) risk-stratification tool by 9.2%-14.6% for relative improvements in area under the receiver operating characteristic curve (AUC).56

With its early success, digital pathology was further investigated for its predictive ability. Spratt et al. utilized four NRG Oncology phase III clinical trials (NRG/RTOG 9202, 9413, 9910, 0126) to develop a similar multimodal AI architecture and validated its performance on the NRG/RTOG 9408 dataset.57 The primary objective of this study was to identify a subgroup of patients who might benefit from adding ADT to radiation.57 The development cohort comprised over 2000 patients, with the majority having intermediate-risk prostate cancer, while the validation cohort consisted of over 1500 patients, with more than half having intermediate-risk prostate cancer.57 Over a third of patients in the validation cohort were classified as predictive model-positive, demonstrating an absolute improvement of 10% by adding ADT for distant metastasis-free survival and prostate-cancer-specific survival at 15 years.57 However, no differential treatment benefits were identified between predictive model subgroups for metastasis-free survival and overall survival.57 Spratt et al. performed a separate analysis using six NRG Oncology clinical trials (NRG/RTOG 9202, 9408, 9413, 9910, 9902, 0521), validating the multimodal AI algorithm as prognostic for distant metastasis and prostate cancer-specific mortality among patients with high-risk prostate cancer.59 Subsequently, the NCCN Guideline for prostate cancer included ArteraAI Prostate as the first AI-based tool with prognostic and predictive benefits from ADT among patients with localized prostate cancer.64

AI in Head and Neck Cancer

Other malignancies targeted by extensive research in AI are head and neck cancers, especially with respect to radiomics. For example, ENE is a known adverse feature associated with poor locoregional control.65, 66 However, ENE identification has been largely based on pathologic evaluation, since radiographic determination has been inconsistent.67 - 69 As a result, 24%-31% of patients with p16+ head and neck cancer receive trimodality therapy.70, 71 To reduce this knowledge gap, Kann et al. developed a deep-learning (DL) algorithm based on 270 patients from a single institution with over 650 lymph nodes segmented.72 The model predicted ENE and nodal metastasis with an AUC of 0.91 for both endpoints.72 Based on such early success, Kann et al. utilized validation datasets of 82 patients with 130 lymph nodes segmented from Mount Sinai Hospital and 62 patients with 70 lymph nodes segmented from The Cancer Genome Atlas imaging data through The Cancer Imaging Archive.28 The DL model predicted ENE with an AUC of 0.84-0.90 on these validation datasets, outperforming diagnostic radiologists and improving interobserver agreement among these radiologists.28 Owing to the small sample size of p16-positive oropharyngeal cancer in these retrospective datasets,28 further validation was performed using a multicenter phase II clinical trial, ECOG-ACRIN 3311.29 The DL model was retrained using three retrospective datasets as mentioned previously, ultimately identifying 178 patients from ECOG-ACRIN 3311 with 313 manually segmented lymph nodes.29 It had an AUC of 0.86 for the identification of ENE, outperforming four radiologists, with a limitation of node level segmentation required prior to independent testing.29

Another evolving paradigm for treatment de-escalation among patients with head and neck cancer is to reduce treatment volume. Several phase II clinical trials and a large retrospective study demonstrated the feasibility of reducing the dose of elective nodal irradiation to 30-40 Gy.73 - 75 To omit elective nodal irradiation, colleagues from the University of Texas Southwestern Medical Center evaluated several DL models using 129 patients and over 700 lymph nodes segmented with AUC of 0.88-0.98,76 - 78 comparable to the AUC of 0.91 from the study by Kann et al.72 Subsequently, Sher et al. incorporated this model in the prospective phase II INRT-AIR trial.50 Of 67 patients with nonmetastatic head and neck cancer who underwent definitive radiation or chemoradiation, an average of 31 lymph nodes per patient were evaluated by the DL model, determining that approximately 10% were involved.50 At a median follow-up of 33 months, overall and progression-free survival at 2 years were favorable at 91% and 82%, respectively.50 One patient with heavy marijuana use had an out-of-field elective nodal recurrence with concurrent distant metastasis, but the study otherwise found favorable quality of life outcomes with no solitary elective nodal failure.50

AI in Supportive Care

In addition to improving oncologic outcomes, another area incorporating AI is the effort to reduce acute care visits, such as emergency department visits and unplanned hospitalizations. Predicting such events has been investigated among patients without a cancer diagnosis.79 - 82

In radiation oncology, Hong et al. initially developed a machine learning (ML) model based on nearly 7000 patients with over 8000 treatment courses at a single institution; this model included variables such as baseline demographics, disease and treatment characteristics, prior acute care visits, laboratory values, and recent vital signs.83 Internal validation demonstrated an AUC of 0.80 for the ML model in predicting acute care visits.83 Subsequently, Hong et al. performed the SHIELD-RT single-institution, prospective quality improvement study.27 This model was utilized to identify high-risk patients, who were defined as having more than a 10% risk of acute care visits, and randomized them to twice-weekly on-treatment visits versus standard of care.27 Of nearly 1000 treatment courses, 311 were evaluated as high-risk courses, with the majority of patients having gastrointestinal cancer or primary brain cancer.27 The ML model had a favorable performance with an AUC of 0.82 for triaging patients to high- versus low-risk for acute care visits, and fewer than 3% of low-risk patients had acute care visits.27 Twice-weekly evaluation led to a reduction from 22% to 12% of acute care visits during radiation therapy, the primary endpoint of this study.27 Furthermore, a post-hoc economic analysis showed that such a reduction in acute care visits translated to lower health care costs.41

