Results for “Modal” — Page 91

DOE OSTI2012

Concurrent segmentation of the prostate on MRI and CT via linked statistical shape models for radiotherapy planning

Purpose: Prostate gland segmentation is a critical step in prostate radiotherapy planning, where dose plans are typically formulated on CT. Pretreatment MRI is now beginning to be acquired at several medical centers. Delineation of the prostate on MRI is acknowledged as being significantly simpler to perform, compared to delineation on CT. In this work, the authors present a novel framework for building a linked statistical shape model (LSSM), a statistical shape model (SSM) that links the shape variation of a structure of interest (SOI) across multiple imaging modalities. This framework is particularly relevant in scenarios where accurate boundary delineations of the SOI on one of the modalities may not be readily available, or difficult to obtain, for training a SSM. In this work the authors apply the LSSM in the context of multimodal prostate segmentation for radiotherapy planning, where the prostate is concurrently segmented on MRI and CT. Methods: The framework comprises a number of logically connected steps. The first step utilizes multimodal registration of MRI and CT to map 2D boundary delineations of the prostate from MRI onto corresponding CT images, for a set of training studies. Hence, the scheme obviates the need for expert delineations of the gland on CT for explicitly constructing a SSM for prostate segmentation on CT. The delineations of the prostate gland on MRI and CT allows for 3D reconstruction of the prostate shape which facilitates the building of the LSSM. In order to perform concurrent prostate MRI and CT segmentation using the LSSM, the authors employ a region-based level set approach where the authors deform the evolving prostate boundary to simultaneously fit to MRI and CT images in which voxels are classified to be either part of the prostate or outside the prostate. The classification is facilitated by using a combination of MRI-CT probabilistic spatial atlases and a random forest classifier, driven by gradient and Haar features. Results: The authors acquire a total of 20 MRI-CT patient studies and use the leave-one-out strategy to train and evaluate four different LSSMs. First, a fusion-based LSSM (fLSSM) is built using expert ground truth delineations of the prostate on MRI alone, where the ground truth for the gland on CT is obtained via coregistration of the corresponding MRI and CT slices. The authors compare the fLSSM against another LSSM (xLSSM), where expert delineations of the gland on both MRI and CT are employed in the model building; xLSSM representing the idealized LSSM. The authors also compare the fLSSM against an exclusive CT-based SSM (ctSSM), built from expert delineations of the gland on CT alone. In addition, two LSSMs trained using trainee delineations (tLSSM) on CT are compared with the fLSSM. The results indicate that the xLSSM, tLSSMs, and the fLSSM perform equivalently, all of them out-performing the ctSSM. Conclusions: The fLSSM provides an accurate alternative to SSMs that require careful expert delineations of the SOI that may be difficult or laborious to obtain. Additionally, the fLSSM has the added benefit of providing concurrent segmentations of the SOI on multiple imaging modalities.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2012

Characterization of the homogeneous tissue mixture approximation in breast imaging dosimetry

Purpose: To compare the estimate of normalized glandular dose in mammography and breast CT imaging obtained using the actual glandular tissue distribution in the breast to that obtained using the homogeneous tissue mixture approximation. Methods: Twenty volumetric images of patient breasts were acquired with a dedicated breast CT prototype system and the voxels in the breast CT images were automatically classified into skin, adipose, and glandular tissue. The breasts in the classified images underwent simulated mechanical compression to mimic the conditions present during mammographic acquisition. The compressed thickness for each breast was set to that achieved during each patient's last screening cranio-caudal (CC) acquisition. The volumetric glandular density of each breast was computed using both the compressed and uncompressed classified images, and additional images were created in which all voxels representing adipose and glandular tissue were replaced by a homogeneous mixture of these two tissues in a proportion corresponding to each breast's volumetric glandular density. All four breast images (compressed and uncompressed; heterogeneous and homogeneous tissue) were input into Monte Carlo simulations to estimate the normalized glandular dose during mammography (compressed breasts) and dedicated breast CT (uncompressed breasts). For the mammography simulations the x-ray spectra used was that used during each patient's last screening CC acquisition. For the breast CT simulations, two x-ray spectra were used, corresponding to the x-ray spectra with the lowest and highest energies currently being used in dedicated breast CT prototype systems under clinical investigation. The resulting normalized glandular dose for the heterogeneous and homogeneous versions of each breast for each modality was compared. Results: For mammography, the normalized glandular dose based on the homogeneous tissue approximation was, on average, 27% higher than that estimated using the true heterogeneous glandular tissue distribution (Wilcoxon Signed Rank Test p= 0.00046). For dedicated breast CT, the overestimation of normalized glandular dose was, on average, 8% (49 kVp spectrum, p= 0.00045) and 4% (80 kVp spectrum, p= 0.000089). Only two cases in mammography and two cases in dedicated breast CT with a tube voltage of 49 kVp resulted in lower dose estimates for the homogeneous tissue approximation compared to the heterogeneous tissue distribution. Conclusions: The normalized glandular dose based on the homogeneous tissue mixture approximation results in a significant overestimation of dose to the imaged breast. This overestimation impacts the use of dose estimates in absolute terms, such as for risk estimates, and may impact some comparative studies, such as when modalities or techniques with different x-ray energies are used. The error introduced by the homogeneous tissue mixture approximation in higher energy x-ray modalities, such as dedicated breast CT, although statistically significant, may not be of clinical concern. Further work is required to better characterize this overestimation and potentially develop new metrics or correction factors to better estimate the true glandular dose to breasts undergoing imaging with ionizing radiation.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2013

