Applications In Radiation Oncology Research Articles

Pilot Study Assessing Student Interest in Combined Residency Training Pathways With Radiation Oncology

Most medical students have limited exposure to radiation oncology (RO) before deciding which specialty to choose for residency. This may limit the number and diversity of RO applicants. The purpose of this study was to determine students' views on a combined pathway program of RO with internal medicine (IM), as well as other related medical specialties, as a potential means of overcoming barriers to interest in RO and the early decision point to solo training in a highly specialized field. In July to August 2022, all 299 United States student and postgraduate year 1 members of the American Society for Radiation Oncology were sent an electronic survey assessing their views on advantages and disadvantages of a combined IM/RO training pathway, and interest in several options of combined pathway programs. Eighty participants completed the survey (response rate 27%). Thirty-four (43%) were very or extremely interested in IM/RO residency (median Likert-type rating 3, IQR 2-4). The most important potential advantages of an IM/RO pathway included greater flexibility in employment options (n = 51, 64%), enhanced general medical knowledge to facilitate ambitions in other career pathways (n = 46, 58%), improved patient care (n = 43, 54%), and having a pathway for combined hematologic/oncology and RO board certification (n = 46, 58%). In comparison to IM/RO, participants were significantly more interested in a combined RO and hematology/oncology program (median Likert-type rating 5, IQR 5-5, P = .005). Among the subgroup of 26 survey participants who believed it less likely they would apply for RO residency, 18 (69%) thought an IM/RO pathway would increase their interest in RO (median Likert-type rating 4, IQR 3-5). Interest in IM/RO did not differ by gender, race, or ethnicity. Combined training pathways involving RO were viewed positively by survey respondents, and may be particularly appealing to those less committed to a career in RO. Further research will help guide recommendations for the creation of these programs.

MR-Linac Dosimetry: Current Approaches and Challenges

The use of MR-integrated linacs (MR-linac) in clinical radiation oncology applications are becoming more prevalent. However, given the novel and radically different designs of these systems from conventional linacs, current radiation dosimetry protocols for high energy photons are no longer appropriate for use in modern MR-linac systems: First, given the presence of the magnetic field and the linac design, traditional reference conditions defined by previous high energy photon dosimetry protocols cannot be met; Second, the presence of the strong magnetic field can affect the performance of conventional equipment used for dosimetry. In this manuscript, we describe some of the challenges faced in radiation dosimetry in external MR-guided radiotherapy delivery systems, summarize some of the publications in this area, and finally discuss the early work by the American Association of physicists in Medicine (AAPM) Task Group 351 which is mandated with producing a protocol for reference dosimetry in MR-linac units.

Automation and artificial intelligence in radiation therapy treatment planning.

Automation and artificial intelligence (AI) is already possible for many radiation therapy planning and treatment processes with the aim of improving workflows and increasing efficiency in radiation oncology departments. Currently, AI technology is advancing at an exponential rate, as are its applications in radiation oncology. This commentary highlights the way AI has begun to impact radiation therapy treatment planning and looks ahead to potential future developments in this space. Historically, radiation therapist's (RT's) role has evolved alongside the adoption of new technology. In Australia, RTs have key clinical roles in both planning and treatment delivery and have been integral in the implementation of automated solutions for both areas. They will need to continue to be informed, to adapt and to transform with AI technologies implemented into clinical practice in radiation oncology departments. RTs will play an important role in how AI-based automation is implemented into practice in Australia, ensuring its application can truly enable personalised and higher-quality treatment for patients. To inform and optimise utilisation of AI, research should not only focus on clinical outcomes but also AI's impact on professional roles, responsibilities and service delivery. Increased efficiencies in the radiation therapy workflow and workforce need to maintain safe improvements in practice and should not come at the cost of creativity, innovation, oversight and safety.