Ongoing Clinical Trials

Table 2 consists of a list of ongoing clinical trials that incorporate AI. In particular, a multimodal AI risk-stratification developed by Spratt et al.57, 59 has been incorporated into two such clinical trials. The HypoElect study (ClinicalTrials.gov ID: NCT06582446) is a single-arm phase II clinical trial that consists of patients with NCCN high-risk, multimodal AI high-risk prostate cancer and is evaluating the role of whole-pelvis radiation in five fractions with radiation dose escalation using brachytherapy and two years of ADT. The second study is the (ClinicalTrials.gov ID: NCT06772441), a single-arm, phase II HypoPro clinical trial comprising patients with NCCN high-risk, multimodal AI low-/intermediate-risk prostate cancer and is investigating SBRT in combination with brachytherapy and concurrent ADT. Additionally, while most ongoing clinical trials leverage AI for adaptive radiation therapy ( Table 2 ), another noteworthy study is a randomized clinical trial by researchers at the University of Hong Kong (ClinicalTrials.gov ID: NCT06636188). It is the first prospective study incorporating a chatbot, Digi-Coach, to help reduce physical and psychological distress versus usual nursing care among patients with head and neck cancer.

Table 2.

Ongoing Prospective Studies

CLINICAL TRIAL CLINICAL TRIALS.GOV ID START DATE ESTIMATED END DATE STUDY DESIGN ROLE OF AI Status Disease site
Artificial IntelligenceI for Prostate Cancer Treatment Planning NCT04441775 2020 2022 Observational Improve consistency and quality of radiation treatment plans. Completed Prostate
Two Studies for Patients With High Risk Prostate Cancer Testing Less Intense Treatment for Patients With a Low Gene Risk Score and Testing a More Intense Treatment for Patients With a High Gene Risk Score, The PREDICT-RT Trial NCT04513717 2020 2033 Interventional Radiation therapy quality assurance using an AI algorithm. Recruiting Prostate
ARtificial Intelligence for Gross Tumor Volume Segmentation (ARGOS) NCT05775068 2021 2024 Observational Autosegmentation of GTV on CT scan. Active, not recruiting Thoracic
Artificial Intelligence in Functional Imaging for Individualized Treatment of Head and Neck Squamous Cell Carcinoma Patients (KIVAL-KHT) NCT05192655 2021 2026 Observational Analysis of diagnostic imaging and clinical and histopathological data to predict outcomes. Recruiting Head/Neck
AI for Head Neck Cancer Treated With Adaptive RadioTherapy (RadiomicART) NCT05081531 2021 2024 Interventional Analysis of diagnostic imaging to predict outcomes and toxicities. Recruiting Head/Neck
PostRadiotherapy MRI-based AI System to Predict Radiation Proctitis for Pelvic Cancers NCT04918992 2021 TBD Observational Analysis of post-radiation MRI scan to predict proctitis. Unknown status General
Clinical Validation of AI-Assisted Radiotherapy Contouring Software for Thoracic Organs At Risk NCT05787522 2022 2024 Observational Autosegmentation of organs at risk on CT scan. Completed Thoracic
Simulation-Free Hippocampal-Avoidance Whole Brain Radiotherapy Using Diagnostic MRI-Based and Cone Beam Computed Tomography-Guided On-Table Adaptive Planning in a Novel Ring Gantry Radiotherapy Device NCT05096286 2022 2022 Interventional Simulation-free workflow using a semi-automated planning based on AI. Completed CNS
The Impact of Radiotherapy on Oligometastatic Cancer NCT05933876 2022 2037 Observational Analysis of clinical data, medical images, and biological samples to predict who will benefit from radiation to oligometastatic sites. Recruiting General
Intensive Symptom Surveillance Guided by Machine Learning-Directed Risk Stratification in Patients With Non-Metastatic Head and Neck Cancer, The INSIGHT Trial NCT05338905 2022 2027 Interventional Analysis of clinical data to identify high-risk patients who will benefit from symptom surveillance Recruiting Head/Neck
Artificial Intelligence in CNS Radiation Oncology (AI-RAD) NCT06036394 2023 2028 Observational Autosegmentation of tumor and organs at risk, use radiomics to predict toxicities and outcomes. Active, not recruiting CNS
Stereotactic Body Radiation Therapy Planning With Artificial Intelligence-Directed Dose Recommendation for Treatment of Primary or Metastatic Lung Tumors, RAD-AI Study NCT05802186 2023 2026 Interventional AI to guide radiation dose for primary lung cancer and lung metastases. Recruiting Thoracic
Adaptive Radiation in Anal Cancer NCT05838391 2023 2025 Interventional Adaptive radiation using AI. Recruiting GI
Randomized Evaluation of Machine Learning Assisted Radiation Treatment Planning versus Standard Radiation Treatment Planning NCT05979883 2023 2026 Interventional-Phase III AI-assisted radiation treatment planning. Recruiting Head/Neck
MR-guidance in Chemoradiotherapy for Cervical Cancer (AIM-C1) NCT06142760 2023 2026 Interventional Adaptive radiation using AI. Recruiting GU
Daily-Adaptive Stereotactic Body Radiation Therapy for Biochemically Recurrent, Radiologic Apparent Prostate Cancer After Radical Prostatectomy NCT05946824 2023 2028 Interventional-Phase II Adaptive radiation using AI. Recruiting Prostate