3D inpatient dose reconstruction from the PET-CT imaging of {sup 90}Y microspheres for metastatic cancer to the liver: Feasibility study

Purpose: The introduction of radioembolization with microspheres represents a significant step forward in the treatment of patients with metastatic disease to the liver. This technique uses semiempirical formulae based on body surface area or liver and target volumes to calculate the required total activity for a given patient. However, this treatment modality lacks extremely important information, which is the three-dimensional (3D) dose delivered by microspheres to different organs after their administration. The absence of this information dramatically limits the clinical efficacy of this modality, specifically the predictive power of the treatment. Therefore, the aim of this study is to develop a 3D dose calculation technique that is based on the PET imaging of the infused microspheres.Methods: The Fluka Monte Carlo code was used to calculate the voxel dose kernel for {sup 90}Y source with voxel size equal to that of the PET scan. The measured PET activity distribution was converted to total activity distribution for the subsequent convolution with the voxel dose kernel to obtain the 3D dose distribution. In addition, dose-volume histograms were generated to analyze the dose to the tumor and critical structures.Results: The 3D inpatient dose distribution can be reconstructed from the PET data of a patient scanned after the infusion of microspheres. A total of seven patients have been analyzed so far using the proposed reconstruction method. Four patients underwent treatment with SIR-Spheres for liver metastases from colorectal cancer and three patients were treated with Therasphere for hepatocellular cancer. A total of 14 target tumors were contoured on post-treatment PET-CT scans for dosimetric evaluation. Mean prescription activity was 1.7 GBq (range: 0.58–3.8 GBq). The resulting mean maximum measured dose to targets was 167 Gy (range: 71–311 Gy). Mean minimum dose to 70% of target (D70) was 68 Gy (range: 25–155 Gy). Mean minimum dose to 90% of target (D90) was 53 Gy (range: 13–125 Gy).Conclusions: A three-dimensional inpatient dose reconstruction method has been developed that is based on the PET/CT data of a patient treated with {sup 90}Y microspheres. It allows for a complete description of the absorbed dose by the tumor and critical structures. It represents the first step in building predictive models for treatment outcomes for patients receiving this therapeutic modality as well as it allows for better analysis of patients' dose response and will ultimately improve future treatment administration.

61 RADIATION PROTECTION AND DOSIMETRY↗

DOE OSTI2014

SU-E-I-39: Molecular Image Guided Cancer Stem Cells Therapy

Purpose: Cancer stem cells resistance to radiation is a problematic issue that has caused a big fail in cancer treatment. Methods: As a primary work, molecular imaging can indicate the main mechanisms of radiation resistance of cancer stem cells. By developing and commissioning new probes and nanomolecules and biomarkers, radiation scientist will able to identify the essential pathways of radiation resistance of cancer stem cells. As the second solution, molecular imaging is a best way to find biological target volume and delineate cancer stem cell tissues. In the other hand, by molecular imaging techniques one can image the treatment response in tumor and also in normal tissue. In this issue, the response of cancer stem cells to radiation during therapy course can be imaged, also the main mechanisms of radiation resistance and finding the best radiation modifiers (sensitizers) can be achieved by molecular imaging modalities. In adaptive radiotherapy the molecular imaging plays a vital role to have higher tumor control probability by delivering high radiation doses to cancer stem cells in any time of treatment. The outcome of a feasible treatment is dependent to high cancer stem cells response to radiation and removing all of which, so a good imaging modality can show this issue and preventing of tumor recurrence and metastasis. Results: Our results are dependent to use of molecular imaging as a new modality in the clinic. We propose molecular imaging as a new radiobiological technique to solve radiation therapy problems due to cancer stem cells. Conclusion: Molecular imaging guided cancer stem cell diagnosis and therapy is a new approach in the field of cancer treatment. This new radiobiological imaging technique should be developed in all clinics as a feasible tool that is more biological than physical imaging.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

