Enhancing the accuracy of tumor response predictions enables the development of tailored therapeutic strategies for patients with breast cancer. In this study, we developed deep radiomic models to enhance the prediction of chemotherapy response after the first treatment cycle. 18F-Fludeoxyglucose PET/CT imaging data and clinical record from 60 breast cancer patients were retrospectively obtained from the Cancer Imaging Archive. PET/CT scans were conducted at three distinct stages of treatment; prior to the initiation of chemotherapy (T1), following the first cycle of chemotherapy (T2), and after the full chemotherapy regimen (T3). The patient's primary gross tumor volume (GTV) was delineated on PET images using a 40% threshold of the maximum standardized uptake value (SUVmax). Radiomic features were extracted from the GTV based on the PET/CT images. In addition, a squeeze-and-excitation network (SENet) deep learning model was employed to generate additional features from the PET/CT images for combined analysis. A XGBoost machine learning model was developed and compared with the conventional machine learning algorithm [random forest (RF), logistic regression (LR) and support vector machine (SVM)]. The performance of each model was assessed using receiver operating characteristics area under the curve (ROC AUC) analysis, and prediction accuracy in a validation cohort. Model performance was evaluated through fivefold cross-validation on the entire cohort, with data splits stratified by treatment response categories to ensure balanced representation. The AUC values for the machine learning models using only radiomic features were 0.85(XGBoost), 0.76 (RF), 0.80 (LR), and 0.59 (SVM), with XGBoost showing the best performance. After incorporating additional deep learning-derived features from SENet, the AUC values increased to 0.92, 0.88, 0.90, and 0.61, respectively, demonstrating significant improvements in predictive accuracy. Predictions were based on pre-treatment (T1) and post-first-cycle (T2) imaging data, enabling early assessment of chemotherapy response after the initial treatment cycle. Integrating deep learning-derived features significantly enhanced the performance of predictive models for chemotherapy response in breast cancer patients. This study demonstrated the superior predictive capability of the XGBoost model, emphasizing its potential to optimize personalized therapeutic strategies by accurately identifying patients unlikely to respond to chemotherapy after the first treatment cycle.

BACKGROUND:Modern radiation therapy for breast cancer has significantly advanced with the adoption of volumetric modulated arc therapy (VMAT), offering enhanced precision and improved treatment efficiency. PURPOSE:To ensure the accuracy and precision of such complex treatments, a robust patient-specific quality assurance (PSQA) protocol is essential. This study investigates the potential of machine learning (ML) models to predict gamma passing rates (GPR), a key metric in PSQA. METHODS:A dataset comprising 863 VMAT plans was used to develop and compare seven ML models: Histogram-based gradient boosting regressor, random forest regressor, extra trees regressor, gradient boosting regressor, linear regression, AdaBoost regressor, and Multi-layer perceptron regressor. These models incorporated anatomical, dosimetric, and plan complexity features. RESULTS:Among the evaluated models, the extra trees regressor (ETR), random forest regressor (RFR), and gradient boosting regressor (GBR) demonstrated the best performance, achieving mean absolute errors (MAEs) of 0.51%, 0.52%, and 0.51%, and mean squared errors (MSEs) of 0.0051%, 0.0051%, and 0.0052%, respectively, on the validation dataset. CONCLUSIONS:This study highlights the promise of ML-based approaches in streamlining PSQA processes, thereby supporting the quality assurance of breast cancer treatments using VMAT.

Synchronous bilateral breast cancers (SBBC) present a considerable issue in external beam radiotherapy because of large fields size and large target volumes. Mono-isocentric volumetric modulated arc therapy (VMAT) appears as an appropriate irradiation technique for these types of tumors. The aim of this study was to demonstrate the utility of a 3D DVH pretreatment quality assurance program in VMAT of SBBC cases. Twenty SBBC patients who underwent radiation therapy in our department were retrospectively enrolled in this study. Fifteen patients were treated exclusively to the mammary glands. Five patients benefited from a dose boost on the tumor bed (60Gy). Nine patients were irradiated on the supraclavicular nodes (50Gy). This dose was delivered in 25 fractions and integrated boost was used when appropriate. Depending on the complexity of the treatment plans; 2 or 4 arcs VMAT plans were used in a mono-isocentric technique. The patient specific quality assurance (PSQA) was evaluated using COMPASS measured data, COMPASS reconstructed (CR) and COMPASS computed (CC) dose compared to treatment planning system (TPS) dose. Clinical evaluation was based on DVH metrics for target volumes and organ at risks. The maximum average dose deviation between TPS, CC, and CR was below 3%. The paired t-test between TPS, CC, and CR shows a strong agreement (p < 0.001). The 3DVH dose distribution comparison between TPS and COMPASS were also performed with good gamma score for global analysis. COMPASS was successfully evaluated as a 3DVH pretreatment system for SBBC despite the large fields size and complex target volumes. It allows the verification of the plan in 3D patient anatomy and the evaluation of dose discrepancies.