Computed Tomography-Guided Stereotactic Adaptive Radiotherapy (CT-STAR) for the Treatment of Central and Ultra-Central Early-Stage Non-Small Cell Lung Cancer NCT05785845 2023 2026 Interventional Adaptive radiation using AI. Recruiting Thoracic
A Chatbot to Reduce Physical and Psychological Distress of Patients With Head and Neck Cancer Undergoing Radiotherapy NCT06636188 2024 2027 Interventional AI-based patient navigator chatbot to reduce physical and psychological distress. Active, not recruiting Head/Neck
Glioma Adaptive Radiotherapy With Development of an Artificial Intelligence Workflow (GLADIATOR) NCT06492486 2024 2028 Interventional-Phase II Adaptive radiation using AI. Not yet recruiting CNS
AI as an Aid for Weekly Symptom Intake in Radiotherapy NCT06525181 2024 2024 Interventional Medical documentation for on-treatment visits to improve accuracy and efficiency. Not yet recruiting General
A phase II Clinical Trial of Artificial Intelligence-assisted One-stop Radiotherapy for Breast Cancer After Breast-conserving Surgery (BC-AIO) NCT06686459 2024 2027 Interventional-Phase II Autosegmentation and radiation treatment planning. Not yet recruiting Breast
Evaluation of a Novel Auto Segmentation Algorithm for Normal Structure Delineation in Radiation Treatment Planning NCT06200116 2024 2026 Observational Autosegmentation. Recruiting General
Online Adaptive Radiotherapy for Nasopharyngeal Carcinoma (OART) NCT06516133 2024 2030 Phase III Clinical Trial Adaptive radiation using AI. Recruiting Head/Neck
One Fraction Simulation-Free Treatment With CT-Guided Stereotactic Adaptive Radiotherapy for Patients With Oligometastatic and Primary Lung Tumors (ONE STOP) NCT06236516 2024 2025 Phase III Clinical Trial Adaptive radiation using AI. Recruiting Thoracic
Artificial Intelligence to Personalize Prostate Cancer Treatment (the HypoElect Trial) (HypoElect) NCT06582446 2024 2027 Interventional-Phase II Patient selection and risk stratification. Recruiting Prostate
Artificial Intelligence Driven Personalisation of Radiotherapy and Concomitant Androgen Deprivation Therapy for Prostate Cancer Patients (the HypoPro Trial) (HypoPro) NCT06772441 2024 2027 Interventional Patient selection and risk stratification. Recruiting Prostate
RAdiotherapy With FDG-PET Guided Dose-PAINTing Compared With Standard Radiotherapy for Primary Head and Neck Cancer-3 (RADPAINT-3) NCT06297902 2024 2030 Interventional Analysis of blood samples to predict tumor response and toxicities. Recruiting Head/Neck
Artificial Intelligence-Guided Radiotherapy Planning for Glioblastoma (ARTPLAN-GLIO) NCT06657027 2025 2027 Observational Analysis of MRI scans to evaluate the extent of tumor infiltration. Not yet recruiting CNS
Locally Optimised Contouring With AI Technology for Radiotherapy (LOCATOR) NCT06546592 2025 2029 Interventional Autosegmentation. Not yet recruiting General

Abbreviations: AI, artificial intelligence; CT, computed tomography; GTV, gross tumor volume; MRI, magnetic resonance imaging.

Limitations

Limitations of this study include its utilization only of prospective studies while excluding retrospective studies and other types of journal articles. The rationale for this decision is that several published reviews already incorporate retrospective studies to discuss the role of AI in radiation oncology.30 - 33 As a result, however, bias may be introduced toward reporting studies from major cancer centers with access to experts with significant AI technical skills. Subsequently, results from these prospective studies may not be generalizable to or implemented in smaller community cancer centers without access to such AI expertise. For instance, significant barriers hindered implementation of the SHIELD-RT trial process; these included labor-intensive, manual verification of treatment course data for each eligible patient, generating and verifying AI predictions by multiple investigators for each enrolled patient, and manually deploying clinical alerts for treating physicians and enrolled patients to ensure that the intervention was completed on time per protocol.84 In addition, discussion of commercially available technologies is beyond the scope of this review. These have been comprehensively discussed by NRG Oncology in its summary of the roles of commercial products in adaptive radiation, autosegmentation, treatment planning, and clinical trial development.85 - 88 Lastly, despite our efforts to include prospective AI data, we may have inadvertently excluded other relevant studies from our review. Further studies are warranted to capture the growing complexity of AI and its impact in radiation oncology.

ADVERTISEMENT

Conclusion

Radiation oncology is poised to be influenced substantially by AI in the coming decades. Emerging AI tools will streamline radiation treatment planning and adaptive radiation, guide treatment recommendations by improving patient selection based on digital pathology and radiomics, and tailor supportive care to reduce acute care visits. As a result, such efforts will translate to further progress in radiation oncology and patient outcomes.

