Search NASA⌕ Search

SEARCH · Search NASA

Results for “Cancer models”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 865 records · Page 48

Volumetric Modulated Arc Therapy: Planning and Evaluation for Prostate Cancer Cases

Purpose: To develop an optimization method using volumetric modulated arc therapy (VMAT) and evaluate VMAT plans relative to the standard intensity-modulated radiotherapy (IMRT) approach in prostate cancer. Methods and Materials: A single gantry rotation was modeled using 177 equispaced beams. Multileaf collimator apertures and dose rates were optimized with respect to gantry angle subject to dose-volume-based objectives. Our VMAT implementation used conjugate gradient descent to optimize dose rate, and stochastic sampling to find optimal multileaf collimator leaf positions. A treatment planning study of 11 prostate cancer patients with a prescription dose of 86.4 Gy was performed to compare VMAT with a standard five-field IMRT approach. Plan evaluation statistics included the percentage of planning target volume (PTV) receiving 95% of prescribed dose (V95), dose to 95% of PTV (D95), mean PTV dose, tumor control probability, and dosimetric endpoints of normal organs, whereas monitor unit (MU) and delivery time were used to assess delivery efficiency. Results: Patient-averaged PTV V95, D95, mean dose, and tumor control probability in VMAT plans were 96%, 82.6 Gy, 88.5 Gy, and 0.920, respectively, vs. 97%, 84.0 Gy, 88.9 Gy, and 0.929 in IMRT plans. All critical structure dose requirements were met. The VMAT plans presented better rectal wall sparing, with a reduction of 1.5% in normal tissue complication probability. An advantage of VMAT plans was that the average number of MUs (290 MU) was less than for IMRT plans (642 MU). Conclusion: The VMAT technique can reduce beam on time by up to 55% while maintaining dosimetric quality comparable to that of the standard IMRT approach.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

User’s Manual for RESRAD-BUILD Code V.4: Vol. 1 – Methodology and Models Used in RESRAD-BUILD Code

The RESRAD-BUILD computer code models radionuclide release and transport in indoor environments and performs pathway analyses to evaluate the potential radiological dose and risk incurred by an individual who works or lives in a building contaminated with radioactive material or housing radioactively contaminated furniture or equipment. The code provides four geometries to characterize a radiation source: point, line, area, and volume, in which radionuclides are homogeneously distributed. Radionuclides contained in a source are considered to be released to the indoor air due to various processes including erosion (mechanically or weathering), diffusion (for tritium and radon in a volume source), or emanation (radon in a point, line, or area source). The release can proceed through different time phases with different rates. In RESRAD-BUILD Version 4.0, a dynamic ventilation model is implemented to simulate the fate and transport of source material particles and radionuclides after their releases. This dynamic ventilation model considers (1) air exchange between rooms in the building and between the rooms and the outdoor environment, (2) deposition from air to floor, (3) resuspension from the floor to the air, and (4) periodical vacuuming that reduces the floor deposition. The fate and transport modeling provides estimates of radionuclide concentrations in the source, in the air, and on the floor at different times, which are then integrated over the exposure duration for the calculation of radiation doses and cancer risks. A single run of the RESRAD-BUILD code can model a building with up to 9 rooms, 10 sources, and 10 receptors. The potential radiation dose and cancer risk incurred by each receptor are calculated for seven exposure pathways: (1) external radiation directly from the sources (accounting for shielding), (2) external radiation from radioactive particles deposited on the floors, (3) external radiation from airborne radionuclides, (4) inhalation of airborne radionuclides, (5) inhalation of radon and radon progenies, (6) inadvertent ingestion of radioactive particles directly from the source, and (7) ingestion of radioactive particles deposited on the floors. Various exposure scenarios can be modeled with RESRAD-BUILD, including but are not limited to, office worker, renovation worker, decontamination worker, building visitor, and resident. Both deterministic and probabilistic analyses can be performed to obtain results in both text reports and graphic displays.

