Fully automated deep learning based auto-contouring of liver segments and spleen on contrast-enhanced CT images.

Manual delineation of liver segments on computed tomography (CT) images for primary/secondary liver cancer (LC) patients is time-intensive and prone to inter/intra-observer variability. Therefore, we developed a deep-learning-based model to auto-contour liver segments and spleen on contrast-enhanced CT (CECT) images. We trained two models using 3d patch-based attention U-Net ([Formula: see text] and 3d full resolution of nnU-Net ([Formula: see text] to determine the best architecture ([Formula: see text]. BA was used with vessels ([Formula: see text] and spleen ([Formula: see text] to assess the impact on segment contouring. Models were trained, validated, and tested on 160 ([Formula: see text]), 40 ([Formula: see text]), 33 ([Formula: see text]), 25 (CCH) and 20 (CPVE) CECT of LC patients. [Formula: see text] outperformed [Formula: see text] across all segments with median differences in Dice similarity coefficients (DSC) ranging 0.03-0.05 (p < 0.05). [Formula: see text], and [Formula: see text] were not statistically different (p > 0.05), however, both were slightly better than [Formula: see text] by DSC up to 0.02. The final model, [Formula: see text], showed a mean DSC of 0.89, 0.82, 0.88, 0.87, 0.96, and 0.95 for segments 1, 2, 3, 4, 5-8, and spleen, respectively on entire test sets. Qualitatively, more than 85% of cases showed a Likert score [Formula: see text] 3 on test sets. Our final model provides clinically acceptable contours of liver segments and spleen which are usable in treatment planning.

Analysis and prediction of liver volume change maps derived from computational tomography scans acquired pre- and post-radiation therapy.

External beam radiation therapy (EBRT) of liver cancers can cause local liver atrophy as a result of tissue damage or hypertrophy as a result of liver regeneration. Predicting those volumetric changes would enable new strategies for liver function preservation during treatment planning. However, understanding of the spatial dose/volume relationship is still limited. This study leverages the use of deep learning-based segmentation and biomechanical deformable image registration (DIR) to analyze and predict this relationship. Pre- and Post-EBRT imaging data were collected for 100 patients treated for hepatocellular carcinomas, cholangiocarcinoma or CRC with intensity-modulated radiotherapy (IMRT) with prescription doses ranging from 50 to 100 Gy delivered in 10-28 fractions. For each patient, DIR between the portal and venous (PV) phase of a diagnostic computed tomography (CT) scan acquired before radiation therapy (RT) planning, and a PV phase of a diagnostic CT scan acquired after the end of RT (on average 147 ± 36 d) was performed to calculate Jacobian maps representing volume changes in the liver. These volume change maps were used: (i): to analyze the dose/volume relationship in the whole liver and individual Couinaud's segments; and (ii): to investigate the use of deep-learning to predict a Jacobian map solely based on the pre-RT diagnostic CT and planned dose distribution. Moderate correlations between mean equivalent dose in 2 Gy fractions (EQD2) and volume change was observed for all liver sub-regions analyzed individually with Pearson correlationrranging from -0.36 to -067. The predicted volume change maps showed a significantly stronger voxel-wise correlation with the DIR-based volume change maps than when considering the original EQD2 distribution (0.63 ± 0.24 versus 0.55 ± 23, respectively), demonstrating the ability of the proposed approach to establish complex relationships between planned dose and liver volume response months after treatment, which represents a promising prediction tool for the development of future adaptive and personalized liver radiation therapy strategies.

Development and validation of a population-based anatomical colorectal model for radiation dosimetry in late effects studies of survivors of childhood cancer.