References

1. Bezjak A , Paulus R , Gaspar LE , et al. Safety and efficacy of a five-fraction stereotactic body radiotherapy schedule for centrally located non-small-cell lung cancer: NRG oncology/RTOG 0813 trial. J Clin Oncol. 2019; 37 ( 15 ): 1316 - 1325. 10.1200/JCO.18.00622
2. Singh AK , Gomez-Suescun JA , Stephans KL , et al. One versus three fractions of stereotactic body radiation therapy for peripheral stage I to II non-small cell lung cancer: a randomized, multi-institution, phase 2 trial. Int J Radiat Oncol Biol Phys. 2019; 105 ( 4 ): 752 - 759. 10.1016/j.ijrobp.2019.08.019
3. Videtic GMM , Hu C , Singh AK , et al. A randomized phase 2 study comparing 2 stereotactic body radiation therapy schedules for medically inoperable patients with stage I peripheral non-small cell lung cancer: NRG oncology RTOG 0915 (NCCTG N0927). Int J Radiat Oncol Biol Phys. 2015; 93 ( 4 ): 757 - 764. 10.1016/j.ijrobp.2015.07.2260
4. Deek MP , Van der Eecken K , Sutera P , et al. Long-term outcomes and genetic predictors of response to metastasis-directed therapy versus observation in oligometastatic prostate cancer: analysis of STOMP and ORIOLE trials. J Clin Oncol. 2022; 40 ( 29 ): 3377 - 3382. 10.1200/JCO.22.00644
5. Palma DA , Olson R , Harrow S , et al. Stereotactic ablative radiotherapy for the comprehensive treatment of oligometastatic cancers: long-term results of the SABR-COMET phase II randomized trial. J Clin Oncol. 2020; 38 ( 25 ): 2830 - 2838. 10.1200/JCO.20.00818
6. Tsai CJ , Yang JT , Shaverdian N , et al. Standard-of-care systemic therapy with or without stereotactic body radiotherapy in patients with oligoprogressive breast cancer or non-small-cell lung cancer (Consolidative Use of Radiotherapy to Block [CURB] oligoprogression): an open-label, randomised, controlled, phase 2 study. Lancet. 2024; 403 ( 10422 ): 171 - 182. 10.1016/S0140-6736(23)01857-3
7. Yang JT , Wijetunga NA , Pentsova E , et al. Randomized phase II trial of proton craniospinal irradiation versus photon involved-field radiotherapy for patients with solid tumor leptomeningeal metastasis. J Clin Oncol. 2022; 40 ( 33 ): 3858 - 3867. 10.1200/JCO.22.01148
8. Kishan AU , Ma TM , Lamb JM , et al. Magnetic resonance imaging-guided vs computed tomography-guided stereotactic body radiotherapy for prostate cancer: the MIRAGE randomized clinical trial. JAMA Oncol. 2023; 9 ( 3 ): 365 - 373. 10.1001/jamaoncol.2022.6558
9. Chuong MD , Lee P , Low DA , et al. Stereotactic MR-guided on-table adaptive radiation therapy (SMART) for borderline resectable and locally advanced pancreatic cancer: a multi-center, open-label phase 2 study. Radiother Oncol. 2024; 191: 110064. 10.1016/j.radonc.2023.110064
10. Castelli J , Thariat J , Benezery K , et al. Weekly adaptive radiotherapy vs standard intensity-modulated radiotherapy for improving salivary function in patients with head and neck cancer: a phase 3 randomized clinical trial. JAMA Oncol. 2023; 9 ( 8 ): 1056 - 1064. 10.1001/jamaoncol.2023.1352
11. Hofman MS , Lawrentschuk N , Francis RJ , et al. Prostate-specific membrane antigen PET-CT in patients with high-risk prostate cancer before curative-intent surgery or radiotherapy (propsma): a prospective, randomised, multicentre study. Lancet. 2020; 395 ( 10231 ): 1208 - 1216. 10.1016/S0140-6736(20)30314-7
12. June CH , Sadelain M . Chimeric antigen receptor therapy. N Engl J Med. 2018; 379 ( 1 ): 64 - 73. 10.1056/NEJMra1706169
13. Lee NY , Sherman EJ , Schöder H , et al. Hypoxia-directed treatment of human papillomavirus-related oropharyngeal carcinoma. J Clin Oncol. 2024; 42 ( 8 ): 940 - 950. 10.1200/JCO.23.01308
14. Dembrower K , Crippa A , Colón E , et al. Artificial intelligence for breast cancer detection in screening mammography in sweden: a prospective, population-based, paired-reader, non-inferiority study. Lancet Digit Health. 2023; 5 ( 10 ): e703 - e711. 10.1016/S2589-7500(23)00153-X
15. Perlow HK , Nalin AP , Handley D , et al. A prospective registry study of68 Ga-DOTATATE PET/CT incorporation into treatment planning of intracranial meningiomas. Int J Radiat Oncol Biol Phys. 2024; 118 ( 4 ): 979 - 985. 10.1016/j.ijrobp.2023.10.014
16. Perlow HK , Raleigh DR , Wang TJC , et al. Consensus radiation treatment planning guidelines using (68)Ga-DOTATATE PET/CT for resected meningiomas. Int J Radiat Oncol Biol Phys. 2025; 122 ( 1 ): 150 - 158. 10.1016/j.ijrobp.2024.12.003