Pilot Study Assessing Student Interest in Combined Residency Training Pathways with Radiation Oncology

Radiation oncology (RO) is a small, highly specialized field, which most medical students have limited exposure to before having to decide what specialty to apply to for their graduate medical education (GME). This may ultimately limit the number and diversity of RO applicants. The purpose of this pilot study was to determine students' views on a combined pathway program between RO with internal medicine (IM), as well as other related medical specialties, as a potential means of overcoming some of these barriers to student interest in RO and the early decision point to solo training in RO. An electronic survey was sent to 299 United States student members of the American Society for Radiation Oncology (ASTRO) between July-August 2022. Multiple choice and Likert-type questions were used to assess students' views on the potential advantages and disadvantages of a combined IM/RO training pathway, and interest in this potential combined pathway as well as other combined pathway programs with RO and other related subspecialties. The Mann-Whitney U and Kruskal-Wallis tests were used to compare the responses of subgroups of students stratified according to their gender identity, race, or ethnicity. The Wilcoxon signed rank test was used to compare all participants' interest in different combined pathway programs compared to RO/IM. Eighty participants completed the survey (response rate 27%). Thirty-four participants (43%) were either very or extremely interested in pursuing a combined IM/RO residency (median Likert-type rating 3, IQR 2-4). The duration of a combined IM/RO training considered most preferable or ideal was five years (n = 57, 71%) or six years (n = 21, 26%). The most important potential advantages of a combined IM/RO pathway included greater flexibility in employment options (n = 51, 64%), enhanced general medical knowledge to facilitate ambitions in other career pathways (n = 46, 58%), improved patient care (n = 43, 54%), and having a pathway for combined hematologic/oncology and RO board certification (n = 46, 58%). In comparison to IM/RO, participants were significantly more interested in a combined RO and hematology/oncology program (median Likert-type rating 5, IQR 5-5, p = 0.005). Among the subgroup of 26 students who believed that it was less likely that they would be applying for RO residency, 18 (69%) believed that the availability of an IM/RO pathway would increase their interest in applying to RO (median Likert-type rating 4, IQR 3-5). Interest in IM/RO did not differ by gender, race, or ethnicity. Combined training pathways involving RO were viewed positively by most survey respondents, comprised of student members of ASTRO. Such programs may be particularly appealing to those students who are interested, but less committed to pursuing, a career in RO. Further research will help guide recommendations for the creation of these programs.

Cancer Informatics for Cancer Centers: Sharing Ideas on How to Build an Artificial Intelligence-Ready Informatics Ecosystem for Radiation Oncology.

In August 2022, the Cancer Informatics for Cancer Centers brought together cancer informatics leaders for its biannual symposium, Precision Medicine Applications in Radiation Oncology, co-chaired by Quynh-Thu Le, MD (Stanford University), and Walter J. Curran, MD (GenesisCare). Over the course of 3 days, presenters discussed a range of topics relevant to radiation oncology and the cancer informatics community more broadly, including biomarker development, decision support algorithms, novel imaging tools, theranostics, and artificial intelligence (AI) for the radiotherapy workflow. Since the symposium, there has been an impressive shift in the promise and potential for integration of AI in clinical care, accelerated in large part by major advances in generative AI. AI is now poised more than ever to revolutionize cancer care. Radiation oncology is a field that uses and generates a large amount of digital data and is therefore likely to be one of the first fields to be transformed by AI. As experts in the collection, management, and analysis of these data, the informatics community will take a leading role in ensuring that radiation oncology is prepared to take full advantage of these technological advances. In this report, we provide highlights from the symposium, which took place in Santa Barbara, California, from August 29 to 31, 2022. We discuss lessons learned from the symposium for data acquisition, management, representation, and sharing, and put these themes into context to prepare radiation oncology for the successful and safe integration of AI and informatics technologies.

Artificial Intelligence in Head and Neck Cancer: A Systematic Review of Systematic Reviews.