CT imaging plays a crucial role in medical diagnostics, but it comes with radiation exposure risks. Diagnostic Reference Levels (DRLs) are essential tools for optimizing imaging protocols and ensuring patient safety. This study evaluates the effective dose (ED) from abdominal, chest, and skull CT exams using advanced MDCT systems across ten radiological centers in southern Nigeria, with the primary objective of assessing the alignment of local imaging protocols with international reference standards to optimize patient safety and reduce radiation exposure. The study quantifies the effective doses for these CT examinations, compares them with national and local DRLs, and analyzes adherence to international benchmarks to identify opportunities for dose reduction and safety improvements. A total of 2,828 CT scans, including abdominal, chest, and skull imaging, were analyzed from MDCT scanners, with dose descriptors extracted from image dose reports to calculate the ED using established dose-length product conversion coefficients. The calculated ED values were compared with national and local DRLs as well as internationally published reference benchmarks. The comparative analysis revealed significant variations in ED values across different types of CT exams, highlighting areas where dose optimization is possible. The findings suggest that some centers need to adjust their protocols to better align with best practices and improve patient safety. This study provides valuable insights into radiation dose levels in southern Nigeria, offering recommendations for protocol adjustments and enhanced patient safety while serving as a guide for improving local imaging practices, reducing radiation exposure, and ensuring better clinical outcomes.

Purpose/Objective(s) This study investigated the performance of machine learning-derived nomograms utilizing logistic regression (LR) and random forest (RF) algorithms, along with biomarkers from 68Ga-PSMA-11 PET, MRI, and the Decipher test, in predicting side-specific extraprostatic extension (EPE) in radical prostatectomy (RP) patients. Materials/Methods In this retrospective study, data from prostate cancer patients who underwent RP at Indiana University Hospital, alongside their 68Ga-PSMA-11 PET, MRI, and Decipher results were analyzed. This included their clinical information (fraction of positive cores on biopsy, prostate-specific antigen density, and ISUP biopsy grade) alongside results from three biomarkers: 68Ga-PSMA-11 PET, MRI, and the Decipher test. Several machine learning-derived nomogram models were developed using LR and RF algorithms. These models incorporated various combinations of the biomarkers (PET-only, MRI-only, Decipher-only, Decipher-only, PET+Decipher, PET+MRI, MRI+Decipher, and PET+MRI+Decipher) along with the clinical data. The performance of each model was assessed using the area under the receiver operating characteristic curve (AUC-ROC). Results In the nomogram development and validation process, we selected 110 patients with a median age of 61.5 years (range = 48 –77 years), PSA score of 2.7–7.15 ng/mL, ISUP biopsy grade of 1–5, and a fraction of positive cores of 0.50 ± 0.36. For the PET-only, MRI-only, Decipher-only, PET+Decipher, PET+MRI, MRI+Decipher, and PET+MRI+Decipher variables, the performance of the RF models was 0.82, 0.76, 0.65, 0.83, 0.76, 0.73, and 0.83, respectively. In addition, the performance of the LR models was 0.80, 0.71, 0.62, 0.78, 0.79, 0.78, and 0.63, respectively. Conclusion From this study, the RF algorithm generally outperformed LR algorithm in this task. 68Ga-PSMA PET-only variables provided better EPE prediction than MRI or Decipher variables used individually among the evaluated biomarkers. However, the most accurate predictions were achieved by combining Decipher and MRI data with PET data, particularly within the RF models. This suggests that a combined approach using multiple biomarkers holds promise for improving EPE prediction. Hence to guide treatment decisions, such as nerve-sparing RP. Further study with a larger patient population is needed to refine the model’s accuracy and validate its clinical utility.

Accurate alignment in the chest region is crucial in VMAT TBI treatment. Efforts should be made to minimize shifts over 1cm in the IS direction.