61 RADIATION PROTECTION AND DOSIMETRY↗

NaroNet: Discovery of tumor microenvironment elements from highly multiplexed images

Understanding the spatial interactions between the elements of the tumor microenvironment -i.e. tumor cells. fibroblasts, immune cells- and how these interactions relate to the diagnosis or prognosis of a tumor is one of the goals of computational pathology. We present NaroNet, a deep learning framework that models the multi-scale tumor microenvironment from multiplex-stained cancer tissue images and provides patient-level interpretable predictions using a seamless end-to-end learning pipeline. Trained only with multiplex-stained tissue images and their corresponding patient-level clinical labels, NaroNet unsupervisedly learns which cell phenotypes, cell neighborhoods, and neighborhood interactions have the highest influence to predict the correct label. To this end, NaroNet incorporates several novel and state-of-the-art deep learning techniques, such as patch-level contrastive learning, multi-level graph embeddings, a novel max-sum pooling operation, or a metric that quantifies the relevance that each microenvironment element has in the individual predictions. We validate NaroNet using synthetic data simulating multiplex-immunostained images where a patient label is artificially associated to the -adjustable- probabilistic incidence of different microenvironment elements. We then apply our model to two sets of images of human cancer tissues: 336 seven-color multiplex-immunostained images from 12 high-grade endometrial cancer patients; and 382 35-plex mass cytometry images from 215 breast cancer patients. In both synthetic and real datasets, NaroNet provides outstanding predictions of relevant clinical information while associating those predictions to the presence of specific microenvironment elements.

60 APPLIED LIFE SCIENCES↗

SU-E-T-206: Improving Radiotherapy Toxicity Based On Artificial Neural Network (ANN) for Head and Neck Cancer Patients

Purpose/Objective(s): The aim of this study is to build the estimator of toxicity using artificial neural network (ANN) for head and neck cancer patients Materials/Methods: An ANN can combine variables into a predictive model during training and considered all possible correlations of variables. We constructed an ANN based on the data from 73 patients with advanced H and N cancer treated with external beam radiotherapy and/or chemotherapy at our institution. For the toxicity estimator we defined input data including age, sex, site, stage, pathology, status of chemo, technique of external beam radiation therapy (EBRT), length of treatment, dose of EBRT, status of post operation, length of follow-up, the status of local recurrences and distant metastasis. These data were digitized based on the significance and fed to the ANN as input nodes. We used 20 hidden nodes (for the 13 input nodes) to take care of the correlations of input nodes. For training ANN, we divided data into three subsets such as training set, validation set and test set. Finally, we built the estimator for the toxicity from ANN output. Results: We used 13 input variables including the status of local recurrences and distant metastasis and 20 hidden nodes for correlations. 59 patients for training set, 7 patients for validation set and 7 patients for test set and fed the inputs to Matlab neural network fitting tool. We trained the data within 15% of errors of outcome. In the end we have the toxicity estimation with 74% of accuracy. Conclusion: We proved in principle that ANN can be a very useful tool for predicting the RT outcomes for high risk H and N patients. Currently we are improving the results using cross validation.

60 APPLIED LIFE SCIENCES↗

An insight into statistical refractive index properties of cell internal structure via low-coherence statistical amplitude microscopy

Refractive index properties, especially at the nanoscale, have shown great potential in cancer diagnosis and screening. Due to the intrinsic complexity and weak refractive index fluctuation, the reconstruction of internal structures of a biological cell has been challenging. In this paper, we propose a simple and practical approach to derive the statistical properties of internal refractive index fluctuations within a biological cell with a new optical microscopy method – Low-coherence Statistical Amplitude Microscopy (SAM). We validated the capability of SAM to characterize the statistical properties of cell internal structures, which is described by numerical models of one-dimensional Gaussian random field. We demonstrated the potential of SAM in cancer detection with an animal model of intestinal carcinogenesis – multiple intestinal neoplasia mouse model. We showed that SAM-derived statistical properties of cell nuclear structures could detect the subtle changes that are otherwise undetectable by conventional cytopathology.