Several studies have emphasized the potential of artificial intelligence (AI) and its subfields, such as machine learning (ML), as emerging and feasible approaches to optimize patient care in oncology. As a result, clinicians and decision-makers are faced with a plethora of reviews regarding the state of the art of applications of AI for head and neck cancer (HNC) management. This article provides an analysis of systematic reviews on the current status, and of the limitations of the application of AI/ML as adjunctive decision-making tools in HNC management. Electronic databases (PubMed, Medline via Ovid, Scopus, and Web of Science) were searched from inception until November 30, 2022. The study selection, searching and screening processes, inclusion, and exclusion criteria followed the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. A risk of bias assessment was conducted using a tailored and modified version of the Assessment of Systematic Review (AMSTAR-2) tool and quality assessment using the Risk of Bias in Systematic Reviews (ROBIS) guidelines. Of the 137 search hits retrieved, 17 fulfilled the inclusion criteria. This analysis of systematic reviews revealed that the application of AI/ML as a decision aid in HNC management can be thematized as follows: (1) detection of precancerous and cancerous lesions within histopathologic slides; (2) prediction of the histopathologic nature of a given lesion from various sources of medical imaging; (3) prognostication; (4) extraction of pathological findings from imaging; and (5) different applications in radiation oncology. In addition, the challenges in implementation of AI/ML models for clinical evaluations include the lack of standardized methodological guidelines for the collection of clinical images, development of these models, reporting of their performance, external validation procedures, and regulatory frameworks. At present, there is a paucity of evidence to suggest the adoption of these models in clinical practice due to the aforementioned limitations. Therefore, this manuscript highlights the need for development of standardized guidelines to facilitate the adoption and implementation of these models in the daily clinical practice. In addition, adequately powered, prospective, randomized controlled trials are urgently needed to further assess the potential of AI/ML models in real-world clinical settings for the management of HNC.

Photon-limited Cherenkov imaging of radiation therapy dose.

Cherenkov imaging is a unique verification tool that could provide both dosimetric and tissue functional information during radiation therapy. However, the number of interrogated Cherenkov photons in tissue is always limited and tangled with stray radiation photons, severely frustrating the measurement the signal-to-noise ratio (SNR). As such, here, a noise-robust photon-limited imaging technique is proposed by comprehensively exploiting the physical rationale of low-flux Cherenkov measurements together with the spatial correlations of the objects. Validation experiments confirmed that the Cherenkov signal could be promisingly recovered with high SNR by irradiating at as few as one x ray pulse from a linear accelerator (10 mGy dose), and the Cherenkov excited luminescence imaging depth can be extended by >100% on average, for most concentrations of phosphorescent probe. This approach demonstrates that improved applications in radiation oncology could be seen when signal amplitude, noise robustness, and temporal resolution are comprehensively considered in the image recovery process.

4DCT is long overdue for improvement.

4DCT is long overdue for improvement.

Automatic quality assurance of radiotherapy treatment plans using Bayesian networks: A multi-institutional study.