17. Kalinsky K , Barlow WE , Gralow JR , et al. 21-gene assay to inform chemotherapy benefit in node-positive breast cancer. N Engl J Med. 2021; 385 ( 25 ): 2336 - 2347. 10.1056/NEJMoa2108873
18. Sparano JA , Gray RJ , Makower DF , et al. Adjuvant chemotherapy guided by a 21-gene expression assay in breast cancer. N Engl J Med. 2018; 379 ( 2 ): 111 - 121. 10.1056/NEJMoa1804710
19. Lu S , Kato T , Dong X , et al. osimertinib after chemoradiotherapy in stage III EGFR -mutated NSCLC. N Engl J Med. 2024; 391 ( 7 ): 585 - 597. 10.1056/NEJMoa2402614
20. Howell MD , Corrado GS , DeSalvo KB . Three epochs of artificial intelligence in health care. JAMA. 2024; 331 ( 3 ): 242 - 244. 10.1001/jama.2023.25057
21. McKinney SM , Sieniek M , Godbole V , et al. International evaluation of an AI system for breast cancer screening. Nature. 2020; 577 ( 7788 ): 89 - 94. 10.1038/s41586-019-1799-6
22. Cao K , Xia Y , Yao J , et al. Large-scale pancreatic cancer detection via non-contrast CT and deep learning. Nat Med. 2023; 29 ( 12 ): 3033 - 3043. 10.1038/s41591-023-02640-w
23. Ardila D , Kiraly AP , Bharadwaj S , et al. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nat Med. 2019; 25 ( 6 ): 954 - 961. 10.1038/s41591-019-0447-x
24. Liu Y , Jain A , Eng C , et al. A deep learning system for differential diagnosis of skin diseases. Nat Med. 2020; 26 ( 6 ): 900 - 908. 10.1038/s41591-020-0842-3
25. Ayers JW , Poliak A , Dredze M , et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Intern Med. 2023; 183 ( 6 ): 589 - 596. 10.1001/jamainternmed.2023.1838
26. Nayak A , Alkaitis MS , Nayak K , et al. Comparison of history of present illness summaries generated by a chatbot and senior internal medicine residents. JAMA Intern Med. 2023; 183 ( 9 ): 1026 - 1027. 10.1001/jamainternmed.2023.2561
27. Hong JC , Eclov NCW , Dalal NH , et al. System for high-intensity evaluation during radiation therapy (SHIELD-RT): a prospective randomized study of machine learning-directed clinical evaluations during radiation and chemoradiation. J Clin Oncol. 2020; 38 ( 31 ): 3652 - 3661. 10.1200/JCO.20.01688
28. Kann BH , Hicks DF , Payabvash S , et al. Multi-institutional validation of deep learning for pretreatment identification of extranodal extension in head and neck squamous cell carcinoma. J Clin Oncol. 2020; 38 ( 12 ): 1304 - 1311. 10.1200/JCO.19.02031
29. Kann BH , Likitlersuang J , Bontempi D , et al. Screening for extranodal extension in HPV-associated oropharyngeal carcinoma: evaluation of a CT-based deep learning algorithm in patient data from a multicentre, randomised de-escalation trial. Lancet Digit Health. 2023; 5 ( 6 ): e360 - e369. 10.1016/S2589-7500(23)00046-8
30. Huynh E , Hosny A , Guthier C , et al. Artificial intelligence in radiation oncology. Nat Rev Clin Oncol. 2020; 17 ( 12 ): 771 - 781. 10.1038/s41571-020-0417-8
31. Kalsi S , French H , Chhaya S , et al. The evolving role of artificial intelligence in radiotherapy treatment planning-a literature review. Clin Oncol (R Coll Radiol). 2024; 36 ( 10 ): 596 - 605. 10.1016/j.clon.2024.06.005
32. Kawamura M , Kamomae T , Yanagawa M , et al. Revolutionizing radiation therapy: the role of AI in clinical practice. J Radiat Res. 2024; 65 ( 1 ): 1 - 9. 10.1093/jrr/rrad090
33. Parkinson C , Matthams C , Foley K , Spezi E . Artificial intelligence in radiation oncology: a review of its current status and potential application for the radiotherapy workforce. Radiography (Lond). 2021; 27 Suppl 1: S63 - S68. 10.1016/j.radi.2021.07.012
34. Zeleznik R , Weiss J , Taron J , et al. Deep-learning system to improve the quality and efficiency of volumetric heart segmentation for breast cancer. NPJ Digit Med. 2021; 4 ( 1 ): 43. 10.1038/s41746-021-00416-5
35. Ma L , Yang Y , Ma J , et al. Correlation between AI-based CT organ features and normal lung dose in adjuvant radiotherapy following breast-conserving surgery: a multicenter prospective study. BMC Cancer. 23 ( 1 ): 1085. 10.1186/s12885-023-11554-2
36. Jayachandran Preetha C , Meredig H , Brugnara G , et al. Deep-learning-based synthesis of post-contrast T1-weighted MRI for tumour response assessment in neuro-oncology: a multicentre, retrospective cohort study. Lancet Digit Health. 2021; 3 ( 12 ): e784 - e794. 10.1016/S2589-7500(21)00205-3