59 BASIC BIOLOGICAL SCIENCES↗

Impact of Statistical Learning Methods on the Predictive Power of Multivariate Normal Tissue Complication Probability Models

Purpose: To study the impact of different statistical learning methods on the prediction performance of multivariate normal tissue complication probability (NTCP) models. Methods and Materials: In this study, three learning methods, stepwise selection, least absolute shrinkage and selection operator (LASSO), and Bayesian model averaging (BMA), were used to build NTCP models of xerostomia following radiotherapy treatment for head and neck cancer. Performance of each learning method was evaluated by a repeated cross-validation scheme in order to obtain a fair comparison among methods. Results: It was found that the LASSO and BMA methods produced models with significantly better predictive power than that of the stepwise selection method. Furthermore, the LASSO method yields an easily interpretable model as the stepwise method does, in contrast to the less intuitive BMA method. Conclusions: The commonly used stepwise selection method, which is simple to execute, may be insufficient for NTCP modeling. The LASSO method is recommended.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Overview of the Graphical User Interface for the GERMcode (GCR Event-Based Risk Model)

The descriptions of biophysical events from heavy ions are of interest in radiobiology, cancer therapy, and space exploration. The biophysical description of the passage of heavy ions in tissue and shielding materials is best described by a stochastic approach that includes both ion track structure and nuclear interactions. A new computer model called the GCR Event-based Risk Model (GERM) code was developed for the description of biophysical events from heavy ion beams at the NASA Space Radiation Laboratory (NSRL). The GERMcode calculates basic physical and biophysical quantities of high-energy protons and heavy ions that have been studied at NSRL for the purpose of simulating space radiobiological effects. For mono-energetic beams, the code evaluates the linear-energy transfer (LET), range (R), and absorption in tissue equivalent material for a given Charge (Z), Mass Number (A) and kinetic energy (E) of an ion. In addition, a set of biophysical properties are evaluated such as the Poisson distribution of ion or delta-ray hits for a specified cellular area, cell survival curves, and mutation and tumor probabilities. The GERMcode also calculates the radiation transport of the beam line for either a fixed number of user-specified depths or at multiple positions along the Bragg curve of the particle. The contributions from primary ion and nuclear secondaries are evaluated. The GERMcode accounts for the major nuclear interaction processes of importance for describing heavy ion beams, including nuclear fragmentation, elastic scattering, and knockout-cascade processes by using the quantum multiple scattering fragmentation (QMSFRG) model. The QMSFRG model has been shown to be in excellent agreement with available experimental data for nuclear fragmentation cross sections, and has been used by the GERMcode for application to thick target experiments. The GERMcode provides scientists participating in NSRL experiments with the data needed for the interpretation of their experiments, including the ability to model the beam line, the shielding of samples and sample holders, and the estimates of basic physical and biological outputs of the designed experiments. We present an overview of the GERMcode GUI, as well as providing training applications.

Kim, Myung-Hee Y.↗

Overview of the Graphical User Interface for the GERM Code (GCR Event-Based Risk Model

The descriptions of biophysical events from heavy ions are of interest in radiobiology, cancer therapy, and space exploration. The biophysical description of the passage of heavy ions in tissue and shielding materials is best described by a stochastic approach that includes both ion track structure and nuclear interactions. A new computer model called the GCR Event-based Risk Model (GERM) code was developed for the description of biophysical events from heavy ion beams at the NASA Space Radiation Laboratory (NSRL). The GERM code calculates basic physical and biophysical quantities of high-energy protons and heavy ions that have been studied at NSRL for the purpose of simulating space radiobiological effects. For mono-energetic beams, the code evaluates the linear-energy transfer (LET), range (R), and absorption in tissue equivalent material for a given Charge (Z), Mass Number (A) and kinetic energy (E) of an ion. In addition, a set of biophysical properties are evaluated such as the Poisson distribution of ion or delta-ray hits for a specified cellular area, cell survival curves, and mutation and tumor probabilities. The GERM code also calculates the radiation transport of the beam line for either a fixed number of user-specified depths or at multiple positions along the Bragg curve of the particle. The contributions from primary ion and nuclear secondaries are evaluated. The GERM code accounts for the major nuclear interaction processes of importance for describing heavy ion beams, including nuclear fragmentation, elastic scattering, and knockout-cascade processes by using the quantum multiple scattering fragmentation (QMSFRG) model. The QMSFRG model has been shown to be in excellent agreement with available experimental data for nuclear fragmentation cross sections, and has been used by the GERM code for application to thick target experiments. The GERM code provides scientists participating in NSRL experiments with the data needed for the interpretation of their experiments, including the ability to model the beam line, the shielding of samples and sample holders, and the estimates of basic physical and biological outputs of the designed experiments. We present an overview of the GERM code GUI, as well as providing training applications.