Artificial intelligence applications in radiation oncology have been the focus of study in the last decade. The introduction of automated and intelligent solutions for routine clinical tasks, such as treatment planning and quality assurance, has the potential to increase safety and efficiency of radiotherapy. In this work, we present a multi-institutional study across three different institutions internationally on a Bayesian network (BN)-based initial plan review assistive tool that alerts radiotherapy professionals for potential erroneous or suboptimal treatment plans. Clinical data were collected from the oncology information systems in three institutes in Europe (Maastro clinic - 8753 patients treated between 2012 and 2020) and the United States of America (University of Vermont Medical Center [UVMMC] - 2733 patients, University of Washington [UW] - 6180 patients, treated between 2018 and 2021). We trained the BN model to detect potential errors in radiotherapy treatment plans using different combinations of institutional data and performed single-site and cross-site validation with simulated plans with embedded errors. The simulated errors consisted of three different categories: i) patient setup, ii) treatment planning and iii) prescription. We also compared the strategy of using only diagnostic parameters or all variables as evidence for the BN. We evaluated the model performance utilizing the area under the receiver-operating characteristic curve (AUC). The best network performance was observed when the BN model is trained and validated using the dataset in the same center. In particular, the testing and validation using UVMMC data has achieved an AUC of 0.92 with all parameters used as evidence. In cross-validation studies, we observed that the BN model performed better when it was trained and validated in institutes with similar technology and treatment protocols (for instance, when testing on UVMMC data, the model trained on UW data achieved an AUC of 0.84, compared with an AUC of 0.64 for the model trained on Maastro data). Also, combining training data from larger clinics (UW and Maastro clinic) and using it on smaller clinics (UVMMC) leads to satisfactory performance with an AUC of 0.85. Lastly, we found that in general the BN model performed better when all variables are considered as evidence. We have developed and validated a Bayesian network model to assist initial treatment plan review using multi-institutional data with different technology and clinical practices. The model has shown good performance even when trained on data from clinics with divergent profiles, suggesting that the model is able to adapt to different data distributions.

Artificial intelligence 101 for veterinary diagnostic imaging.

The prevalence and pervasiveness of artificial intelligence (AI) with medical images in veterinary and human medicine is rapidly increasing. This article provides essential definitions of AI with medical images with a focus on veterinary radiology. Machine learning methods common in medical image analysis are compared, and a detailed description of convolutional neural networks commonly used in deep learning classification and regression models is provided. A brief introduction to natural language processing (NLP) and its utility in machine learning is also provided. NLP can economize the creation of "truth-data" needed when training AI systems for both diagnostic radiology and radiation oncology applications. The goal of this publication is to provide veterinarians, veterinary radiologists, and radiation oncologists the necessary background needed to understand and comprehend AI-focused research projects and publications.

Effects of the Virtual Match Process on Radiation Oncology Applicants' Interview Experience and their Approach to the Match

<h3>Purpose/Objective(s)</h3> COVID drastically changed the 2020-2021 match process experience by transitioning from in-person interviews to virtual interview processes. We evaluated how the virtual match influenced radiation oncology applicants interview selection and interview experiences. <h3>Materials/Methods</h3> A list of 140 radiation oncology applicants was curated from events throughout the application process. The optional survey was sent via e-mail to the applicants, followed by two reminder e-mails. Collected demographics were medical education, additional degrees, gender, race, ethnicity, and couples match status. The number of programs applied, interviews received, interviews attended, projected additional interviews canceled for an in-person process was recorded. Information was collected regarding preferred virtual interview experiences including information such length of interviews, number of interviews, interview format, and preferred method of obtaining program information. <h3>Results</h3> Of the 140 contacted applicants, 62 completed the survey. Demographics of the applicants is listed in the table below. For medical training, the applicants contained 50 MDs, 5 DOs, and 7 IMGs. Respondents were 66.1% Male, 32.3% Female, and 1.6% Non-Binary/Non-Conforming. Additional degrees or research experiences from the applicants included 16 research years, 15 PhDs, 9 MS, 3 MPH, and 1 M.Ed. Three applicants couples matched. The number of applications submitted, interviews received, and interviews attended are listed in the table below. When asked if interviews had been in person, most estimated that they would have canceled additional interviews with 6 applicants (9.7%) canceling 0 interviews, 30 applicants (48.4%) canceling 1-5, 17 applicants (27.4%) canceling 6-10, 5 applicants (8.1%) canceling 11-15, 3 applicants (4.8%) canceling 16-20, and 1 applicant (1.6%) canceling >20 interviews. The preferred interview length for most applicants was 15 minutes (27, 43.5%) or 20 minutes (30, 48.4%), while most applicants preferred interview days with 4-6 interviews (47, 75.8). One-on-one or two-on-one interviews were rated as preferred experiences over panel interviews. The highest-rated break method was 5–10-minute breaks between interviews with cameras off. <h3>Conclusion</h3> The virtual NRMP interview experience was overall a positive experience for radiation oncology applicants, allowing applicants to attend more virtual interviews than if they were attending interviews in person. Additional analyses are needed to determine if increased number of interviews per applicant affected unfilled residency positions.