37. Tsang DS , Tsui G , Santiago AT , et al. A prospective study of machine learning-assisted radiation therapy planning for patients receiving 54 Gy to the brain. Int J Radiat Oncol Biol Phys. 2024; 119 ( 5 ): 1429 - 1436. 10.1016/j.ijrobp.2024.02.022
38. George E , Flagg E , Chang K , et al. Radiomics-based machine learning for outcome prediction in a multicenter phase II study of programmed death-ligand 1 inhibition immunotherapy for glioblastoma. AJNR Am J Neuroradiol. 2022; 43 ( 5 ): 675 - 681. 10.3174/ajnr.A7488
39. Friesner ID , Feng J , Kalnicki S , et al. Machine learning-based prediction of hospitalization during chemoradiotherapy with daily step counts. JAMA Oncol. 2024; 10 ( 5 ): 642 - 647. 10.1001/jamaoncol.2024.0014
40. Kehayias CE , Yan Y , Bontempi D , et al. Prospective deployment of an automated implementation solution for artificial intelligence translation to clinical radiation oncology. Front Oncol. 2023; 13: 1305511. 10.3389/fonc.2023.1305511
41. Natesan D , Eisenstein EL , Thomas SM , et al. Health care cost reductions with machine learning-directed evaluations during radiation therapy—an economic analysis of a randomized controlled study. NEJM AI. 2024; 1 ( 4 ): AIoa2300118. 10.1056/aioa2300118
42. Wang D , Lee SH , Geng H , et al. Interpretable machine learning for predicting pathologic complete response in patients treated with chemoradiation therapy for rectal adenocarcinoma. Front Artif Intell. 2022; 5: 1059033. 10.3389/frai.2022.1059033
43. Wesdorp NJ , Zeeuw JM , Postma SCJ , et al. Deep learning models for automatic tumor segmentation and total tumor volume assessment in patients with colorectal liver metastases. Eur Radiol Exp. 2023; 7 ( 1 ): 75. 10.1186/s41747-023-00383-4
44. Fremond S , Andani S , Barkey Wolf J , et al. Interpretable deep learning model to predict the molecular classification of endometrial cancer from haematoxylin and eosin-stained whole-slide images: a combined analysis of the PORTEC randomised trials and clinical cohorts. Lancet Digit Health. 2023; 5 ( 2 ): e71 - e82. 10.1016/S2589-7500(22)00210-2
45. Walker GV , Awan M , Tao R , et al. Prospective randomized double-blind study of atlas-based organ-at-risk autosegmentation-assisted radiation planning in head and neck cancer. Radiother Oncol. 2014; 112 ( 3 ): 321 - 325. 10.1016/j.radonc.2014.08.028
46. Men K , Geng H , Zhong H , et al. A deep learning model for predicting xerostomia due to radiation therapy for head and neck squamous cell carcinoma in the RTOG 0522 clinical trial. Int J Radiat Oncol Biol Phys. 2019; 105 ( 2 ): 440 - 447. 10.1016/j.ijrobp.2019.06.009
47. Sher DJ , Godley A , Park Y , et al. Prospective study of artificial intelligence-based decision support to improve head and neck radiotherapy plan quality. Clin Transl Radiat Oncol. 2021; 29: 65 - 70. 10.1016/j.ctro.2021.05.006
48. Osapoetra LO , Dasgupta A , DiCenzo D , et al. Assessment of clinical radiosensitivity in patients with head-neck squamous cell carcinoma from pre-treatment quantitative ultrasound radiomics. Sci Rep. 2021; 11 ( 1 ): 6117. 10.1038/s41598-021-85221-6
49. Mashayekhi M , McBeth R , Nguyen D , et al. Artificial intelligence guided physician directive improves head and neck planning quality and practice uniformity: a prospective study. Clin Transl Radiat Oncol. 2023; 40: 100616. 10.1016/j.ctro.2023.100616
50. Sher DJ , Moon DH , Vo D , et al. Efficacy and quality-of-life following involved nodal radiotherapy for head and neck squamous cell carcinoma: the INRT-AIR phase II clinical trial. Clin Cancer Res. 2023; 29 ( 17 ): 3284 - 3291. 10.1158/1078-0432.CCR-23-0334
51. Nicolae A , Semple M , Lu L , et al. Conventional vs machine learning-based treatment planning in prostate brachytherapy: results of a phase I randomized controlled trial. Brachytherapy. 2020; 19 ( 4 ): 470 - 476. 10.1016/j.brachy.2020.03.004
52. McIntosh C , Conroy L , Tjong MC , et al. Clinical integration of machine learning for curative-intent radiation treatment of patients with prostate cancer. Nat Med. 2021; 27 ( 6 ): 999 - 1005. 10.1038/s41591-021-01359-w
53. Sanders JW , Kudchadker RJ , Tang C , et al. Prospective evaluation of prostate and organs at risk segmentation software for MRI-based prostate radiation therapy. Radiol Artif Intell. 2022; 4 ( 2 ): e210151. 10.1148/ryai.210151