Most modern LINAC are installed with a variety of photon energies varying from 6 MV to 18 MV. The conventional wisdom in external radiotherapy has been that higher energies (>10 MV) are preferred for deep seated pelvic and/or abdominal lesions, especially for larger target volume or larger patients. 1. Laughlin JS Mohan R Kutcher GJ. Choice of optimum megavoltage for accelerators for photon beam treatment. Int J Radiat Oncol Biol Phys. 1986; 12: 1551-1557 Abstract Full Text PDF PubMed Scopus (48) Google Scholar The usefulness for higher energy is more evident with 3DCRT, as compared to IMRT and VMAT. 2. Soderstrom S Eklof A Brahme A. Aspects on the optimal photon beam energy for radiation therapy. Acta Oncol. 1999; 38: 179-187 Crossref PubMed Scopus (33) Google Scholar The recent paper by Eskens et al. 3. Eskens M Nguyen H Deere W et al. A retrospective evaluation of mixed energy volumetric modulated arc therapy for anal cancers with lymph node involvement. Med Dosim. 2020; 45 (Winter): 339-345 Abstract Full Text Full Text PDF Scopus (0) Google Scholar shades more light into the usefulness of using different energies or a combination of energies for anal cancers with lymph node involvement. This is a well written paper; however, certain useful information was not addressed in the manuscript. 1-Comparison between 6 MV and 10 MV

Whole pelvic radiotherapy (WPRT) can sterilize microscopic lymph node metastases in treatment of prostate cancer. WPRT, compared to prostate only radiotherapy (PORT), is associated with increased acute gastrointestinal, and hematological toxicities. To further explore minimizing normal tissue toxicities associated with WPRT in definitive IMRT for prostate cancer, this planning study compared dosimetric differences between static 9-field-IMRT, full arc VMAT, and mixed partial-full arc VMAT techniques. In this retrospective study, 12 prostate cancer patients who met the criteria for WPRT were randomly selected for this study. The initial volume, PTV46, included the prostate, seminal vesicles, and pelvic nodes with margin and was prescribed to 4600 cGy. The cone-down volume, PTV78, included the prostate and proximal seminal vesicles with margin to a total dose of 7800 cGy. For each CT image set, 3 plans were generated for each of the PTVs: an IMRT plan, a full arc (FA) VMAT plan, and a mixed partial-full arc (PFA) VMAT plan, using 6MV photons energy. According to RTOG protocols none of the plans had a major Conformity Index (CI) violation by any of the 3 planning techniques. PFA plan had the best mean CI index of 1.00 and significantly better than IMRT (p = 0.03) and FA (p = 0.007). For equivalent PTV coverage, the average composite gradient index of the PFA plans was better than the IMRT and the FA plans with values 1.92, 2.03, and 2.01 respectively. The defference was statistically significant between PFA/IMRT and PFA/FA, with p- values of < 0.001. The IMRT plans and the PFA plans provided very similar doses to the rectum, bladder, sigmoid colon, and femoral heads, which were lower than the dose in the FA plans. There was a significant decrease in the mean dose to the rectum from 4524 cGy with the FA to 4182 cGy with the PFA and 4091 cGy with IMRT (p < 0.001). The percent of rectum receiving 4000 cGy was also the highest with FA at 66.1% compared to 49.9% (PFA) and 47.5% (IMRT). There was a significant decrease in the mean dose to the bladder from 3922 cGy (FA) to 3551 cGy (PFA) and 3612 cGy (IMRT) (p < 0.001). The percent of bladder receiving 4000 cGy was also the highest with FA at 45.4% compared to 36.6% (PFA) and 37.4% (IMRT). The average mean dose to the sigmoid colon decreased from 4177 cGy (FA) to 3893 cGy (PFA) and 3819 cGy (IMRT). The average mean dose to the femoral heads decreased from 2091 cGy (FA) to 2026 cGy (PFA) and 1987 cGy (IMRT). Considering the improvement in plan quality indices recorded in this study including the dose gradient and the dose to organs at risk, mixed partial-full arc plans may be the preferred VMAT treatment technique over full arc plans for prostate cancer treatments that include nodal volumes.