9-Year Analysis of Trends in the Radiation Oncology Residency Match

<h3>Purpose/Objective(s)</h3> To assess the trends in Radiation Oncology residency applicants, programs, and positions from 2013 to 2021. <h3>Materials/Methods</h3> United States Radiation Oncology residency program and applicant data were obtained from the National Resident Matching Program, Main Residency Match, Charting Outcomes, and Electronic Residency Application Service (ERAS) reports from 2013 to 2021. <h3>Results</h3> The total number of U.S. MD senior Radiation Oncology applicants decreased from 339 in 2017 to 193 in 2021. This corresponded to a decrease in the mean applications received per program from 167.0 to 100.2 applications. In contrast, the number of applicants who were not U.S. MD Seniors (DO & IMG) increased from 38 (11.2%) in 2017 to 106 (54.9%) in 2021. The total number of positions offered (PGY-1 & PGY-2) increased steadily from 183 in 2013 to 207 in 2019, after which there was a sharp decline to 189 and 188 in 2020 and 2021, respectively. The mean number of Radiation Oncology-specific applications per applicant decreased from 39.1 in 2017 to 20.7 in 2021. The number of positions per U.S. senior remained steady from 2014 to 2018 at 1.0. However, this ratio has increased each year since 2018 and reached 1.9 positions per U.S. MD senior applicant in 2021. The mean number of unfilled programs (PGY-1 & PGY-2) prior to Supplemental Offer and Acceptance Program (SOAP) from 2013 to 2018 was 4.8 programs (5.4% of all programs) compared to a now-increased mean of 25.0 unfilled programs (27.7% of all programs) from 2019 to 2021. From 2014 to 2020, the match rate for Radiation Oncology residency applicants increased from 89% to 99%. <h3>Conclusion</h3> Despite an increase in interest among DO and IMG applicants and recent efforts to decrease the number of available Radiation Oncology residency positions, a continued rise in the number of positions offered relative to U.S. applicants has led to a growing number of unfilled programs. As a result, measures should be taken to increase exposure to Radiation Oncology during medical school and to make our field more appealing to medical students who may be considering Radiation Oncology as their ultimate career choice.

Fifteen-Year Trends in Canadian Female Medical Student Interest in Radiation Oncology

<h3>Purpose/Objective(s)</h3> The proportion of female radiation oncology residents in Canada remains disproportionately low. The objective of this study was to examine 15 years of resident match data to assess trends in gender-specific medical student interest and match characteristics in radiation oncology (RO). <h3>Materials/Methods</h3> Public Canadian Residency Matching Service (CaRMS) data between 2007-21 were compiled. Gender-based analysis focused on Canadian medical graduates (CMGs) ranking RO as their first-choice discipline in the first iteration of the match. Gender data were not available for applicants and match characteristics for second iteration outcomes. Pearson's chi-square test was used to compare 15-year average female medical student interest in RO versus all training programs. Paired t-test was used to compare total, female, and male applicant numbers over five-year increments (2007-11; 2012-16; 2017-21). Statistical analysis was performed using statistical software. <h3>Results</h3> Total applicants to RO programs ranged between 24-51 per year over the study period. The mean number of applicants ranking RO as their first-choice discipline significantly decreased from 24 (range: 20-27) in 2007-11 to 16 (12-18) in 2012-16 (p<0.01), with an increase to 22 (9-28) in 2017-21 (p<0.05). The 15-year average of female RO applicants was 9 per year, ranging from 2/9 (22%) in 2017 to 12/18 (67%) applicants in 2015. Female applicant numbers were lower in 2012-16 (mean 7; 4-12) compared to 2007-11 (mean 14; 11-16; p<0.01). Female applicant numbers in 2017-21 (mean 7; 2-10) were comparable to 2012-16 (p=0.46), but remained lower than 2007-11 (p<0.01). Male applicant numbers were similar (p=0.12) in 2007-11 (mean 10; 9-12) and 2012-16 (mean 8; 5-11), but increased in 2017-21 (mean= 15; 7-20) compared to 2007-11 (p=0.03) and 2012-16 (p=0.01). Overall, 308 CMG applicants ranked RO as their first choice over the past 15 years; 141 females (46%) and 167 males (54%). Over the same period, 56% (24472/43667) of CMG applicants who participated in the CaRMS match were female. Female CMG application rates to RO were significantly lower compared to application rates to all specialties combined (p<0.01). <h3>Conclusion</h3> Despite increasing male applicant numbers, the number of RO-preferring female medical students remains disproportionately low. Female medical student interest in RO over the last ten years has also fallen significantly from 2007-11 levels. These findings highlight the need for early exposure of female medical students and improved mentorship efforts. A deeper understanding of factors influencing female medical students' career selection and CaRMS ranking decisions is required to improve female representation in RO. These data have led to a special interest group proposal for women in RO, now being considered by the Canadian Association of Radiation Oncology.

