Improving the Quality of Care in Radiation Oncology using Artificial Intelligence

Radiation therapy is a complex process involving multiple professionals and steps from simulation to treatment planning to delivery, and these procedures are prone to error. Additionally, the imaging and treatment delivery equipment in radiotherapy is highly complex and interconnected and represents another risk point in the quality of care. Numerous quality assurance tasks are carried out to ensure quality and to detect and prevent potential errors in the process of care. Recent developments in artificial intelligence provide potential tools to the radiation oncology community to improve the efficiency and performance of quality assurance efforts. Targets for artificial intelligence enhancement include the quality assurance of treatment plans, target and tissue structure delineation used in the plans, delivery of the plans and the radiotherapy delivery equipment itself. Here we review recent developments of artificial intelligence applications that aim to improve quality assurance processes in radiation therapy and discuss some of the challenges and limitations that require further development work to realise the potential of artificial intelligence for quality assurance.     

Preparing to sit the Royal Australia and New Zealand College of Radiologists Faculty of Radiation Oncology Fellowship Part 2 examination: The value of a workshop including practice and feedback

A workshop has been held annually to help prepare candidates to sit the Royal Australia and New Zealand College of Radiologists Part 2 Faculty of Radiation Oncology examination. This study examined the value of such a course and its component parts and assessed attendees’ learning environments. We collected detailed information from participants before and after the training workshop in 2005. A specific feature of this workshop included the use of an examination technique feedback form to facilitate the provision of systematic and comprehensive feedback to individual candidates after mock examination. Participants completed course evaluation forms and a learning environment survey. There were 22 candidate participants. The course and its components of this course were perceived very positively – including the examination technique feedback forms and written advice. Only three of the 24 questions regarding the registrars learning environment had less than 80% favourable responses – two of these questions related to workload. The course design described seems reasonably satisfactory in that it included the components ranked most highly by candidates. We also identified a number of variations that may be useful for future workshops. Although learning environments were generally good, we identified a perceived problem with workloads affecting a significant number of registrars.     

Quality assurance in a Radiation Oncology Unit: The Chart Round experience

Quality assurance ensures that planned treatments eventuate. Programmes must include feedback loops to promptly correct any shortfall in predetermined standards. In March 1999, a weekly Chart Round was introduced to verify that certain items relevant to quality care were being completed for patients of the Head and Neck Radiotherapy Unit at the Peter MacCallum Cancer Institute. The experience was reviewed after 1 year and it was found that the initiation of Chart Rounds has assisted in raising the level of item completion from 80% to 99% in similar groups of patients treated before and after the initiation of the Chart Round. Initiation of the Chart Round has also provided a useful forum for in‐house peer‐review, education and effective real‐time communication between medical and allied health personnel, all of which has further added to the quality of patient care.     

Comparison of prostate set‐up accuracy and margins with off‐line bony anatomy corrections and online implanted fiducial‐based corrections

The aim of the study was to determine prostate set‐up accuracy and set‐up margins with off‐line bony anatomy‐based imaging protocols, compared with online implanted fiducial marker‐based imaging with daily corrections. Eleven patients were treated with implanted prostate fiducial markers and online set‐up corrections. Pretreatment orthogonal electronic portal images were acquired to determine couch shifts and verification images were acquired during treatment to measure residual set‐up error. The prostate set‐up errors that would result from skin marker set‐up, off‐line bony anatomy‐based protocols and online fiducial marker‐based corrections were determined. Set‐up margins were calculated for each set‐up technique using the percentage of encompassed isocentres and a margin recipe. The prostate systematic set‐up errors in the medial–lateral, superior–inferior and anterior–posterior directions for skin marker set‐up were 2.2, 3.6 and 4.5 mm (1 standard deviation). For our bony anatomy‐based off‐line protocol the prostate systematic set‐up errors were 1.6, 2.5 and 4.4 mm. For the online fiducial based set‐up the results were 0.5, 1.4 and 1.4 mm. A prostate systematic error of 10.2 mm was uncorrected by the off‐line bone protocol in one patient. Set‐up margins calculated to encompass 98% of prostate set‐up shifts were 11–14 mm with bone off‐line set‐up and 4–7 mm with online fiducial markers. Margins from the van Herk margin recipe were generally 1–2 mm smaller. Bony anatomy‐based set‐up protocols improve the group prostate set‐up error compared with skin marks; however, large prostate systematic errors can remain undetected or systematic errors increased for individual patients. The margin required for set‐up errors was found to be 10–15 mm unless implanted fiducial markers are available for treatment guidance.     