PURPOSE: The purposes of this study were to develop and integrate a colorectal model that incorporates anatomical variations of pediatric patients into the age-scalable MD Anderson Late Effects (MDA-LE) computational phantom, and validate the model for pediatric radiation therapy (RT) dose reconstructions. METHODS: Colorectal contours were manually derived from whole-body non-contrast computed tomography (CT) scans of 114 pediatric patients (age range: 2.1-21.6 years, 74 males, 40 females). One contour was used for an anatomical template, 103 for training and 10 for testing. Training contours were used to create a colorectal principal component analysis (PCA)-based statistical shape model (SSM) to extract the population's dominant deformations. The SSM was integrated into the MDA-LE phantom. Geometric accuracy was assessed between patient-specific and SSM contours using several overlap metrics. Two alternative colorectal shapes were generated using the first 17 dominant modes of the PCA-based SSM. Dosimetric accuracy was assessed by comparing colorectal doses from test patients' CT-based RT plans (ground truth) with reconstructed doses for the mean and two alternative models in age-matched MDA-LE phantoms. RESULTS: When using all 103 PCA modes, the mean (min-max) Dice similarity coefficient, distance-to-agreement and Hausdorff distance between the patient-specific and reconstructed contours for the test patients were 0.89 (0.85-0.91), 2.1 mm (1.7-3.0), and 8.6 mm (5.7-14.3), respectively. The average percent difference between reconstructed and ground truth mean and maximum colorectal doses for the mean (alternative 1, 2) model were 6.3% (8.1%, 6.1%) and 4.4% (4.3%, 4.7%), respectively. CONCLUSIONS: We developed, validated and integrated a colorectal PCA-based SSM into the MDA-LE phantom and demonstrated its dosimetric performance for accurate pediatric RT dose reconstruction.

Body region-specific 3D age-scaling functions for scaling whole-body computed tomography anatomy for pediatric late effects studies.

Purpose.Radiation epidemiology studies of childhood cancer survivors treated in the pre-computed tomography (CT) era reconstruct the patients' treatment fields on computational phantoms. For such studies, the phantoms are commonly scaled to age at the time of radiotherapy treatment because age is the generally available anthropometric parameter. Several reference size phantoms are used in such studies, but reference size phantoms are only available at discrete ages (e.g.: newborn, 1, 5, 10, 15, and Adult). When such phantoms are used for RT dose reconstructions, the nearest discrete-aged phantom is selected to represent a survivor of a specific age. In this work, we (1) conducted a feasibility study to scale reference size phantoms at discrete ages to various other ages, and (2) evaluated the dosimetric impact of using exact age-scaled phantoms as opposed to nearest age-matched phantoms at discrete ages.Methods.We have adopted the University of Florida/National Cancer Institute (UF/NCI) computational phantom library for our studies. For the feasibility study, eight male and female reference size UF/NCI phantoms (5, 10, 15, and 35 years) were downscaled to fourteen different ages which included next nearest available lower discrete ages (1, 5, 10 and 15 years) and the median ages at the time of RT for Wilms' tumor (3.9 years), craniospinal (8.0 years), and all survivors (9.1 years old) in the Childhood Cancer Survivor Study (CCSS) expansion cohort treated with RT. The downscaling was performed using our in-house age scaling functions (ASFs). To geometrically validate the scaling, Dice similarity coefficient (DSC), mean distance to agreement (MDA), and Euclidean distance (ED) were calculated between the scaled and ground-truth discrete-aged phantom (unscaled UF/NCI) for whole-body, brain, heart, liver, pancreas, and kidneys. Additionally, heights of the scaled phantoms were compared with ground-truth phantoms' height, and the Centers for Disease Control and Prevention (CDC) reported 50th percentile height. Scaled organ masses were compared with ground-truth organ masses. For the dosimetric assessment, one reference size phantom and seventeen body-size dependent 5-year-old phantoms (9 male and 8 female) of varying body mass indices (BMI) were downscaled to 3.9-year-old dimensions for two different radiation dose studies. For the first study, we simulated a 6 MV photon right-sided flank field RT plan on a reference size 5-year-old and 3.9-year-old (both of healthy BMI), keeping the field size the same in both cases. Percent of volume receiving dose ≥15 Gy (V15) and the mean dose were calculated for the pancreas, liver, and stomach. For the second study, the same treatment plan, but with patient anatomy-dependent field sizes, was simulated on seventeen body-size dependent 5- and 3.9-year-old phantoms with varying BMIs. V15, mean dose, and minimum dose received by 1% of the volume (D1), and by 95% of the volume (D95) were calculated for pancreas, liver, stomach, left kidney (contralateral), right kidney, right and left colons, gallbladder, thoracic vertebrae, and lumbar vertebrae. A non-parametric Wilcoxon rank-sum test was performed to determine if the dose to organs of exact age-scaled and nearest age-matched phantoms were significantly different (p < 0.05).Results.In the feasibility study, the best DSCs were obtained for the brain (median: 0.86) and whole-body (median: 0.91) while kidneys (median: 0.58) and pancreas (median: 0.32) showed poorer agreement. In the case of MDA and ED, whole-body, brain, and kidneys showed tighter distribution and lower median values as compared to other organs. For height comparison, the overall agreement was within 2.8% (3.9 cm) and 3.0% (3.2 cm) of ground-truth UF/NCI and CDC reported 50th percentile heights, respectively. For mass comparison, the maximum percent and absolute differences between the scaled and ground-truth organ masses were within 31.3% (29.8 g) and 211.8 g (16.4%), respectively (across all ages). In the first dosimetric study, absolute difference up to 6% and 1.3 Gy was found for V15and mean dose, respectively. In the second dosimetric study, V15and mean dose were significantly different (p < 0.05) for all studied organs except the fully in-beam organs. D1and D95were not significantly different for most organs (p > 0.05).Conclusion.We have successfully evaluated our ASFs by scaling UF/NCI computational phantoms from one age to another age, which demonstrates the feasibility of scaling any CT-based anatomy. We have found that dose to organs of exact age-scaled and nearest aged-matched phantoms are significantly different (p < 0.05) which indicates that using the exact age-scaled phantoms for retrospective dosimetric studies is a better approach.