54. Thomas C , Dregely I , Oksuz I , et al. Neural network dose prediction for rectal spacer stratification in dose‐escalated prostate radiotherapy. Med Phys. 2022; 49 ( 4 ): 2172 - 2182. 10.1002/mp.15575
55. Johnsson K , Brynolfsson J , Sahlstedt H , et al. Analytical performance of apromise: automated anatomic contextualization, detection, and quantification of [(18)F]DCFPyL (PSMA) imaging for standardized reporting. Eur J Nucl Med Mol Imaging. 2022; 49 ( 3 ): 1041 - 1051. 10.1007/s00259-021-05497-8
56. Esteva A , Feng J , van der Wal D , et al. Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials. NPJ Digit Med. 2022; 5 ( 1 ): 71. 10.1038/s41746-022-00613-w
57. Spratt DE , Tang S , Sun Y , et al. Artificial intelligence predictive model for hormone therapy use in prostate cancer. NEJM Evid. 2023; 2 ( 8 ): EVIDoa2300023. 10.1056/EVIDoa2300023
58. Ross AE , Zhang J , Huang H-C , et al. External validation of a digital pathology-based multimodal artificial intelligence architecture in the NRG/RTOG 9902 phase 3 trial. Eur Urol Oncol. 2024; 7 ( 5 ): 1024 - 1033. 10.1016/j.euo.2024.01.004
59. Spratt DE , Liu VYT , Jia AY , et al. Meta-analysis of individual patient-level data for a multimodal artificial intelligence biomarker in high-risk prostate cancer: results from six NRG/RTOG phase 3 randomized trials. Eur Urol. 2024; 86 ( 4 ): 369 - 371. 10.1016/j.eururo.2024.06.019
60. Wong J , Fong A , McVicar N , et al. Comparing deep learning-based auto-segmentation of organs at risk and clinical target volumes to expert inter-observer variability in radiotherapy planning. Radiother Oncol. 2020; 144: 152 - 158. 10.1016/j.radonc.2019.10.019
61. Glaessgen A , Hamberg H , Pihl C-G , et al. Interobserver reproducibility of percent gleason grade 4/5 in total prostatectomy specimens. J Urol. 2002; 168 ( 5 ): 2006 - 2010. 10.1016/S0022-5347(05)64283-8
62. Sadimin ET , Khani F , Diolombi M , Meliti A , Epstein JI . Interobserver reproducibility of percent gleason pattern 4 in prostatic adenocarcinoma on prostate biopsies. Am J Surg Pathol. 2016; 40 ( 12 ): 1686 - 1692. 10.1097/PAS.0000000000000714
63. Zhu L , Pan J , Mou W , et al. Harnessing artificial intelligence for prostate cancer management. Cell Rep Med. 2024; 5 ( 4 ): 101506. 10.1016/j.xcrm.2024.101506
64. Network NCC. Prostate cancer (Version 1.2025). Accessed 22 January 2025. https://www.nccn.org/professionals/physician_gls/pdf/prostate.pdf
65. Bernier J , Domenge C , Ozsahin M , et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med. 2004; 350 ( 19 ): 1945 - 1952. 10.1056/NEJMoa032641
66. Cooper JS , Pajak TF , Forastiere AA , et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med. 2004; 350 ( 19 ): 1937 - 1944. 10.1056/NEJMoa032646
67. Chai RL , Rath TJ , Johnson JT , et al. Accuracy of computed tomography in the prediction of extracapsular spread of lymph node metastases in squamous cell carcinoma of the head and neck. JAMA Otolaryngol Head Neck Surg. 2013; 139 ( 11 ): 1187 - 1194. 10.1001/jamaoto.2013.4491
68. Maxwell JH , Rath TJ , Byrd JK , et al. Accuracy of computed tomography to predict extracapsular spread in p16-positive squamous cell carcinoma. Laryngoscope. 2015; 125 ( 7 ): 1613 - 1618. 10.1002/lary.25140
69. Url C , Schartinger VH , Riechelmann H , et al. Radiological detection of extracapsular spread in head and neck squamous cell carcinoma (HNSCC) cervical metastases. Eur J Radiol. 2013; 82 ( 10 ): 1783 - 1787. 10.1016/j.ejrad.2013.04.024
70. Nichols AC , Theurer J , Prisman E , et al. Randomized trial of radiotherapy versus transoral robotic surgery for oropharyngeal squamous cell carcinoma: long-term results of the ORATOR trial. J Clin Oncol. 2022; 40 ( 8 ): 866 - 875. 10.1200/JCO.21.01961
71. Ferris RL , Flamand Y , Weinstein GS , et al. Phase II randomized trial of transoral surgery and low-dose intensity modulated radiation therapy in resectable p16+ locally advanced oropharynx cancer: an ECOG-ACRIN cancer research group trial (E3311). J Clin Oncol. 2022; 40 ( 2 ): 138 - 149. 10.1200/JCO.21.01752
72. Kann BH , Aneja S , Loganadane GV , et al. Pretreatment identification of head and neck cancer nodal metastasis and extranodal extension using deep learning neural networks. Sci Rep. 2018; 8 ( 1 ): 14036. 10.1038/s41598-018-32441-y