Assessment of a daily online implanted fiducial marker‐guided prostate radiotherapy process

The aims of this study were to investigate whether intrafraction prostate motion can affect the accuracy of online prostate positioning using implanted fiducial markers and to determine the effect of prostate rotations on the accuracy of the software‐predicted set‐up correction shifts. Eleven patients were treated with implanted prostate fiducial markers and online set‐up corrections. Orthogonal electronic portal images were acquired to determine couch shifts before treatment. Verification images were also acquired during treatment to assess whether intrafraction motion had occurred. A limitation of the online image registration software is that it does not allow for in‐plane prostate rotations (evident on lateral portal images) when aligning marker positions. The accuracy of couch shifts was assessed by repeating the registration measurements with separate software that incorporates full in‐plane prostate rotations. Additional treatment time required for online positioning was also measured. For the patient group, the overall postalignment systematic prostate errors were less than 1.5 mm (1 standard deviation) in all directions (range 0.2–3.9 mm). The random prostate errors ranged from 0.8 to 3.3 mm (1 standard deviation). One patient exhibited intrafraction prostate motion, resulting in a postalignment prostate set‐up error of more than 10 mm for one fraction. In 14 of 35 fractions, the postalignment prostate set‐up error was greater than 5 mm in the anterior–posterior direction for this patient. Maximum prostate rotations measured from the lateral images varied from 2° to 20° for the patients. The differences between set‐up shifts determined by the online software without in‐plane rotations to align markers, and with rotations applied, was less than 1 mm (root mean square), with a maximum difference of 4.1 mm. Intrafraction prostate motion was found to reduce the effectiveness of the online set‐up for one of the patients. A larger study is required to determine the magnitude of this problem for the patient population. The inability in the current software to incorporate in‐plane prostate rotations is a limitation that should not introduce large errors, provided that the treatment isocentre is positioned near the centre of the prostate.     

全面的调强适形放射治疗质量保证概念

调强适形放射治疗的质量保证是一个重要的研究课题.其内容非常广泛,文章主要介绍全面的调强适形放射治疗质量保证的概念.     

Faculty of Radiation Oncology 2010 workforce survey

Introduction: This paper outlines the key results of the Faculty of Radiation Oncology 2010 workforce survey and compares these results with earlier data. Methods: The workforce survey was conducted in mid‐2010 using a custom‐designed 17‐question survey. The overall response rate was 76%. Results: The majority of radiation oncologist respondents were male (n = 212, 71%), but the majority of trainee respondents were female (n = 59, 52.7%). The age range of fellows was 32–92 years (median: 47 years; mean: 49 years) and that of trainees was 27–44 years (median: 31 years; mean: 31.7 years). Most radiation oncologists worked at more than one practice (average: two practices). The majority of radiation oncologists worked in the public sector (n = 169, 64.5%), with some working in ‘combination’ of public and private sectors (n = 65, 24.8%) and a minority working in the private sector only (n = 28, 10.7%). The hours worked per week ranged from 1 to 85 (mean: 44 h; median: 45 h) for radiation oncologists, while for trainees the range was 16–90 (mean: 47 h; median: 45 h). The number of new cases seen in a year ranged from 1 to 1100 (mean: 275; median: 250). Most radiation oncologists considered themselves generalists with a preferred sub‐specialty (43.3%) or specialists (41.9%), while a minority considered themselves as generalists (14.8%). Conclusions: There are a relatively large and increasing number of radiation oncologists and trainees compared with previous years. The excessive workloads evident in previous surveys appear to have diminished. However, further work is required on assessing the impact of ongoing feminisation and sub‐specialisation.     