Kim, Myung-Hee↗

Radiological risk assessment of environmental radon

Measurements of radon gas ({sup 222}Rn) in the environmental are important to assess indoor air quality and to study the potential risk to human health. Generally known that exposure to radon is considered the second leading cause of lung cancer after smoking. The environmental radon concentration depends on the {sup 226}Ra concentration, indoor atmosphere, cracking on rocks and building materials. This study was carried out to determine the indoor radon concentration from selected samples of tin tailings (amang) and building materials in an airtight sealed homemade radon chamber. The radiological risk assessment for radon gas was also calculated based on the annual exposure dose, effective dose equivalent, radon exhalation rates and fatal cancer risk. The continuous radon monitor Sun Nuclear model 1029 was used to measure the radon concentration emanates from selected samples for 96 hours. Five types of tin tailings collected from Kampar, Perak and four samples of building materials commonly used in Malaysia dwellings or building constructions were analysed for radon concentration. The indoor radon concentration determined in ilmenite, monazite, struverite, xenotime and zircon samples varies from 219.6 ± 76.8 Bq m{sup −3} to 571.1 ± 251.4 Bq m{sup −3}, 101.0 ± 41.0 Bq m{sup −3} to 245.3 ± 100.2 Bq m{sup −3}, 53.1 ± 7.5 Bq m{sup −3} to 181.8 ± 9.7 Bq m{sup −3}, 256.1 ± 59.3 Bq m{sup −3} to 652.2 ± 222.2 Bq m{sup −3} and 164.5 ± 75.9 Bq m{sup −3} to 653.3 ± 240.0 Bq m{sup −3}, respectively. Whereas, in the building materials, the radon concentration from cement brick, red-clay brick, gravel aggregate and cement showed 396.3 ± 194.3 Bq m{sup −3}, 192.1 ± 75.4 Bq m{sup −3}, 176.1 ± 85.9 Bq m{sup −3} and 28.4 ± 5.7 Bq m{sup −3}, respectively. The radon concentration in tin tailings and building materials were found to be much higher in xenotime and cement brick samples than others. All samples in tin tailings were exceeded the action level for radon gas of 148 Bq m{sup −3} proposed by EPA except monazite 0.15 kg, struverite 0.15 kg and 0.25 kg. Whereas, all building material samples have exceeded the radon concentration in concrete and building materials of 3 to 7 Bq m{sup −3} estimated by ICRP. The annual effective dose, effective dose equivalent, and radon exhalation rates in tin tailings were calculated to be in the range of 2.47 to 11.46 mSv, 5.94 to 1090.56 mSv y{sup −1}, and 0.23 to 1.18 mBq kg{sup −1} h{sup −1}. For building materials, the calculated risk assessment of the annual effective dose, effective dose equivalent, radon exhalation rates and fatal cancer risk were 0.72 to 10.00 mSv, 1.73 to 24.00 mSv y{sup −1}, 0.010 to 0.06 mBq kg{sup −1} h{sup −1} and 40 to 550 chances of persons will suffer the cancer per million (1 × 10{sup 6}), respectively.

