To define a male and female pelvic normal tissue contouring atlas for Radiation Therapy Oncology Group (RTOG) trials.
One male pelvis computed tomography (CT) data set and one female pelvis CT data ...set were shared via the Image-Guided Therapy QA Center. A total of 16 radiation oncologists participated. The following organs at risk were contoured in both CT sets: anus, anorectum, rectum (gastrointestinal and genitourinary definitions), bowel NOS (not otherwise specified), small bowel, large bowel, and proximal femurs. The following were contoured in the male set only: bladder, prostate, seminal vesicles, and penile bulb. The following were contoured in the female set only: uterus, cervix, and ovaries. A computer program used the binomial distribution to generate 95% group consensus contours. These contours and definitions were then reviewed by the group and modified.
The panel achieved consensus definitions for pelvic normal tissue contouring in RTOG trials with these standardized names: Rectum, AnoRectum, SmallBowel, Colon, BowelBag, Bladder, UteroCervix, Adnexa_R, Adnexa_L, Prostate, SeminalVesc, PenileBulb, Femur_R, and Femur_L. Two additional normal structures whose purpose is to serve as targets in anal and rectal cancer were defined: AnoRectumSig and Mesorectum. Detailed target volume contouring guidelines and images are discussed.
Consensus guidelines for pelvic normal tissue contouring were reached and are available as a CT image atlas on the RTOG Web site. This will allow uniformity in defining normal tissues for clinical trials delivering pelvic radiation and will facilitate future normal tissue complication research.
Locally advanced non-small cell lung cancer (NSCLC) patients suffer from a high local failure rate following radiotherapy. Despite many efforts to develop new dose-volume models for early detection ...of tumor local failure, there was no reported significant improvement in their application prospectively. Based on recent studies of biomarker proteins' role in hypoxia and inflammation in predicting tumor response to radiotherapy, we hypothesize that combining physical and biological factors with a suitable framework could improve the overall prediction. To test this hypothesis, we propose a graphical Bayesian network framework for predicting local failure in lung cancer. The proposed approach was tested using two different datasets of locally advanced NSCLC patients treated with radiotherapy. The first dataset was collected retrospectively, which comprises clinical and dosimetric variables only. The second dataset was collected prospectively in which in addition to clinical and dosimetric information, blood was drawn from the patients at various time points to extract candidate biomarkers as well. Our preliminary results show that the proposed method can be used as an efficient method to develop predictive models of local failure in these patients and to interpret relationships among the different variables in the models. We also demonstrate the potential use of heterogeneous physical and biological variables to improve the model prediction. With the first dataset, we achieved better performance compared with competing Bayesian-based classifiers. With the second dataset, the combined model had a slightly higher performance compared to individual physical and biological models, with the biological variables making the largest contribution. Our preliminary results highlight the potential of the proposed integrated approach for predicting post-radiotherapy local failure in NSCLC patients.
To evaluate variability in the definition of preoperative radiotherapy gross tumor volume (GTV) and clinical target volume (CTV) delineated by sarcoma radiation oncologists.
Extremity sarcoma ...planning CT images along with the corresponding diagnostic MRI from two patients were distributed to 10 Radiation Therapy Oncology Group sarcoma radiation oncologists with instructions to define GTV and CTV using standardized guidelines. The CT data with contours were then returned for central analysis. Contours representing statistically corrected 95% (V95) and 100% (V100) agreement were computed for each structure.
For the GTV, the minimum, maximum, mean (SD) volumes (mL) were 674, 798, 752±35 for the lower extremity case and 383, 543, 447±46 for the upper extremity case. The volume (cc) of the union, V95 and V100 were 882, 761, and 752 for the lower, and 587, 461, and 455 for the upper extremity, respectively. The overall GTV agreement was judged to be almost perfect in both lower and upper extremity cases (kappa=0.9 p<0.0001 and kappa=0.86 p<0.0001). For the CTV, the minimum, maximum, mean (SD) volumes (mL) were 1145, 1911, 1605±211 for the lower extremity case and 637, 1246, 1006±180 for the upper extremity case. The volume (cc) of the union, V95, and V100 were 2094, 1609, and 1593 for the lower, and 1533, 1020, and 965 for the upper extremity cases, respectively. The overall CTV agreement was judged to be almost perfect in the lower extremity case (kappa=0.85 p<0.0001) but only substantial in the upper extremity case (kappa=0.77 p<0.0001).
Almost perfect agreement existed in the GTV of these two representative cases. There was no significant disagreement in the CTV of the lower extremity, but variation in the CTV of upper extremity was seen, perhaps related to the positional differences between the planning CT and the diagnostic MRI.
Estimating the proper margins for the planning target volume (PTV) could be a challenging task in cases where the organ undergoes significant changes during the course of radiotherapy treatment. ...Developments in image-guidance and the presence of onboard imaging technologies facilitate the process of correcting setup errors. However, estimation of errors to organ motions remain an open question due to the lack of proper software tools to accompany these imaging technological advances. Therefore, we have developed a new tool for visualization and quantification of deformations from daily images. The tool allows for estimation of tumor coverage and normal tissue exposure as a function of selected margin (isotropic or anisotropic). Moreover, the software allows estimation of the optimal margin based on the probability of an organ being present at a particular location. Methods based on swarm intelligence, specifically Ant Colony Optimization (ACO) are used to provide an efficient estimate of the optimal margin extent in each direction. ACO can provide global optimal solutions in highly nonlinear problems such as margin estimation. The proposed method is demonstrated using cases from gastric lymphoma with daily TomoTherapy megavoltage CT (MVCT) contours. Preliminary results using Dice similarity index are promising and it is expected that the proposed tool will be very helpful and have significant impact for guiding future margin definition protocols.
Radiation pneumonitis (RP) is a potentially fatal side effect arising in lung cancer patients who receive radiotherapy as part of their treatment. For the modeling of RP outcomes data, several ...predictive models based on traditional statistical methods and machine learning techniques have been reported. However, no guidance to variation in performance has been provided to date. In this study, we explore several machine learning algorithms for classification of RP data. The performance of these classification algorithms is investigated in conjunction with several feature selection strategies and the impact of the feature selection strategy on performance is further evaluated. The extracted features include patients demographic, clinical and pathological variables, treatment techniques, and dose-volume metrics. In conjunction, we have been developing an in-house Matlab-based open source software tool, called DREES, customized for modeling and exploring dose response in radiation oncology. This software has been upgraded with a popular classification algorithm called support vector machine (SVM), which seems to provide improved performance in our exploration analysis and has strong potential to strengthen the ability of radiotherapy modelers in analyzing radiotherapy outcomes data.
PDF