SU-E-J-86: Lobar Lung Function Quantification by PET Galligas and CT Ventilation Imaging in Lung Cancer Patients

Purpose: The purpose of this study was to quantify the lobar lung function using the novel PET Galligas ([68Ga]-carbon nanoparticle) ventilation imaging and the investigational CT ventilation imaging in lung cancer patients pre-treatment. Methods: We present results on our first three lung cancer patients (2 male, mean age 78 years) as part of an ongoing ethics approved study. For each patient a PET Galligas ventilation (PET-V) image and a pair of breath hold CT images (end-exhale and end-inhale tidal volumes) were acquired using a Siemens Biograph PET CT. CT-ventilation (CT-V) images were created from the pair of CT images using deformable image registration (DIR) algorithms and the Hounsfield Unit (HU) ventilation metric. A comparison of ventilation quantification from each modality was done on the lobar level and the voxel level. A Bland-Altman plot was used to assess the difference in mean percentage contribution of each lobe to the total lung function between the two modalities. For each patient, a voxel-wise Spearmans correlation was calculated for the whole lungs between the two modalities. Results: The Bland-Altman plot demonstrated strong agreement between PET-V and CT-V for assessment of lobar function (r=0.99, p<0.001; range mean difference: −5.5 to 3.0). The correlation between PET-V and CT-V at the voxel level was moderate(r=0.60, p<0.001). Conclusion: This preliminary study on the three patients data sets demonstrated strong agreement between PET and CT ventilation imaging for the assessment of pre-treatment lung function at the lobar level. Agreement was only moderate at the level of voxel correlations. These results indicate that CT ventilation imaging has potential for assessing pre-treatment lobar lung function in lung cancer patients.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

SU-E-T-28: A Treatment Planning Comparison of Volumetric Modulated Arc Therapy Vs. Proton Therapy for Post-Mastectomy Radiotherapy

Purpose: The delivery of post-mastectomy radiotherapy (PMRT) can be challenging for patients with left-sided breast cancer due to the PTV size and proximity to critical organs. This study investigates the use of protons for PMRT in a clinically-representative cohort of patients, and quantitatively compares volumetric modulated arc therapy (VMAT) to proton therapy to have an evidence-based rationale for selecting a treatment modality for these patients. Methods: Eight left-sided PMRT patients previously treated at our clinic with VMAT were identified for the study. PTVs included the chest wall and regional lymph nodes. Passively scattered (PS) and intensity modulated proton therapy (IMPT) plans were constructed using the Eclipse proton planning system. The resulting plans were compared to the original VMAT plan on the basis of PTV coverage; dose homogeneity index (DHI) and conformity index (CI); dose to organs at risk (OAR); tumor control probability (TCP), normal tissue complication probability (NTCP) and secondary cancer complication probability (SCCP). Differences were tested for significance using the paired Student's t-test (p<0.01). Results: All modalities produced clinically acceptable PMRT plans. The comparison demonstrated proton treatment plans provide significantly lower NTCP values for the heart and the lung while maintaining significantly better CI and DHI. At a prescribed dose of 50.4 Gy (RBE) in the PTV, the calculated mean NTCP value for the patients decreased from 1.3% to 0.05% for the whole heart (cardiac mortality) and from 3.8% to 1.1% for the lungs (radiation pneumonitis) for both proton therapy plans from VMAT plans. Both proton modalities showed a significantly lower SCCP for the contralateral breast compared to VMAT. Conclusion: All three plans (VMAT, PS, and IMPT) provide acceptable treatment plans for PMRT. However, proton therapy shows a significant advantage over VMAT with regards to sparing OARs and may be more advantageous for patients with unfavorable anatomy.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2014