Faculty of Radiation Oncology survey of work practices

The aim of the present paper was to determine the current working conditions of practising radiation oncologists (RO) in Australasia and their attitudes towards their work and work environment. The authors were requested by the Faculty of Radiation Oncology to conduct a survey of Fellows’ work conditions and attitudes. The need for such a survey arose from a workshop of the Faculty held in Sydney in 1995, to determine future directions of the Faculty. Issues of potential interest were identified at the workshop and supplemented by the authors into a survey consisting of both directed and open questions. Respondents were free to remain anonymous. An address list of RO was supplied by the Royal Australian and New Zealand College of Radiologists (RANZCR). Two mailouts were performed to increase the response rate. The survey was completed during the second half of 1996 and analysed in 1997. The response rate was 79% (63–100% according to state/country). The age range was 30–69 years (median: 43 years; mean: 44 years), and 78% of the respondents were male. A public centre was identified as the sole or main place of work for 84% of respondents. The number of RO per practice varied from one to 25 (median: 5). The estimated hours worked per week ranged from 20 to 79 (mean: 52 h; median: 50 h). There were significant differences in allocation of hours between public and private (more clinical hours for private (P = 0.008), more teaching hours for public (P = 0.007)) but no difference in total hours. The responses for clinical work profile were: ‘general’ 39%, ‘largely subspecialty’ 37% and both 2% (23% did not respond). The proportion whose practice was ‘largely subspecialty’ differed between public and private (53%vs 13%, respectively; P = 0.06), and varied according to the number of RO in the practice (62% for > five RO vs 35% for ≤ five RO, P = 0.03). The need for subspecialization for the treatment of common tumours (breast, gynaecological etc.) was held by 78% of respondents and was related to the number of RO in the practice (100% for > five RO vs 81% for ≤ five RO, P = 0.002). The number of new patients seen per RO in 1995 ranged from 0 to 700 (mean: 342; median: 350), with a significant difference between the mean numbers seen in public versus private settings (331 vs 409, respectively; P = 0.008). Administration of cytotoxics was supervised by 36% of RO across Australasia but was dependent strongly on the state/country of practice (P = 0.0002). The current and preferred roles in clinical management were scored as means of 7.9 and 8.7, respectively, on a linear scale from 1 (mainly technical role) to 10 (significant role in overall management). A total of 40% of RO were content with their current time allocation; the remainder generally wished to reduce clinical time and increase self-education and/or clinical research. Most indicated that they wanted some ‘protected time’. Most were content to be called a ‘radiation oncologist’ and wished to retain this title; the most common alternative was ‘clinical oncologist’. Much of the general comment related to perceived loss of control of clinical and academic aspects of oncology to other specialties. The present survey is the first to define current work practices and to explore attitudes to work. It is clear that many RO consider clinical workloads to be excessive, to the exclusion of other work-related activities. There was a strong feeling that significant changes will be required if RO are to maintain the role in cancer management for which they are trained.     

Survey of Radiation Oncology Centres in Australia: Report of the Radiation Oncology Treatment Quality Program*

One of the first steps towards the development of a comprehensive quality program for radiation oncology in Australia has been a survey of practice. This paper reports on the results of the survey that should inform the development of standards for radiation oncology in Australia. A questionnaire of 108 questions spanning aspects of treatment services, equipment, staff, infrastructure and available quality systems was mailed to all facilities providing radiation treatment services in Australia (n = 45). Information of 42 sites was received by June 2006 providing data on 113 operational linear accelerators of which approximately 2/3 are equipped with multi‐leaf collimators. More than 75% of facilities were participating in a formal quality assurance (QA) system, with 63% following a nationally or internationally recognised system. However, there was considerable variation in the availability of policies and procedures specific to quality aspects, and the review of these. Policies for monitoring patient waiting times for treatment were documented at just 71% of all facilities. Although 85% of all centres do, in fact, monitor machine throughput, the number and types of efficiency measures varied markedly, thereby limiting the comparative use of these results. Centres identified workload as the single most common factor responsible for limiting staff involvement in both QA processes and clinical trial participation. The data collected in this ‘snapshot’ survey provide a unique and comprehensive baseline for future comparisons and evaluation of changes.