There is no universally accepted definition of quality improvement (QI). However, the American Board of Radiology (ABR) defines “QI” as “a systematic approach to the study of healthcare and/or a commitment to efforts to continuously improve performance and outcomes in healthcare”. According to Kruskal et al.,[1] QI in radiation oncology includes “(a) quality assurance programs for continuous improvements in quality, (b) processes to improve staff and patient safety, and (c) procedures to improve the clinical, technical, and therapy performance of all staff”.[1] Fundamentally, QI techniques are, well founded methods to drive change and improve efficiency. The goal of QI is therefore to create practical processes and structures that will introduce positive change into a work environment in a reproducible and sustainable way that is non-disruptive and at an acceptable cost. There are many forces that can drive the creation of QI programs in radiation oncology. The first is the desire to provide high-quality patient care, which is defined by the Institute of Medicine as “safe, effective, patient-centered, timely, efficient, and equitable care”.[2] The second is the mandate of accrediting bodies such as the Joint Commission and the American College of Radiology (ACR), whose accrediting standards further support this goal. The third is the economic incentives to provide high-quality care at an affordable cost.[3] Clinical medical physicists (MPs) are often viewed as the custodians of quality in radiation therapy department. Radiation therapy is a long-complicated process and therefore has numerous avenues for potential QI endeavors.[4, 5] These QI initiatives demand time and resources to be successful. More often, when time is not reserved, these initiatives become administrative burdens on the staff adding to their already established workflow. To make QI relevant, feasible and sustainable, it is necessary to embed it into MP workflow. This act transforms QI from a burden, which places an extra demand on physicists’ time, into an exercise of team ingenuity. Clinical MPs dedicated time is therefore recommended to support QI. The justifications for this recommendation are presented in this commentary. Hospital accreditation is an external systematic assessment of a hospital's structures, processes, and results by an independent professional body using pre-established accepted optimum standards. Accreditation has an important role in establishing standards and in improving the quality, safety, effectiveness, and efficiency of hospital services.[6] There are three professional organizations that may provide radiation oncology accreditation: the American College of Radiology (ACR), the American Society for Radiation Oncology (ASTRO), and the American College of Radiation Oncology (ACRO).[7-9] The accreditation programs from ACR, ASTRO, and ACRO are Radiation Oncology Practice Accreditation (ROPA), Accreditation Programs for Excellence (APEX), and Practice Accreditation Program (PAP), respectively. These programs provide radiation oncologists with an independent and impartial peer review. Facility staff, equipment, treatment-planning, treatment records, patient-safety policies, and quality control/quality assessment activities are all assessed.[7-9] ACR established ROPA in 1986. ACRO's PAP was initiated in 1996 as a service to ACRO members.[7] ASTRO unveiled its accreditation program for excellence in late 2015. The American Board of Medical Specialties (ABMS) – a 24-member board, representing all medical subspecialties in the USA, in March 2000, agreed to initiate specialty-specific Maintenance of Certification (MOC) programs. Diplomats are no longer issued lifetime certification, instead they need to provide documentations of continued learning and QI. MOC recognizes that in addition to medical knowledge, several essential elements such as communication skills involved in the delivering quality care must be developed and maintained throughout one's career.[10, 11] The ABR representing diagnostic radiology, radiation oncology, and radiation physics has developed their own MOC program, which was approved by the ABMS, and initiated with full implementation for all three disciplines starting in 2007.[10, 11] The ABR MOC has four components: professional standing, lifelong learning and self-assessment, cognitive expertise, and evaluation of practice performance. The self-evaluation of practice performance includes the process of continuing QI and is entitled “Practice Quality Improvement” (PQI).[12] The ABR's guidelines state that “every radiologic physics diplomate must complete a PQI project. The choice of PQI activities and projects are to meet the spirit of the definition of QI”.[10, 14] Medical Physics embodies a wide range of clinical aspects and ABR has identified five possible PQI areas: safety of patients, employees and public, accuracy of analysis and calculation, report turnaround and communication issues, and practice guidelines and standards and surveys.[10] In addition, the diplomate must demonstrate a commitment to maintaining competency as a radiologic physicist.[10] Effective from 15 March 2016, the ABR instituted the continuous certification and annual “look-back” processes which in part requires diplomates to have completed at least one PQI project in the previous 3 years. In a Medical Physics point/counter point article, Njeh et al.[15] argued that PQI project could provide background material for research and publication. Methodical PQI has never been a mainstream MPs’ activity, but the forgone sections have established the need for MPs to be active participants in these projects. This has been echoed in the growing request for PQI training to be included in the medical physicist residence syllabus.[16] The benefits to patients, clinicians, and healthcare providers of engaging in PQI are considerable, but there are many challenges involved in designing, delivering, and sustaining a QI intervention.