MO-E-12A-01: Quantitative Imaging: Techniques, Applications, and Challenges

The first symposium in the Quantitative Imaging Track focused on the introduction of quantitative imaging (QI) by illustrating the potential of QI in diagnostic and therapeutic applications in research and patient care, highlighting key challenges in implementation of such QI applications, and reviewing QI efforts of selected national and international agencies and organizations, including the FDA, NCI, NIST, and RSNA. This second QI symposium will focus more specifically on the techniques, applications, and challenges of QI. The first talk of the session will focus on modalityagnostic challenges of QI, beginning with challenges of the development and implementation of QI applications in single-center, single-vendor settings and progressing to the challenges encountered in the most general setting of multi-center, multi-vendor settings. The subsequent three talks will focus on specific QI challenges and opportunities in the modalityspecific settings of CT, PET/CT, and MR. Each talk will provide information on modality-specific QI techniques, applications, and challenges, including current efforts focused on solutions to such challenges. Learning Objectives: Understand key general challenges of QI application development and implementation, regardless of modality. Understand selected QI techniques and applications in CT, PET/CT, and MR. Understand challenges, and potential solutions for such challenges, for the applications presented for each modality.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

SU-E-J-89: Comparative Analysis of MIM and Velocity’s Image Deformation Algorithm Using Simulated KV-CBCT Images for Quality Assurance

Purpose: Deformable image registration (DIR) is used routinely in the clinic without a formalized quality assurance (QA) process. Using simulated deformations to digitally deform images in a known way and comparing to DIR algorithm predictions is a powerful technique for DIR QA. This technique must also simulate realistic image noise and artifacts, especially between modalities. This study developed an algorithm to create simulated daily kV cone-beam computed-tomography (CBCT) images from CT images for DIR QA between these modalities. Methods: A Catphan and physical head-and-neck phantom, with known deformations, were used. CT and kV-CBCT images of the Catphan were utilized to characterize the changes in Hounsfield units, noise, and image cupping that occur between these imaging modalities. The algorithm then imprinted these changes onto a CT image of the deformed head-and-neck phantom, thereby creating a simulated-CBCT image. CT and kV-CBCT images of the undeformed and deformed head-and-neck phantom were also acquired. The Velocity and MIM DIR algorithms were applied between the undeformed CT image and each of the deformed CT, CBCT, and simulated-CBCT images to obtain predicted deformations. The error between the known and predicted deformations was used as a metric to evaluate the quality of the simulated-CBCT image. Ideally, the simulated-CBCT image registration would produce the same accuracy as the deformed CBCT image registration. Results: For Velocity, the mean error was 1.4 mm for the CT-CT registration, 1.7 mm for the CT-CBCT registration, and 1.4 mm for the CT-simulated-CBCT registration. These same numbers were 1.5, 4.5, and 5.9 mm, respectively, for MIM. Conclusion: All cases produced similar accuracy for Velocity. MIM produced similar values of accuracy for CT-CT registration, but was not as accurate for CT-CBCT registrations. The MIM simulated-CBCT registration followed this same trend, but overestimated MIM DIR errors relative to the CT-CBCT registration.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-CD-207-12: Impact of Anatomical Noise On Detection Performance of Microcalcifications in Multi-Contrast Breast Imaging

Purpose: The anatomical noise power spectra (NPS) for differential phase contrast (DPC) and dark field (DF) imaging have recently been characterized using a power-law model with two parameters, alpha and beta, an innovative extension to the methodology used in x-ray attenuation based breast imaging such as mammography, DBT, or cone-beam CT. Beta values of 3.6, 2.6, and 1.3 have been measured for absorption, DPC, and DF respectively for cadaver breasts imaged in the coronal plane; these dramatic differences should be reflected in their detection performance. The purpose of this study was to determine the impact of anatomical noise on breast calcification detection and compare the detection performance of the three contrast mechanisms of a multi-contrast x-ray imaging system. Methods: In our studies, a calcification image object was segmented out of the multi-contrast images of a cadaver breast specimen. 50 measured total NPS were measured from breast cadavers directly. The ideal model observer detectability was calculated for a range of doses (5–100%) and a range of calcification sizes (diameter = 0.25–2.5 mm). Results: Overall we found the highest average detectability corresponded to DPC imaging (7.4 for 1 mm calc.), with DF the next highest (3.8 for 1 mm calc.), and absorption the lowest (3.2 for 1 mm calc.). However, absorption imaging also showed the slowest dependence on dose of the three modalities due to the significant anatomical noise. DPC showed a peak detectability for calcifications ∼1.25 mm in diameter, DF showed a peak for calcifications around 0.75 mm in diameter, and absorption imaging had no such peak in the range explored. Conclusion: Understanding imaging performance for DPC and DF is critical to transition these modalities to the clinic. The results presented here offer new insight into how these modalities complement absorption imaging to maximize the likelihood of detecting early breast cancers. J. Garrett, Y. Ge, K. Li: Nothing to disclose. G.-H. Chen: Research funded, GE Healthcare; Research funded, Siemens AX.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-G-303-00: Radiomics: Advances in the Use of Quantitative Imaging Used for Predictive Modeling