61 RADIATION PROTECTION AND DOSIMETRY↗

Deep active learning for classifying cancer pathology reports

Abstract Background Automated text classification has many important applications in the clinical setting; however, obtaining labelled data for training machine learning and deep learning models is often difficult and expensive. Active learning techniques may mitigate this challenge by reducing the amount of labelled data required to effectively train a model. In this study, we analyze the effectiveness of 11 active learning algorithms on classifying subsite and histology from cancer pathology reports using a Convolutional Neural Network as the text classification model. Results We compare the performance of each active learning strategy using two differently sized datasets and two different classification tasks. Our results show that on all tasks and dataset sizes, all active learning strategies except diversity-sampling strategies outperformed random sampling, i.e., no active learning. On our large dataset (15K initial labelled samples, adding 15K additional labelled samples each iteration of active learning), there was no clear winner between the different active learning strategies. On our small dataset (1K initial labelled samples, adding 1K additional labelled samples each iteration of active learning), marginal and ratio uncertainty sampling performed better than all other active learning techniques. We found that compared to random sampling, active learning strongly helps performance on rare classes by focusing on underrepresented classes. Conclusions Active learning can save annotation cost by helping human annotators efficiently and intelligently select which samples to label. Our results show that a dataset constructed using effective active learning techniques requires less than half the amount of labelled data to achieve the same performance as a dataset constructed using random sampling.

59 BASIC BIOLOGICAL SCIENCES↗

Real-time out-of-plane artifact subtraction tomosynthesis imaging using prior CT for scanning beam digital x-ray system

Purpose: The scanning beam digital x-ray system (SBDX) is an inverse geometry fluoroscopic system with high dose efficiency and the ability to perform continuous real-time tomosynthesis in multiple planes. This system could be used for image guidance during lung nodule biopsy. However, the reconstructed images suffer from strong out-of-plane artifact due to the small tomographic angle of the system. Methods: The authors propose an out-of-plane artifact subtraction tomosynthesis (OPAST) algorithm that utilizes a prior CT volume to augment the run-time image processing. A blur-and-add (BAA) analytical model, derived from the project-to-backproject physical model, permits the generation of tomosynthesis images that are a good approximation to the shift-and-add (SAA) reconstructed image. A computationally practical algorithm is proposed to simulate images and out-of-plane artifacts from patient-specific prior CT volumes using the BAA model. A 3D image registration algorithm to align the simulated and reconstructed images is described. The accuracy of the BAA analytical model and the OPAST algorithm was evaluated using three lung cancer patients’ CT data. The OPAST and image registration algorithms were also tested with added nonrigid respiratory motions. Results: Image similarity measurements, including the correlation coefficient, mean squared error, and structural similarity index, indicated that the BAA model is very accurate in simulating the SAA images from the prior CT for the SBDX system. The shift-variant effect of the BAA model can be ignored when the shifts between SBDX images and CT volumes are within ±10 mm in the x and y directions. The nodule visibility and depth resolution are improved by subtracting simulated artifacts from the reconstructions. The image registration and OPAST are robust in the presence of added respiratory motions. The dominant artifacts in the subtraction images are caused by the mismatches between the real object and the prior CT volume. Conclusions: Their proposed prior CT-augmented OPAST reconstruction algorithm improves lung nodule visibility and depth resolution for the SBDX system.

60 APPLIED LIFE SCIENCES↗

2,3,7,8-Tetrachlorodibenzo-p-dioxin has both pro-carcinogenic and anti-carcinogenic effects on neuroendocrine prostate carcinoma formation in TRAMP mice