[17] However, to have successful PQI projects some of these challenges need to be addressed: training, leadership support, time allocation, appropriate tools, mechanism for data collection, financial resources, human resources, and selecting the right project.[13, 17] MP needs to be trained in the process and procedures of PQI as they affect an individual's practice of radiologic Physics.[10] This education requirement is echoed by Medical Physics residency programs accrediting agencies. Nonetheless, recent surveys indicate that most programs lack a formal program to support this learning.[16, 18] The success of a PQI project depends on proper training on effective use of QI methodology.[13] Six Sigma and Lean are more common QI methods.[5, 19] Six Sigma reduce process variation by decreasing defects to a specific statistical measure. Six Sigma projects use a five-phased process known as DMAIC (define, measure, analyze, improve, and control).[20] The main emphasis of Lean is on cutting out unnecessary and wasteful steps in the delivery of a service. Lean uses a technique called value stream mapping. The next step is to apply the 5S (sort, simplify, sweep, standardize, and self-discipline).[21-23] Six Sigma and Lean have a complementary relationship with each other and can be combined as Lean Six Sigma. The synergetic adoption of these methods allows the creation of a continuous process flow that eliminates waste (Lean) and reduces process variation (Six Sigma), to achieve and maintain the best quality.[24] Education and training are also required in other quality control tools like root cause analysis (RCA), failure mode and effect analysis (FMEA),[25] and incident reporting and learning (IRL). RCA is a reactive retrospective approach used to ascertain the “root cause” of a problem that has already occurred, whereas FMEA is a proactive prospective systematic approach that is used to identify and understand causes, contributing factors, and effects of potential failures on a process, system, or practice. IRL is about using the opportunities from reported actual or potential incidents and analyzing them to determine the systemic and human factors involved.[22, 26, 27] IRL is a reactive and retrospective look at a known error. Critical elements in any quality program include leadership willingness to experiment and take risks. Institutional leadership and support send the message that all quality-related efforts are valued and constitute a central component of the institution's mission. This important message should be enhanced by tangible support. This support may be financial such as the provision of human resources such as a departmental quality coordinator, or administrative, such as establishing and facilitating interdepartmental quality forums or adverse event reporting systems. Further leadership support can be demonstrated by the acknowledgment of efforts and successes.[1, 13, 28-30] Leadership support is more critical when there is a bump in the road. As Hawkins[31] eloquently states “There will be moments during all performance improvement projects when things do not go as planned or unforeseeable obstacles arise. If there is not a buy-in from leadership—from people to whom members of your department look for guidance—then the initiative will fail. Simply engaging these individuals is likely not enough. As a QI project leader, you must clearly show key leadership stakeholders why the desired change is necessary, and how you plan to achieve the desired results. Hopefully, they have established a culture that supports such efforts”.[31] The magnitude of resources required to support quality improvements is often underestimated, but without adequate financial support, infrastructure, managerial skills, and dedicated time, efforts to improve quality can quickly run into difficulties.[17] Time has been identified by many as a critical component of successful PQI projects.[3, 17, 30, 32-34] Broder et al.[3] advocated in their article that extra staffing is a prerequisite for successful QI projects. Extra staffing can then be used to give the required time needed to – identify and define the process or problem, collect and analyze the data, generate and prioritize solutions, and finally implement change and monitor results.[1] After determination of the proper staffing and skills needed, roles and time allocation should be clearly defined.[3] Kaplan et al.[32] conducted a literature review of factors affecting the success of QI projects. The most frequently examined contextual factors were funding, general resources, and time. Studies that assessed time resources for QI found positive associations in 60% of the associations tested.[35] Choudhery et al.[33] examined radiology resident participation in PQI projects and reported that resident with dedicated time were more likely to complete a PQI and to publish their results. In a survey of 25 healthcare professionals who had recently carried out PQI projects, having limited time to perform the initiative was considered the most important barrier.[30] In day-to-day activities, one is likely to have competing priorities and will need support to make time for QI. By providing MPs dedicated time for PQI projects, management upholds its core values which include a commitment to excellence, by ingraining QI into the fabric of all clinical processes and in all aspects of the services we provide, excellence in quality and safety of clinical care, and service to our patients and customers and adherence to regulatory compliance for QI initiatives. Furthermore, it demands accountability for the dedicated time and thus more likely the success of the PQI projects. Most of the PQI projects will result in improved patient care and reduce costs in the provision of radiation therapy. Some PQI projects will generate background data for research and publications. Last, clinical MP will meet their MOC part IV requirement. We thank Deputy Editors-in-Chief Timothy Solberg and Per Halvorsen for their valuable and perceptive comments. All authors contributed significantly to the drafting and the final manuscript. None to declare.