Radiation therapy related cardiac disease risk in childhood cancer survivors: Updated dosimetry analysis from the Childhood Cancer Survivor Study.

BACKGROUND AND PURPOSE: We previously evaluated late cardiac disease in long-term survivors in the Childhood Cancer Survivor Study (CCSS) based on heart radiation therapy (RT) doses estimated from an age-scaled phantom with a simple atlas-based heart model (HAtlas). We enhanced our phantom with a high-resolution CT-based anatomically realistic and validated age-scalable cardiac model (HHybrid). We aimed to evaluate how this update would impact our prior estimates of RT-related late cardiac disease risk in the CCSS cohort. METHODS: We evaluated 24,214 survivors from the CCSS diagnosed from 1970 to 1999. RT fields were reconstructed on an age-scaled phantom with HHybrid and mean heart dose (Dm), percent volume receiving ≥ 20 Gy (V20) and ≥ 5 Gy with V20 = 0 ( [Formula: see text] ) were calculated. We reevaluated cumulative incidences and adjusted relative rates of grade 3-5 Common Terminology Criteria for Adverse Events outcomes for any cardiac disease, coronary artery disease (CAD), and heart failure (HF) in association with Dm, V20, and [Formula: see text] (as categorical variables). Dose-response relationships were evaluated using piecewise-exponential models, adjusting for attained age, sex, cancer diagnosis age, race/ethnicity, time-dependent smoking history, diagnosis year, and chemotherapy exposure and doses. For relative rates, Dm was also considered as a continuous variable. RESULTS: Consistent with previous findings with HAtlas, reevaluation using HHybrid dosimetry found that, Dm ≥ 10 Gy, V20 ≥ 0.1%, and [Formula: see text]  ≥ 50% were all associated with increased cumulative incidences and relative rates for any cardiac disease, CAD, and HF. While updated risk estimates were consistent with previous estimates overall without statistically significant changes, there were some important and significant (P < 0.05) increases in risk with updated dosimetry for Dm in the category of 20 to 29.9 Gy and V20 in the category of 30% to 79.9%. When changes in the linear dose-response relationship for Dm were assessed, the slopes of the dose response were steeper (P < 0.001) with updated dosimetry. Changes were primarily observed among individuals with chest-directed RT with prescribed doses ≥ 20 Gy. CONCLUSION: These findings present a methodological advancement in heart RT dosimetry with improved estimates of RT-related late cardiac disease risk. While results are broadly consistent with our prior study, we report that, with updated cardiac dosimetry, risks of cardiac disease are significantly higher in two dose and volume categories and slopes of the Dm-specific RT-response relationships are steeper. These data support the use of contemporary RT to achieve lower heart doses for pediatric patients, particularly those requiring chest-directed RT.

Development and validation of an age-scalable cardiac model with substructures for dosimetry in late-effects studies of childhood cancer survivors.