‘Radiomics’ refers to studies that extract a large amount of quantitative information from medical imaging studies as a basis for characterizing a specific aspect of patient health. Radiomics models can be built to address a wide range of outcome predictions, clinical decisions, basic cancer biology, etc. For example, radiomics models can be built to predict the aggressiveness of an imaged cancer, cancer gene expression characteristics (radiogenomics), radiation therapy treatment response, etc. Technically, radiomics brings together quantitative imaging, computer vision/image processing, and machine learning. In this symposium, speakers will discuss approaches to radiomics investigations, including: longitudinal radiomics, radiomics combined with other biomarkers (‘pan-omics’), radiomics for various imaging modalities (CT, MRI, and PET), and the use of registered multi-modality imaging datasets as a basis for radiomics. There are many challenges to the eventual use of radiomics-derived methods in clinical practice, including: standardization and robustness of selected metrics, accruing the data required, building and validating the resulting models, registering longitudinal data that often involve significant patient changes, reliable automated cancer segmentation tools, etc. Despite the hurdles, results achieved so far indicate the tremendous potential of this general approach to quantifying and using data from medical images. Specific applications of radiomics to be presented in this symposium will include: the longitudinal analysis of patients with low-grade gliomas; automatic detection and assessment of patients with metastatic bone lesions; image-based monitoring of patients with growing lymph nodes; predicting radiotherapy outcomes using multi-modality radiomics; and studies relating radiomics with genomics in lung cancer and glioblastoma. Learning Objectives: Understanding the basic image features that are often used in radiomic models. Understanding requirements for reliable radiomic models, including robustness of metrics, adequate predictive accuracy, and generalizability. Understanding the methodology behind radiomic-genomic (’radiogenomics’) correlations. Research supported by NIH (US), CIHR (Canada), and NSERC (Canada)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-G-303-01: Radiomics: Quantitative Imaging in the Service of Improved Treatment Decision Making

‘Radiomics’ refers to studies that extract a large amount of quantitative information from medical imaging studies as a basis for characterizing a specific aspect of patient health. Radiomics models can be built to address a wide range of outcome predictions, clinical decisions, basic cancer biology, etc. For example, radiomics models can be built to predict the aggressiveness of an imaged cancer, cancer gene expression characteristics (radiogenomics), radiation therapy treatment response, etc. Technically, radiomics brings together quantitative imaging, computer vision/image processing, and machine learning. In this symposium, speakers will discuss approaches to radiomics investigations, including: longitudinal radiomics, radiomics combined with other biomarkers (‘pan-omics’), radiomics for various imaging modalities (CT, MRI, and PET), and the use of registered multi-modality imaging datasets as a basis for radiomics. There are many challenges to the eventual use of radiomics-derived methods in clinical practice, including: standardization and robustness of selected metrics, accruing the data required, building and validating the resulting models, registering longitudinal data that often involve significant patient changes, reliable automated cancer segmentation tools, etc. Despite the hurdles, results achieved so far indicate the tremendous potential of this general approach to quantifying and using data from medical images. Specific applications of radiomics to be presented in this symposium will include: the longitudinal analysis of patients with low-grade gliomas; automatic detection and assessment of patients with metastatic bone lesions; image-based monitoring of patients with growing lymph nodes; predicting radiotherapy outcomes using multi-modality radiomics; and studies relating radiomics with genomics in lung cancer and glioblastoma. Learning Objectives: Understanding the basic image features that are often used in radiomic models. Understanding requirements for reliable radiomic models, including robustness of metrics, adequate predictive accuracy, and generalizability. Understanding the methodology behind radiomic-genomic (’radiogenomics’) correlations. Research supported by NIH (US), CIHR (Canada), and NSERC (Canada)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-G-303-02: Robust Radiomics Methods for PET and CT Imaging

