Search

With the era of big data, the utilization of machine learning algorithms in radiation oncology is rapidly growing with applications including: treatment response modeling, treatment planning, contouring, organ segmentation, image-guidance, motion tracking, quality assurance, and more. Despite this interest, practical clinical implementation of machine learning as part of the day-to-day clinical operations is still lagging. The aim of this white paper is to further promote progress in this new field of machine learning in radiation oncology by highlighting its untapped advantages and potentials for clinical advancement, while also presenting current challenges and open questions for future research. The targeted audience of this paper includes newcomers as well as practitioners in the field of medical physics/radiation oncology. The paper also provides general recommendations to avoid common pitfalls when applying these powerful data analytic tools to medical physics and radiation oncology problems and suggests some guidelines for transparent and informative reporting of machine learning results.

Background: Intrafraction motion monitoring in External Beam Radiation Therapy (EBRT) is usually accomplished by establishing a correlation between the tumor and the surrogates such as an external infrared reflector, implanted fiducial markers, or patient skin surface. These techniques either have unstable surrogate‐tumor correlation or are invasive. Markerless real‐time onboard imaging is a noninvasive alternative that directly images the target motion. However, the low target visibility due to overlapping tissues along the X‐ray projection path makes tumor tracking challenging. Purpose: To enhance the target visibility in projection images, a patient‐specific model was trained to synthesize the Target Specific Digitally Reconstructed Radiograph (TS‐DRR). Methods: Patient‐specific models were built using a conditional Generative Adversarial Network (cGAN) to map the onboard projection images to TS‐DRR. The standard Pix2Pix network was adopted as our cGAN model. We synthesized the TS‐DRR based on the onboard projection images using phantom and patient studies for spine tumors and lung tumors. Using previously acquired CT images, we generated DRR and its corresponding TS‐DRR to train the network. For data augmentation, random translations were applied to the CT volume when generating the training images. For the spine, separate models were trained for an anthropomorphic phantom and a patient treated with paraspinal stereotactic body radiation therapy (SBRT). For lung, separate models were trained for a phantom with a spherical tumor insert and a patient treated with free‐breathing SBRT. The models were tested using Intrafraction Review Images (IMR) for the spine and CBCT projection images for the lung. The performance of the models was validated using phantom studies with known couch shifts for the spine and known tumor deformation for the lung. Results: Both the patient and phantom studies showed that the proposed method can effectively enhance the target visibility of the projection images by mapping them into synthetic TS‐DRR (sTS‐DRR). For the spine phantom with known shifts of 1 mm, 2 mm, 3 mm, and 4 mm, the absolute mean errors for tumor tracking were 0.11 ± 0.05 mm in the x direction and 0.25 ± 0.08 mm in the y direction. For the lung phantom with known tumor motion of 1.8 mm, 5.8 mm, and 9 mm superiorly, the absolute mean errors for the registration between the sTS‐DRR and ground truth are 0.1 ± 0.3 mm in both the x and y directions. Compared to the projection images, the sTS‐DRR has increased the image correlation with the ground truth by around 83% and increased the structural similarity index measure with the ground truth by around 75% for the lung phantom. Conclusions: The sTS‐DRR can greatly enhance the target visibility in the onboard projection images for both the spine and lung tumors. The proposed method could be used to improve the markerless tumor tracking accuracy for EBRT. [ABSTRACT FROM AUTHOR]

An increasing number of bachelor's degrees in physics were awarded over the period shown in the figure, with 9283 total physics bachelor's degrees awarded in 2018. In my introductory physics courses in my small, liberal arts college, there were many women, but fewer than 20% of them graduated with a physics major in my cohort year. I was the first and only woman hired as physics faculty in my first position after graduate school, which I left to transition to a medical physics career. The bachelor's degree in physics is an entry-level degree into the field of medical physics, along with other science, technology, engineering, and mathematics (STEM) majors. [Extracted from the article]

Purpose: Dose accuracy in the buildup region for radiotherapy treatment planning suffers from challenges in both measurement and calculation. This study investigates the dosimetry in the buildup region at normal and oblique incidences for open and IMRT fields and assesses the quality of the treatment planning calculations. Methods: This study was divided into three parts. First, percent depth doses and profiles (for 5×5, 10×10, 20×20, and 30×30 cm2 field sizes at 0°, 45°, and 70° incidences) were measured in the buildup region in Solid Water using an Attix parallel plate chamber and Kodak XV film, respectively. Second, the parameters in the empirical contamination (EC) term of the convolution/superposition (CVSP) calculation algorithm were fitted based on open field measurements. Finally, seven segmental head-and-neck IMRT fields were measured on a flat phantom geometry and compared to calculations using γ and dose-gradient compensation (C) indices to evaluate the impact of residual discrepancies and to assess the adequacy of the contamination term for IMRT fields. Results: Local deviations between measurements and calculations for open fields were within 1% and 4% in the buildup region for normal and oblique incidences, respectively. The C index with 5%/1 mm criteria for IMRT fields ranged from 89% to 99% and from 96% to 98% at 2 mm and 10 cm depths, respectively. The quality of agreement in the buildup region for open and IMRT fields is comparable to that in nonbuildup regions. Conclusions: The added EC term in CVSP was determined to be adequate for both open and IMRT fields. Due to the dependence of calculation accuracy on (1) EC modeling, (2) internal convolution and density grid sizes, (3) implementation details in the algorithm, and (4) the accuracy of measurements used for treatment planning system commissioning, the authors recommend an evaluation of the accuracy of near-surface dose calculations as a part of treatment planning commissioning. [ABSTRACT FROM AUTHOR]

The dosimetric performance of a direct-detection active matrix flat panel dosimeter (AMFPD) is reported for intensity modulated radiation therapy (IMRT) measurements. The AMFPD consists of a-Si:H photodiodes and thin-film transistors deposited on a glass substrate with no overlying scintillator screen or metal plate. The device is operated at 0.8 frames per second in a continuous acquisition or fluoroscopic mode. The effect of the applied bias voltage across the photodiodes on the response of the AMFPD was evaluated because this parameter affects dark signal, lag contributions, and pixel sensitivity. In addition, the AMPFD response was evaluated as a function of dose, dose rate, and energy, for static fields at 10 cm depth. In continuous acquisition mode, the AMFPD maintained a linear dose response (r2>0.99999) up to at least 1040 cGy. In order to obtain reliable integrated dose results for IMRT fields, the effects of lag on the radiation signal were minimized by operating the system at the highest frame rate and at an appropriate reverse bias voltage. Segmental MLC and dynamic MLC IMRT fields were measured with the AMFPD, and the results were compared to film, using standard methods for reliable film dosimetry. Both AMFPD and film measurements were independently converted to dose in cGy. γ and χ values were calculated as indices of agreement. The results from the AMFPD were in excellent agreement with those from film. When 2% of Dmax and 2 mm of distance to agreement were used as the criteria, 98% of the region of interest (defined as the region where dose is greater than 5% of Dmax) satisfied |χ|≤1 on average across the cases that were tested. [ABSTRACT FROM AUTHOR]