BACKGROUND AND PURPOSE: Radiation therapy is a risk factor for late cardiac disease in childhood cancer survivors. Several pediatric cohort studies have established whole heart dose and dose-volume response models. Emerging data suggest that dose to cardiac substructures may be more predictive than whole heart metrics. In order to develop substructure dose-response models, the heart model previously used for pediatric cohort dosimetry needed enhancement and substructure delineation. METHODS: To enhance our heart model, we combined the age-scalable capability of our computational phantom with the anatomically-delineated (with substructures) heart models from an international humanoid phantom series. We examined cardiac volume similarity/overlap between registered age-scaled phantoms (1, 5, 10, and 15 years) with the enhanced heart model and the reference phantoms of the same age; dice similarity coefficient (DSC) and overlap coefficient (OC) were calculated for each matched pair. To assess the accuracy of our enhanced heart model, we compared doses from computed tomography-based planning (ground truth) with reconstructed heart doses. We also compared doses calculated with the prior and enhanced heart models for a cohort of nearly 5000 childhood cancer survivors. RESULTS: We developed a realistic cardiac model with 14-substructures, scalable across a broad age range (1-15 years); average DSC and OC were 0.84 ± 0.05 and 0.90 ± 0.05, respectively. The average percent difference between reconstructed and ground truth mean heart doses was 4.2%. In the cohort dosimetry analysis, dose and dose-volume metrics were approximately 10% lower on average when the enhanced heart model was used for dose reconstructions. CONCLUSION: We successfully developed and validated an anatomically realistic age-scalable cardiac model that can be used to establish substructure dose-response models for late cardiac disease in childhood cancer survivor cohorts.

Development of an age-scalable 3D computational phantom in DICOM standard for late effects studies of childhood cancer survivors.

Purpose: We previously developed an age-scalable 3D computational phantom that has been widely used for retrospective whole-body dose reconstructions of conventional two-dimensional historic radiation therapy (RT) treatments in late effects studies of childhood cancer survivors. This phantom is modeled in the FORTRAN programming language and is not readily applicable for dose reconstructions for survivors treated with contemporary RT whose treatment plans were designed using computed tomography images and complex treatment fields. The goal of this work was to adapt the current FORTRAN model of our age-scalable computational phantom into Digital Imaging and Communications in Medicine (DICOM) standard so that it can be used with any treatment planning system (TPS) to reconstruct contemporary RT. Additionally, we report a detailed description of the phantom's age-based scaling functions, information that was not previously published. Method: We developed a Python script that adapts our phantom model from FORTRAN to DICOM. To validate the conversion, we compared geometric parameters for the phantom modeled in FORTRAN and DICOM scaled to ages 1 month, 6 months, 1, 2, 3, 5, 8, 10, 15, and 18 years. Specifically, we calculated the percent differences between the corner points and volume of each body region and the normalized mean square distance (NMSD) between each of the organs. In addition, we also calculated the percent difference between the heights of our DICOM age-scaled phantom and the heights (50th percentile) reported by the World Health Organization (WHO) and Centers for Disease Control and Prevention (CDC) for male and female children of the same ages. Additionally, we calculated the difference between the organ masses for our DICOM phantom and the organ masses for two reference phantoms (from International Comission on Radiation Protection (ICRP) 89 and the University of Florida/National Cancer Institute reference hybrid voxel phantoms) for ages newborn, 1, 5, 10, 15 and adult. Lastly, we conducted a feasibility study using our DICOM phantom for organ dose calculations in a commercial TPS. Specifically, we simulated a 6 MV photon right-sided flank field RT plan for our DICOM phantom scaled to age 3.9 years; treatment field parameters and age were typical of a Wilms tumor RT treatment in the Childhood Cancer Survivor Study. For comparison, the same treatment was simulated using our in-house dose calculation system with our FORTRAN phantom. The percent differences (between FORTRAN and DICOM) in mean dose and percent of volume receiving dose ⩾5 Gy were calculated for two organs at risk, liver and pancreas. Results: The percent differences in corner points and the volumes of head, neck, and trunk body regions between our phantom modeled in FORTRAN and DICOM agreed within 3%. For all of the ages, the NMSDs were negliglible with a maximum NMSD of 7.80 × 10-2 mm for occiptital lobe of 1 month. The heights of our age-scaled phantom agreed with WHO/CDC data within 7% from infant to adult, and within 2% agreement for ages 5 years and older. We observed that organ masses in our phantom are less than the organ masses for other reference phantoms. Dose calculations done with our in-house calculation system (with FORTRAN phantom) and commercial TPS (with DICOM phantom) agreed within 7%. Conclusion: We successfully adapted our phantom model from the FORTRAN language to DICOM standard and validated its geometric consistency. We also demonstrated that our phantom model is representative of population height data for infant to adult, but that the organ masses are smaller than in other reference phantoms and need further refinement. Our age-scalable computational phantom modeled in DICOM standard can be scaled to any age at RT and used within a commercial TPS to retrospectively reconstruct doses from contemporary RT in childhood cancer survivors.