‘Radiomics’ refers to studies that extract a large amount of quantitative information from medical imaging studies as a basis for characterizing a specific aspect of patient health. Radiomics models can be built to address a wide range of outcome predictions, clinical decisions, basic cancer biology, etc. For example, radiomics models can be built to predict the aggressiveness of an imaged cancer, cancer gene expression characteristics (radiogenomics), radiation therapy treatment response, etc. Technically, radiomics brings together quantitative imaging, computer vision/image processing, and machine learning. In this symposium, speakers will discuss approaches to radiomics investigations, including: longitudinal radiomics, radiomics combined with other biomarkers (‘pan-omics’), radiomics for various imaging modalities (CT, MRI, and PET), and the use of registered multi-modality imaging datasets as a basis for radiomics. There are many challenges to the eventual use of radiomics-derived methods in clinical practice, including: standardization and robustness of selected metrics, accruing the data required, building and validating the resulting models, registering longitudinal data that often involve significant patient changes, reliable automated cancer segmentation tools, etc. Despite the hurdles, results achieved so far indicate the tremendous potential of this general approach to quantifying and using data from medical images. Specific applications of radiomics to be presented in this symposium will include: the longitudinal analysis of patients with low-grade gliomas; automatic detection and assessment of patients with metastatic bone lesions; image-based monitoring of patients with growing lymph nodes; predicting radiotherapy outcomes using multi-modality radiomics; and studies relating radiomics with genomics in lung cancer and glioblastoma. Learning Objectives: Understanding the basic image features that are often used in radiomic models. Understanding requirements for reliable radiomic models, including robustness of metrics, adequate predictive accuracy, and generalizability. Understanding the methodology behind radiomic-genomic (’radiogenomics’) correlations. Research supported by NIH (US), CIHR (Canada), and NSERC (Canada)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-G-303-03: Machine Learning to Improve Human Learning From Longitudinal Image Sets

‘Radiomics’ refers to studies that extract a large amount of quantitative information from medical imaging studies as a basis for characterizing a specific aspect of patient health. Radiomics models can be built to address a wide range of outcome predictions, clinical decisions, basic cancer biology, etc. For example, radiomics models can be built to predict the aggressiveness of an imaged cancer, cancer gene expression characteristics (radiogenomics), radiation therapy treatment response, etc. Technically, radiomics brings together quantitative imaging, computer vision/image processing, and machine learning. In this symposium, speakers will discuss approaches to radiomics investigations, including: longitudinal radiomics, radiomics combined with other biomarkers (‘pan-omics’), radiomics for various imaging modalities (CT, MRI, and PET), and the use of registered multi-modality imaging datasets as a basis for radiomics. There are many challenges to the eventual use of radiomics-derived methods in clinical practice, including: standardization and robustness of selected metrics, accruing the data required, building and validating the resulting models, registering longitudinal data that often involve significant patient changes, reliable automated cancer segmentation tools, etc. Despite the hurdles, results achieved so far indicate the tremendous potential of this general approach to quantifying and using data from medical images. Specific applications of radiomics to be presented in this symposium will include: the longitudinal analysis of patients with low-grade gliomas; automatic detection and assessment of patients with metastatic bone lesions; image-based monitoring of patients with growing lymph nodes; predicting radiotherapy outcomes using multi-modality radiomics; and studies relating radiomics with genomics in lung cancer and glioblastoma. Learning Objectives: Understanding the basic image features that are often used in radiomic models. Understanding requirements for reliable radiomic models, including robustness of metrics, adequate predictive accuracy, and generalizability. Understanding the methodology behind radiomic-genomic (’radiogenomics’) correlations. Research supported by NIH (US), CIHR (Canada), and NSERC (Canada)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

TU-G-303-04: Radiomics and the Coming Pan-Omics Revolution

‘Radiomics’ refers to studies that extract a large amount of quantitative information from medical imaging studies as a basis for characterizing a specific aspect of patient health. Radiomics models can be built to address a wide range of outcome predictions, clinical decisions, basic cancer biology, etc. For example, radiomics models can be built to predict the aggressiveness of an imaged cancer, cancer gene expression characteristics (radiogenomics), radiation therapy treatment response, etc. Technically, radiomics brings together quantitative imaging, computer vision/image processing, and machine learning. In this symposium, speakers will discuss approaches to radiomics investigations, including: longitudinal radiomics, radiomics combined with other biomarkers (‘pan-omics’), radiomics for various imaging modalities (CT, MRI, and PET), and the use of registered multi-modality imaging datasets as a basis for radiomics. There are many challenges to the eventual use of radiomics-derived methods in clinical practice, including: standardization and robustness of selected metrics, accruing the data required, building and validating the resulting models, registering longitudinal data that often involve significant patient changes, reliable automated cancer segmentation tools, etc. Despite the hurdles, results achieved so far indicate the tremendous potential of this general approach to quantifying and using data from medical images. Specific applications of radiomics to be presented in this symposium will include: the longitudinal analysis of patients with low-grade gliomas; automatic detection and assessment of patients with metastatic bone lesions; image-based monitoring of patients with growing lymph nodes; predicting radiotherapy outcomes using multi-modality radiomics; and studies relating radiomics with genomics in lung cancer and glioblastoma. Learning Objectives: Understanding the basic image features that are often used in radiomic models. Understanding requirements for reliable radiomic models, including robustness of metrics, adequate predictive accuracy, and generalizability. Understanding the methodology behind radiomic-genomic (’radiogenomics’) correlations. Research supported by NIH (US), CIHR (Canada), and NSERC (Canada)