It is well established that the prototypical aryl hydrocarbon receptor (AHR) agonist 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) can both cause and protect against carcinogenesis in non-transgenic rodents. But because these animals almost never develop prostate cancer with old age or after carcinogen exposure, whether AHR activation can affect cancer of the prostate remained unknown. We used animals designed to develop this disease, Transgenic Adenocarcinoma of the Mouse Prostate (TRAMP) mice, to investigate the potential role of AHR signaling in prostate cancer development. We previously reported that AHR itself has prostate tumor suppressive functions in TRAMP mice; i.e., TRAMP mice in which Ahr was knocked out developed neuroendocrine prostate carcinomas (NEPC) with much greater frequency than did those with both Ahr alleles. In the present study we investigated effects of AHR activation by three different xenobiotics. In utero and lactational TCDD exposure significantly increased NEPC tumor incidence in TRAMP males, while chronic TCDD treatment in adulthood had the opposite effect, a significant reduction in NEPC incidence. Chronic treatment of adult TRAMP mice with the low-toxicity selective AHR modulators indole-3-carbinol or 3,3′-diindolylmethane did not significantly protect against these tumors. Thus, we demonstrate, for the first time, that ligand-dependent activation of the AHR can alter prostate cancer incidence. The nature of the responses depended on the timing of AHR activation and ligand structures. - Highlights: • TRAMP mice model aggressive neuroendocrine prostate carcinomas in men • In utero/lactational TCDD exposure raised prostate cancer incidence in TRAMP mice. • TCDD treatment in adulthood lowered prostate cancer incidence in TRAMP mice. • No significant protection was seen in TRAMP mice given I3C or DIM in adulthood. • This is the first report that TCDD alters prostate cancer incidence in lab animals.

60 APPLIED LIFE SCIENCES↗

The influence of TRP53 in the dose response of radiation-induced apoptosis, DNA repair and genomic stability in murine haematopoietic cells

Apoptotic and DNA damage endpoints are frequently used as surrogate markers of cancer risk, and have been well-studied in the Trp53+/- mouse model. We report the effect of differing Trp53 gene status on the dose response of ionizing radiation exposures (0.01-2 Gy), with the unique perspective of determining if effects of gene status remain at extended time points. Here we report no difference in the dose response for radiation-induced DNA double-strand breaks in bone marrow and genomic instability (MN-RET levels) in peripheral blood, between wild-type ( Trp53+/+ ) and heterozygous ( Trp53+/- ) mice. The dose response for Trp53+/+ mice showed higher initial levels of radiation-induced lymphocyte apoptosis relative to Trp53+/- between 0 and 1 Gy. Although this trend was observed up to 12 hours post-irradiation, both genotypes ultimately reached the same level of apoptosis at 14 hours, suggesting the importance of late-onset p53-independent apoptotic responses in this mouse model. Expected radiation-induced G1 cell cycle delay was observed in Trp53+/+ but not Trp53+/- . Although p53 has an important role in cancer risk, we have shown its influence on radiation dose response can be temporally variable. This research highlights the importance of caution when using haematopoietic endpoints as surrogates to extrapolate radiation-induced cancer risk estimation.

59 BASIC BIOLOGICAL SCIENCES↗

Phenotype heterogeneity in cancer cell populations

Phenotype heterogeneity in cancer cell populations, be it of genetic, epigenetic or stochastic origin, has been identified as a main source of resistance to drug treatments and a major source of therapeutic failures in cancers. The molecular mechanisms of drug resistance are partly understood at the single cell level (e.g., overexpression of ABC transporters or of detoxication enzymes), but poorly predictable in tumours, where they are hypothesised to rely on heterogeneity at the cell population scale, which is thus the right level to describe cancer growth and optimise its control by therapeutic strategies in the clinic. We review a few results from the biological literature on the subject, and from mathematical models that have been published to predict and control evolution towards drug resistance in cancer cell populations. We propose, based on the latter, optimisation strategies of combined treatments to limit emergence of drug resistance to cytotoxic drugs in cancer cell populations, in the monoclonal situation, which limited as it is still retains consistent features of cell population heterogeneity. The polyclonal situation, that may be understood as “bet hedging” of the tumour, thus protecting itself from different sources of drug insults, may lie beyond such strategies and will need further developments. In the monoclonal situation, we have designed an optimised therapeutic strategy relying on a scheduled combination of cytotoxic and cytostatic treatments that can be adapted to different situations of cancer treatments. Finally, we review arguments for biological theoretical frameworks proposed at different time and development scales, the so-called atavistic model (diachronic view relying on Darwinian genotype selection in the coursof billions of years) and the Waddington-like epigenetic landscape endowed with evolutionary quasi-potential (synchronic view relying on Lamarckian phenotype instruction of a given genome by reversible mechanisms), to represent evolution towards heterogeneity, possibly polyclonal, in cancer cell populations and propose innovative directions for therapeutic strategies based on such frameworks.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Feature‐shared adaptive‐boost deep learning for invasiveness classification of pulmonary subsolid nodules in CT images