The landscape of digital resources in radiation oncology.

In recent years, the number of educational medical resources accessible to residents and practicing radiation oncologists online has grown exponentially to include discussion boards, wikis, videos, podcasts, journal clubs, online communities, and interactive experiences to augment medical education. In this review, we identify, catalog, and critically evaluate educational websites, smartphone applications, web-based multimedia, and podcasts for radiation oncologists. Literature searches were conducted over a 2-month period (April to May 2022) using OVID-MEDLINE and PubMed with a combination of relevant search terms. Websites of relevant radiation oncology societies were reviewed for e-learning resources. Internet searches including the Google search engine, application stores, and podcast-publisher websites were conducted to identify digital resources for radiation oncology education. To ensure credibility, resources were assessed by two independent reviewers utilizing the criteria of authority, accuracy, objectivity, currency, depth, and appearance per suggested formats for evaluating digital resources in medical literature. Literature searches using OVID-MEDLINE and PubMed yielded 425 citations. Those pertinent to radiation oncology provide examples of resource development, integrations into curriculum, interactive modules, case studies, and learner experiences. The multilevel search identified 47 free digital education resources including online radiation oncology tutorials, podcasts, videos, slide sets, applications, and other interactive resources, some requiring membership or a fee for full access. The myriad online educational tools available to radiation oncology residents represent excellent resources for continuing education. This review represents the first comprehensive summary of available online education resources for radiation oncologists to guide clinicians who are increasingly reliant on digital resources, especially during the COVID-19 pandemic.

Deep learning methods for enhancing cone‐beam CT image quality toward adaptive radiation therapy: A systematic review

The use of deep learning (DL) to improve cone‐beam CT (CBCT) image quality has gained popularity as computational resources and algorithmic sophistication have advanced in tandem. CBCT imaging has the potential to facilitate online adaptive radiation therapy (ART) by utilizing up‐to‐date patient anatomy to modify treatment parameters before irradiation. Poor CBCT image quality has been an impediment to realizing ART due to the increased scatter conditions inherent to cone‐beam acquisitions. Given the recent interest in DL applications in radiation oncology, and specifically DL for CBCT correction, we provide a systematic theoretical and literature review for future stakeholders. The review encompasses DL approaches for synthetic CT generation, as well as projection domain methods employed in the CBCT correction literature. We review trends pertaining to publications from January 2018 to April 2022 and condense their major findings—with emphasis on study design and DL techniques. Clinically relevant endpoints relating to image quality and dosimetric accuracy are summarized, highlighting gaps in the literature. Finally, we make recommendations for both clinicians and DL practitioners based on literature trends and the current DL state‐of‐the‐art methods utilized in radiation oncology.