60 APPLIED LIFE SCIENCES↗

DOE OSTI2015

Vision 20/20: Simultaneous CT-MRI — Next chapter of multimodality imaging

Multimodality imaging systems such as positron emission tomography-computed tomography (PET-CT) and MRI-PET are widely available, but a simultaneous CT-MRI instrument has not been developed. Synergies between independent modalities, e.g., CT, MRI, and PET/SPECT can be realized with image registration, but such postprocessing suffers from registration errors that can be avoided with synchronized data acquisition. The clinical potential of simultaneous CT-MRI is significant, especially in cardiovascular and oncologic applications where studies of the vulnerable plaque, response to cancer therapy, and kinetic and dynamic mechanisms of targeted agents are limited by current imaging technologies. The rationale, feasibility, and realization of simultaneous CT-MRI are described in this perspective paper. The enabling technologies include interior tomography, unique gantry designs, open magnet and RF sequences, and source and detector adaptation. Based on the experience with PET-CT, PET-MRI, and MRI-LINAC instrumentation where hardware innovation and performance optimization were instrumental to construct commercial systems, the authors provide top-level concepts for simultaneous CT-MRI to meet clinical requirements and new challenges. Simultaneous CT-MRI fills a major gap of modality coupling and represents a key step toward the so-called “omnitomography” defined as the integration of all relevant imaging modalities for systems biology and precision medicine.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

Automatic anatomy recognition in whole-body PET/CT images

Purpose: Whole-body positron emission tomography/computed tomography (PET/CT) has become a standard method of imaging patients with various disease conditions, especially cancer. Body-wide accurate quantification of disease burden in PET/CT images is important for characterizing lesions, staging disease, prognosticating patient outcome, planning treatment, and evaluating disease response to therapeutic interventions. However, body-wide anatomy recognition in PET/CT is a critical first step for accurately and automatically quantifying disease body-wide, body-region-wise, and organwise. This latter process, however, has remained a challenge due to the lower quality of the anatomic information portrayed in the CT component of this imaging modality and the paucity of anatomic details in the PET component. In this paper, the authors demonstrate the adaptation of a recently developed automatic anatomy recognition (AAR) methodology [Udupa et al., “Body-wide hierarchical fuzzy modeling, recognition, and delineation of anatomy in medical images,” Med. Image Anal. 18, 752–771 (2014)] to PET/CT images. Their goal was to test what level of object localization accuracy can be achieved on PET/CT compared to that achieved on diagnostic CT images. Methods: The authors advance the AAR approach in this work in three fronts: (i) from body-region-wise treatment in the work of Udupa et al. to whole body; (ii) from the use of image intensity in optimal object recognition in the work of Udupa et al. to intensity plus object-specific texture properties, and (iii) from the intramodality model-building-recognition strategy to the intermodality approach. The whole-body approach allows consideration of relationships among objects in different body regions, which was previously not possible. Consideration of object texture allows generalizing the previous optimal threshold-based fuzzy model recognition method from intensity images to any derived fuzzy membership image, and in the process, to bring performance to the level achieved on diagnostic CT and MR images in body-region-wise approaches. The intermodality approach fosters the use of already existing fuzzy models, previously created from diagnostic CT images, on PET/CT and other derived images, thus truly separating the modality-independent object assembly anatomy from modality-specific tissue property portrayal in the image. Results: Key ways of combining the above three basic ideas lead them to 15 different strategies for recognizing objects in PET/CT images. Utilizing 50 diagnostic CT image data sets from the thoracic and abdominal body regions and 16 whole-body PET/CT image data sets, the authors compare the recognition performance among these 15 strategies on 18 objects from the thorax, abdomen, and pelvis in object localization error and size estimation error. Particularly on texture membership images, object localization is within three voxels on whole-body low-dose CT images and 2 voxels on body-region-wise low-dose images of known true locations. Surprisingly, even on direct body-region-wise PET images, localization error within 3 voxels seems possible. Conclusions: The previous body-region-wise approach can be extended to whole-body torso with similar object localization performance. Combined use of image texture and intensity property yields the best object localization accuracy. In both body-region-wise and whole-body approaches, recognition performance on low-dose CT images reaches levels previously achieved on diagnostic CT images. The best object recognition strategy varies among objects; the proposed framework however allows employing a strategy that is optimal for each object.