73. Maguire PD , Neal CR , Hardy SM , Schreiber AM . Single-arm phase 2 trial of elective nodal dose reduction for patients with locoregionally advanced squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys. 2018; 100 ( 5 ): 1210 - 1216. 10.1016/j.ijrobp.2017.12.277
74. Sher DJ , Pham N-L , Shah JL , et al. Prospective phase 2 study of radiation therapy dose and volume de-escalation for elective neck treatment of oropharyngeal and laryngeal cancer. Int J Radiat Oncol Biol Phys. 2021; 109 ( 4 ): 932 - 940. 10.1016/j.ijrobp.2020.09.063
75. Tsai CJ , McBride SM , Riaz N , et al. Evaluation of substantial reduction in elective radiotherapy dose and field in patients with human papillomavirus-associated oropharyngeal carcinoma treated with definitive chemoradiotherapy. JAMA Oncol. 2022; 8 ( 3 ): 364 - 372. 10.1001/jamaoncol.2021.6416
76. Chen L , Dohopolski M , Zhou Z , et al. Attention guided lymph node malignancy prediction in head and neck cancer. Int J Radiat Oncol Biol Phys. 2021; 110 ( 4 ): 1171 - 1179. 10.1016/j.ijrobp.2021.02.004
77. Chen L , Zhou Z , Sher D , et al. Combining many-objective radiomics and 3D convolutional neural network through evidential reasoning to predict lymph node metastasis in head and neck cancer. Phys Med Biol. 64 ( 7 ): 075011. 10.1088/1361-6560/ab083a
78. Dohopolski M , Chen L , Sher D , Wang J . Predicting lymph node metastasis in patients with oropharyngeal cancer by using a convolutional neural network with associated epistemic and aleatoric uncertainty. Phys Med Biol. 2020; 65 ( 22 ): 225002. 10.1088/1361-6560/abb71c
79. Donzé JD , Williams MV , Robinson EJ , et al. International validity of the HOSPITAL score to predict 30-day potentially avoidable HOSPITAL readmissions. JAMA Intern Med. 2016; 176 ( 4 ): 496 - 502. 10.1001/jamainternmed.2015.8462
80. Frizzell JD , Liang L , Schulte PJ , et al. Prediction of 30-day all-cause readmissions in patients hospitalized for heart failure: comparison of machine learning and other statistical approaches. JAMA Cardiol. 2017; 2 ( 2 ): 204 - 209. 10.1001/jamacardio.2016.3956
81. Nguyen OK , Makam AN , Clark C , et al. Predicting all-cause readmissions using electronic health record data from the entire hospitalization: model development and comparison. J Hosp Med. 2016; 11 ( 7 ): 473 - 480. 10.1002/jhm.2568
82. Tevis SE , Weber SM , Kent KC , Kennedy GD . Nomogram to predict postoperative readmission in patients who undergo general surgery. JAMA Surg. 2015; 150 ( 6 ): 505 - 510. 10.1001/jamasurg.2014.4043
83. Hong JC , Niedzwiecki D , Palta M , Tenenbaum JD . Predicting emergency visits and hospital admissions during radiation and chemoradiation: an internally validated pretreatment machine learning algorithm. JCO Clin Cancer Inform. 2018; 2: 1 - 11. 10.1200/CCI.18.00037
84. Hong JC , Eclov NCW , Stephens SJ , Mowery YM , Palta M . Implementation of machine learning in the clinic: challenges and lessons in prospective deployment from the system for high intensity evaluation during radiation therapy (SHIELD-RT) randomized controlled study. BMC Bioinformatics. 2022; 23 ( suppl 12 ): 408. 10.1186/s12859-022-04940-3
85. Glide-Hurst CK , Lee P , Yock AD , et al. Adaptive radiation therapy (ART) strategies and technical considerations: a state of the ART review from NRG oncology. Int J Radiat Oncol Biol Phys. 2021; 109 ( 4 ): 1054 - 1075. 10.1016/j.ijrobp.2020.10.021
86. Jia X , Rong Y , Wu Q , et al. NRG oncology assessment of artificial intelligence for automatic treatment planning in radiation therapy clinical trials: present and future. Int J Radiat Oncol Biol Phys. 2025. 10.1016/j.ijrobp.2025.03.045
87. Rong Y , Chen Q , Fu Y , et al. NRG oncology assessment of artificial intelligence deep learning-based auto-segmentation for radiation therapy: current developments, clinical considerations, and future directions. Int J Radiat Oncol Biol Phys. 2024; 119 ( 1 ): 261 - 280. 10.1016/j.ijrobp.2023.10.033
88. Xiao Y , Benedict S , Cui Y , et al. Embracing the future of clinical trials in radiation therapy: an NRG oncology CIRO technology retreat whitepaper on pioneering technologies and AI-driven solutions. Int J Radiat Oncol Biol Phys. 2025; 122 ( 2 ): 443 - 457. 10.1016/j.ijrobp.2025.01.006

Citation

Virk J, Zhu ;S, Sim ;AJ, Ma JSJ, ;2*. Recent Trends of Artificial Intelligence in Radiation Oncology: A Narrative Review of Prospective Studies. Appl Radiat Oncol. 2025;(3):1 - 13.
doi:10.37549/ARO-D-25-0004

August 1, 2025