Three discipline collaborative radiation therapy (3DCRT) special debate: FLASH radiotherapy needs ongoing basic and animal research before implementing it to a large clinical scale.

Three discipline collaborative radiation therapy (3DCRT) special debate: FLASH radiotherapy needs ongoing basic and animal research before implementing it to a large clinical scale.

Survival Analysis Using the Covid-death Mean-imputation (CoDMI) Algorithm: A First Clinical Application in Radiation Oncology.

To report long-term survival results after trimodal approach for locally advanced rectal cancer (LARC) in the Covid-19 era. We herein illustrate a clinical application of Covid-death mean-imputation (CoDMI) algorithm in LARC patients with Covid-19 infection. We analyzed 94 patients treated for primary LARC. Overall survival was calculated in months from diagnosis to first event (last follow-up/death). Because Covid-19 death events potentially bias survival estimation, to eliminate skewed data due to Covid-19 death events, the observed lifetime of Covid-19 cases was replaced by its corresponding expected lifetime in absence of the Covid-19 event using the CoDMI algorithm. Patients who died of Covid-19 (DoC) are mean-imputed by the Kaplan-Meier estimator. Under this approach, the observed lifetime of each DoC patient is considered as an "incomplete data" and is extended by an additional expected lifetime computed using the classical Kaplan-Meier model. Sixteen patients were dead of disease (DoD), 1 patient was DoC and 77 cases were censored (Cen). The DoC patient died of Covid-19 52 months after diagnosis. The CoDMI algorithm computed the expected future lifetime provided by the Kaplan-Meier estimator applied to the no-DoC observations as well as to the DoC data itself. Given the DoC event at 52 months, the CoDMI algorithm estimated that this patient would have died after 79.5 months of follow-up. The CoDMI algorithm leads to "unbiased" probability of overall survival in LARC patients with Covid-19 infection, compared to that provided by a naïve application of Kaplan-Meier approach. This allows for a proper interpretation/use of Covid-19 events in survival analysis. A user-friendly version of CoDMI is freely available at https://github.com/alef-innovation/codmi.

Multi-Institutional Outcomes and Perspectives From Medical Students After Virtual Radiation Oncology Electives

<h3>Purpose/Objective(s)</h3> Applications in radiation oncology residency have been on the decline in the past 3 years. This was attributed in part due to limited exposure to the field. We hypothesize that improving medical student enrollment into radiation oncology electives through virtual clerkships will increase interest and knowledge in the field. <h3>Materials/Methods</h3> IRB-approved surveys were distributed via a data collection web application to medical students who attended virtual radiation oncology clerkships offered at 7 academic institutions. The surveys consisted of a series of fixed-response, checkboxes, and open-ended questions. The open-ended responses were grouped into common themes. Surveys were analyzed with 2-sided t-tests and Wilcoxon rank test. <h3>Results</h3> Seventy-two (72) medical students from 7 institutions were sent a survey 1-7 months after completing the clerkship. Response rate was 54.1% (n = 39). 69.2% (n = 27) of respondents were MS4 and 30.8% (n = 12) were MS3 with equal distribution in gender. The median length of the elective rotation was 2 weeks (range 1 to 4). 69.2% (n = 27) of the respondents attended one virtual elective, 10.3% (n = 4) attended 2 virtual electives, and 20.5% (n = 8) more than 2 virtual electives. Majority (88.9%) of students attended the elective entirely virtually and 11.1% a combination of virtual and on-site rotation. 30.8% (n = 12) of the respondents, all of whom were MS4, planned on or applied for radiation oncology residency to an average of 59.5 (range 40-89) programs. 92% (n = 11) of these radiation oncology applicants were interviewed in at least one of the programs in which they attended a virtual away rotation. Students who applied to radiation oncology residency attended a median of 3 electives (range 2-4) while those who did not apply, a median of 1 elective (<i>P</i> < 0.001). There was a significant increase in interest in the field after the rotation among the students who applied or planned to apply (<i>P</i> = 0.002) and all respondents (<i>P</i> < 0.05). None of the MS3 respondents planned to apply to radiation oncology and 55.6% (n = 15) of MS4 respondents also did not apply or plan to apply to the field. 93% (n = 37) of respondents stated the elective was valuable for their future practice. 100 % (n = 39) of medical students indicated they were likely or most likely to recommend the virtual elective to other medical students. <h3>Conclusion</h3> Virtual radiation oncology electives have increased access to home and away rotations for medical students. The number of electives attended correlated with the interest of the student in the field and residency applications. Although a majority of medical students who took a virtual elective did not apply in radiation oncology, all recommended the elective and felt that the elective added valuable knowledge to their future practice, even outside of radiation oncology. Virtual elective platforms can improve education, awareness, exposure in radiation oncology as well as augment in-person rotations for medical students.