Purpose In clinical practice, invasiveness is an important reference indicator for differentiating the malignant degree of subsolid pulmonary nodules. These nodules can be classified as atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IAC). The automatic determination of a nodule's invasiveness based on chest CT scans can guide treatment planning. However, it is challenging, owing to the insufficiency of training data and their interclass similarity and intraclass variation. To address these challenges, we propose a two‐stage deep learning strategy for this task: prior‐feature learning followed by adaptive‐boost deep learning. Methods The adaptive‐boost deep learning is proposed to train a strong classifier for invasiveness classification of subsolid nodules in chest CT images, using multiple 3D convolutional neural network (CNN)‐based weak classifiers. Because ensembles of multiple deep 3D CNN models have a huge number of parameters and require large computing resources along with more training and testing time, the prior‐feature learning is proposed to reduce the computations by sharing the CNN layers between all weak classifiers. Using this strategy, all weak classifiers can be integrated into a single network. Results Tenfold cross validation of binary classification was conducted on a total of 1357 nodules, including 765 noninvasive (AAH and AIS) and 592 invasive nodules (MIA and IAC). Ablation experimental results indicated that the proposed binary classifier achieved an accuracy of with an AUC of 81.3 . These results are superior compared to those achieved by three experienced chest imaging specialists who achieved an accuracy of , , and , respectively. About 200 additional nodules were also collected. These nodules covered 50 cases for each category (AAH, AIS, MIA, and IAC, respectively). Both binary and multiple classifications were performed on these data and the results demonstrated that the proposed method definitely achieves better performance than the performance achieved by nonensemble deep learning methods. Conclusions It can be concluded that the proposed adaptive‐boost deep learning can significantly improve the performance of invasiveness classification of pulmonary subsolid nodules in CT images, while the prior‐feature learning can significantly reduce the total size of deep models. The promising results on clinical data show that the trained models can be used as an effective lung cancer screening tool in hospitals. Moreover, the proposed strategy can be easily extended to other similar classification tasks in 3D medical images.

Wang, Jun↗

Late Rectal Toxicity on RTOG 94-06: Analysis Using a Mixture Lyman Model

Purpose: To estimate the parameters of the Lyman normal-tissue complication probability model using censored time-to-event data for Grade {>=}2 late rectal toxicity among patients treated on Radiation Therapy Oncology Group 94-06, a dose-escalation trial designed to determine the maximum tolerated dose for three-dimensional conformal radiotherapy of prostate cancer. Methods and Materials: The Lyman normal-tissue complication probability model was fitted to data from 1,010 of the 1,084 patients accrued on Radiation Therapy Oncology Group 94-06 using an approach that accounts for censored observations. Separate fits were obtained using dose-volume histograms for whole rectum and dose-wall histograms for rectal wall. Results: With a median follow-up of 7.2 years, the crude incidence of Grade {>=}2 late rectal toxicity was 15% (n = 148). The parameters of the Lyman model fitted to dose-volume histograms data, with 95% profile-likelihood confidence intervals, were TD{sub 50} = 79.1 Gy (75.3 Gy, 84.3 Gy), m = 0.146 (0.107, 0.225), and n = 0.077 (0.041, 0.156). The fit based on dose-wall histogram data was not significantly different. Patients with cardiovascular disease had a significantly higher incidence of late rectal toxicity (p = 0.015), corresponding to a dose-modifying factor of 5.3%. No significant association with late rectal toxicity was found for diabetes, hypertension, rectal volume, rectal length, neoadjuvant hormone therapy, or prescribed dose per fraction (1.8 Gy vs. 2 Gy). Conclusions: These results, based on a large cohort of patients from a multi-institutional trial, are expected to be widely representative of the ability of the Lyman model to describe the long-term risk of Grade {>=}2 late rectal toxicity after three-dimensional conformal radiotherapy of prostate cancer.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Predicted Rates of Secondary Malignancies From Proton Versus Photon Radiation Therapy for Stage I Seminoma