61 RADIATION PROTECTION AND DOSIMETRY↗

DOE OSTI2016

Dosimetric comparison of stopping power calibration with dual-energy CT and single-energy CT in proton therapy treatment planning

Purpose: The accuracy of proton dose calculation is dependent on the ability to correctly characterize patient tissues with medical imaging. The most common method is to correlate computed tomography (CT) numbers obtained via single-energy CT (SECT) with proton stopping power ratio (SPR). CT numbers, however, cannot discriminate between a change in mass density and change in chemical composition of patient tissues. This limitation can have consequences on SPR calibration accuracy. Dual-energy CT (DECT) is receiving increasing interest as an alternative imaging modality for proton therapy treatment planning due to its ability to discriminate between changes in patient density and chemical composition. In the current work we use a phantom of known composition to demonstrate the dosimetric advantages of proton therapy treatment planning with DECT over SECT. Methods: A phantom of known composition was scanned with a clinical SECT radiotherapy CT-simulator. The phantom was rescanned at a lower X-ray tube potential to generate a complimentary DECT image set. A set of reference materials similar in composition to the phantom was used to perform a stoichiometric calibration of SECT CT number to proton SPRs. The same set of reference materials was used to perform a DECT stoichiometric calibration based on effective atomic number. The known composition of the phantom was used to assess the accuracy of SPR calibration with SECT and DECT. Intensity modulated proton therapy (IMPT) treatment plans were generated with the SECT and DECT image sets to assess the dosimetric effect of the imaging modality. Isodose difference maps and root mean square (RMS) error calculations were used to assess dose calculation accuracy. Results: SPR calculation accuracy was found to be superior, on average, with DECT relative to SECT. Maximum errors of 12.8% and 2.2% were found for SECT and DECT, respectively. Qualitative examination of dose difference maps clearly showed the dosimetric advantages of DECT imaging, compared to SECT imaging for IMPT dose calculation for the case investigated. Quantitatively, the maximum dose calculation error in the SECT plan was 7.8%, compared to a value of 1.4% in the DECT plan. When considering the high dose target region, the root mean square (RMS) error in dose calculation was 2.1% and 0.4% for SECT and DECT, respectively. Conclusions: DECT-based proton treatment planning in a commercial treatment planning system was successfully demonstrated for the first time. DECT is an attractive imaging modality for proton therapy treatment planning owing to its ability to characterize density and chemical composition of patient tissues. SECT and DECT scans of a phantom of known composition have been used to demonstrate the dosimetric advantages obtainable in proton therapy treatment planning with DECT over the current approach based on SECT.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2016

SU-F-I-53: Coded Aperture Coherent Scatter Spectral Imaging of the Breast: A Monte Carlo Evaluation of Absorbed Dose

Purpose: Coherent scatter based imaging has shown improved contrast and molecular specificity over conventional digital mammography however the biological risks have not been quantified due to a lack of accurate information on absorbed dose. This study intends to characterize the dose distribution and average glandular dose from coded aperture coherent scatter spectral imaging of the breast. The dose deposited in the breast from this new diagnostic imaging modality has not yet been quantitatively evaluated. Here, various digitized anthropomorphic phantoms are tested in a Monte Carlo simulation to evaluate the absorbed dose distribution and average glandular dose using clinically feasible scan protocols. Methods: Geant4 Monte Carlo radiation transport simulation software is used to replicate the coded aperture coherent scatter spectral imaging system. Energy sensitive, photon counting detectors are used to characterize the x-ray beam spectra for various imaging protocols. This input spectra is cross-validated with the results from XSPECT, a commercially available application that yields x-ray tube specific spectra for the operating parameters employed. XSPECT is also used to determine the appropriate number of photons emitted per mAs of tube current at a given kVp tube potential. With the implementation of the XCAT digital anthropomorphic breast phantom library, a variety of breast sizes with differing anatomical structure are evaluated. Simulations were performed with and without compression of the breast for dose comparison. Results: Through the Monte Carlo evaluation of a diverse population of breast types imaged under real-world scan conditions, a clinically relevant average glandular dose for this new imaging modality is extrapolated. Conclusion: With access to the physical coherent scatter imaging system used in the simulation, the results of this Monte Carlo study may be used to directly influence the future development of the modality to keep breast dose to a minimum while still maintaining clinically viable image quality.

60 APPLIED LIFE SCIENCES↗