Reasons for Declining Applicant Numbers in Radiation Oncology From the Applicants' Perspective: Results From the Applicant Concerns and Radiation Oncology Sources Survey (ACROSS).

Applications to radiation oncology in the United States have decreased the past 3 years, resulting in unfilled residency positions (30 [14.5%] in 2019 and 35 [18.5%] in 2020). The aim of this study is to understand the concerns among radiation oncology applicants and whether these concerns may have led to a decline in applications. An Internet-based survey was e-mailed to all radiation oncology applicants participating in the 2020 National Resident Matching Program Match for whom e-mail addresses could be obtained (n=145, 78.8%). The survey was open from May to September 2020. Survey responses were received for 98 applicants out of 145 (67.6%) available applicant e-mails. Applicants' top-rated concerns were location of practice restrictions and residency expansion/oversupply of practicing physicians. Applicants were less concerned about a lack of exposure to the field of radiation oncology, competitiveness of the match, and the content of residency training. The sources of information which applicants used most, which applicants trusted most, and which promoted the most optimism among applicants were radiation oncology attending and resident physicians. The sources of information which applicants used least, which applicants trusted least, and which promoted the most pessimism among applicants were online resources. As a result of their concerns, 27 (28%) respondents strongly considered choosing a different field of medicine. Applicants identified 93 fellow graduating medical students who were interested in radiation oncology but chose not to apply; applicants reported that potential applicants chose not to apply most often due to concerns about location of practice restrictions and residency expansion/oversupply of practicing physicians. Applicant concerns about residency expansion and location of practice restrictions are prevalent and may have led to declining applications to radiation oncology. To assuage applicant concerns, radiation oncology stakeholders must address these concerns by mitigating underlying problems.

Machine learning applications in radiation oncology.

Machine learning technology has a growing impact on radiation oncology with an increasing presence in research and industry. The prevalence of diverse data including 3D imaging and the 3D radiation dose delivery presents potential for future automation and scope for treatment improvements for cancer patients. Harnessing this potential requires standardization of tools and data, and focused collaboration between fields of expertise. The rapid advancement of radiation oncology treatment technologies presents opportunities for machine learning integration with investments targeted towards data quality, data extraction, software, and engagement with clinical expertise. In this review, we provide an overview of machine learning concepts before reviewing advances in applying machine learning to radiation oncology and integrating these techniques into the radiation oncology workflows. Several key areas are outlined in the radiation oncology workflow where machine learning has been applied and where it can have a significant impact in terms of efficiency, consistency in treatment and overall treatment outcomes. This review highlights that machine learning has key early applications in radiation oncology due to the repetitive nature of many tasks that also currently have human review. Standardized data management of routinely collected imaging and radiation dose data are also highlighted as enabling engagement in research utilizing machine learning and the ability integrate these technologies into clinical workflow to benefit patients. Physicists need to be part of the conversation to facilitate this technical integration.