Purpose: Photon radiotherapy has been the standard adjuvant treatment for stage I seminoma. Single-dose carboplatin therapy and observation have emerged as alternative options due to concerns for acute toxicities and secondary malignancies from radiation. In this institutional review board-approved study, we compared photon and proton radiotherapy for stage I seminoma and the predicted rates of excess secondary malignancies for both treatment modalities. Methods and Material: Computed tomography images from 10 consecutive patients with stage I seminoma were used to quantify dosimetric differences between photon and proton therapies. Structures reported to be at increased risk for secondary malignancies and in-field critical structures were contoured. Reported models of organ-specific radiation-induced cancer incidence rates based on organ equivalent dose were used to determine the excess absolute risk of secondary malignancies. Calculated values were compared with tumor registry reports of excess secondary malignancies among testicular cancer survivors. Results: Photon and proton plans provided comparable target volume coverage. Proton plans delivered significantly lower mean doses to all examined normal tissues, except for the kidneys. The greatest absolute reduction in mean dose was observed for the stomach (119 cGy for proton plans vs. 768 cGy for photon plans; p < 0.0001). Significantly more excess secondary cancers per 10,000 patients/year were predicted for photon radiation than for proton radiation to the stomach (4.11; 95% confidence interval [CI], 3.22-5.01), large bowel (0.81; 95% CI, 0.39-1.01), and bladder (0.03; 95% CI, 0.01-0.58), while no difference was demonstrated for radiation to the pancreas (0.02; 95% CI, -0.01-0.06). Conclusions: For patients with stage I seminoma, proton radiation therapy reduced the predicted secondary cancer risk compared with photon therapy. We predict a reduction of one additional secondary cancer for every 50 patients with a life expectancy of 40 years from the time of radiation treatment with protons instead of photons. Proton radiation therapy also allowed significant sparing of most critical structures examined and warrants further study for patients with seminoma, to decrease radiation-induced toxicity.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Quantification of Contralateral Breast Dose and Risk Estimate of Radiation-Induced Contralateral Breast Cancer Among Young Women Using Tangential Fields and Different Modes of Breathing

Purpose: Whole breast irradiation with deep-inspiration breath-hold (DIBH) technique among left-sided breast cancer patients significantly reduces cardiac irradiation; however, a potential disadvantage is increased incidental irradiation of the contralateral breast. Methods and Materials: Contralateral breast dose (CBD) was calculated by comparing 400 treatment plans of 200 left-sided breast cancer patients whose tangential fields had been planned on gated and nongated CT data sets. Various anatomic and field parameters were analyzed for their impact on CBD. For a subgroup of patients (aged {<=}45 years) second cancer risk in the contralateral breast (CB) was modeled by applying the linear quadratic model, compound models, and compound models considering dose-volume information (DVH). Results: The mean CBD was significantly higher in DIBH with 0.69 Gy compared with 0.65 Gy in normal breathing (P=.01). The greatest impact on CBD was due to a shift of the inner field margin toward the CB in DIBH (mean 0.4 cm; range, 0-2), followed by field size in magnitude. Calculation with different risk models for CBC revealed values of excess relative risk/Gy ranging from 0.48-0.65 vs 0.46-0.61 for DIBH vs normal breathing, respectively. Conclusion: Contralateral breast dose, although within a low dose range, was mildly but significantly increased in 200 treatment plans generated under gated conditions, predominately due to a shift in the medial field margin. Risk modeling for CBC among women aged {<=}45 years also pointed to a higher risk when comparing DIBH with normal breathing. This risk, however, was substantially lower in the model considering DVH information. We think that clinical decisions should not be affected by this small increase in CBD with DIBH because DIBH is effective in reducing the dose to the heart in all patients.

62 RADIOLOGY AND NUCLEAR MEDICINE↗