The 50 most-cited articles in Medical Physics, 2020

Radiomics is an emerging area in quantitative image analysis that aims to relate large-scale extracted imaging information to clinical and biological endpoints. The development of quantitative imaging methods along with machine learning has enabled the opportunity to move data science research towards translation for more personalized cancer treatments. Accumulating evidence has indeed demonstrated that noninvasive advanced imaging analytics, that is, radiomics, can reveal key components of tumor phenotype for multiple three-dimensional lesions at multiple time points over and beyond the course of treatment. These developments in the use of CT, PET, US, and MR imaging could augment patient stratification and prognostication buttressing emerging targeted therapeutic approaches. In recent years, deep learning architectures have demonstrated their tremendous potential for image segmentation, reconstruction, recognition, and classification. Many powerful open-source and commercial platforms are currently available to embark in new research areas of radiomics. Quantitative imaging…

Computer-aided diagnosis (CAD) has been a major field of research for the past few decades. CAD uses machine learning methods to analyze imaging and/or nonimaging patient data and makes assessment of the patient's condition, which can then be used to assist clinicians in their decision-making process. The recent success of the deep learning technology in machine learning spurs new research and development efforts to improve CAD performance and to develop CAD for many other complex clinical tasks. In this paper, we discuss the potential and challenges in developing CAD tools using deep learning technology or artificial intelligence (AI) in general, the pitfalls and lessons learned from CAD in screening mammography and considerations needed for future implementation of CAD or AI in clinical use. It is hoped that the past experiences and the deep learning technology will lead to successful advancement and lasting growth in this new era of CAD, thereby enabling CAD to deliver intelligent aids to improve health care.

The use of radiochromic film (RCF) dosimetry in radiation therapy is extensive due to its high level of achievable accuracy for a wide range of dose values and its suitability under a variety of measurement conditions. However, since the publication of the 1998 AAPM Task Group 55, Report No. 63 on RCF dosimetry, the chemistry, composition, and readout systems for RCFs have evolved steadily. There are several challenges in using the new RCFs, readout systems and validation of the results depending on their applications. Accurate RCF dosimetry requires understanding of RCF selection, handling and calibration methods, calibration curves, dose conversion methods, correction methodologies as well as selection, operation and quality assurance (QA) programs of the readout systems. Acquiring this level of knowledge is not straight forward, even for some experienced users. This Task Group report addresses these issues and provides a basic understanding of available RCF models, dosimetric characteristics and properties, advantages and limitations, configurations, and overall elemental…

In recent years, significant progress has been made in developing more accurate and efficient machine learning algorithms for segmentation of medical and natural images. In this review article, we highlight the imperative role of machine learning algorithms in enabling efficient and accurate segmentation in the field of medical imaging. We specifically focus on several key studies pertaining to the application of machine learning methods to biomedical image segmentation. We review classical machine learning algorithms such as Markov random fields, k-means clustering, random forest, etc. Although such classical learning models are often less accurate compared to the deep-learning techniques, they are often more sample efficient and have a less complex structure. We also review different deep-learning architectures, such as the artificial neural networks (ANNs), the convolutional neural networks (CNNs), and the recurrent neural networks (RNNs), and present the segmentation results attained by those learning models that were published in the past 3 yr. We highlight the successes and…

PURPOSE: To describe a large, publicly available dataset comprising computed tomography (CT) projection data from patient exams, both at routine clinical doses and simulated lower doses. ACQUISITION AND VALIDATION METHODS: The library was developed under local ethics committee approval. Projection and image data from 299 clinically performed patient CT exams were archived for three types of clinical exams: noncontrast head CT scans acquired for acute cognitive or motor deficit, low-dose noncontrast chest scans acquired to screen high-risk patients for pulmonary nodules, and contrast-enhanced CT scans of the abdomen acquired to look for metastatic liver lesions. Scans were performed on CT systems from two different CT manufacturers using routine clinical protocols. Projection data were validated by reconstructing the data using several different reconstruction algorithms and through use of the data in the 2016 Low Dose CT Grand Challenge. Reduced dose projection data were simulated for each scan using a validated noise-insertion method. Radiologists marked location and diagnosis for…

PURPOSE: Accurate segmentation of lung and infection in COVID-19 computed tomography (CT) scans plays an important role in the quantitative management of patients. Most of the existing studies are based on large and private annotated datasets that are impractical to obtain from a single institution, especially when radiologists are busy fighting the coronavirus disease. Furthermore, it is hard to compare current COVID-19 CT segmentation methods as they are developed on different datasets, trained in different settings, and evaluated with different metrics. METHODS: To promote the development of data-efficient deep learning methods, in this paper, we built three benchmarks for lung and infection segmentation based on 70 annotated COVID-19 cases, which contain current active research areas, for example, few-shot learning, domain generalization, and knowledge transfer. For a fair comparison among different segmentation methods, we also provide standard training, validation and testing splits, evaluation metrics and, the corresponding code. RESULTS: Based on the state-of-the-art…

In x-ray computed tomography (CT), materials with different elemental compositions can have identical CT number values, depending on the mass density of each material and the energy of the detected x-ray beam. Differentiating and classifying different tissue types and contrast agents can thus be extremely challenging. In multienergy CT, one or more additional attenuation measurements are obtained at a second, third or more energy. This allows the differentiation of at least two materials. Commercial dual-energy CT systems (only two energy measurements) are now available either using sequential acquisitions of low- and high-tube potential scans, fast tube-potential switching, beam filtration combined with spiral scanning, dual-source, or dual-layer detector approaches. The use of energy-resolving, photon-counting detectors is now being evaluated on research systems. Irrespective of the technological approach to data acquisition, all commercial multienergy CT systems circa 2020 provide dual-energy data. Material decomposition algorithms are then used to identify specific materials…

OBJECTIVE: Computed tomography (CT) provides rich diagnosis and severity information of COVID-19 in clinical practice. However, there is no computerized tool to automatically delineate COVID-19 infection regions in chest CT scans for quantitative assessment in advanced applications such as severity prediction. The aim of this study was to develop a deep learning (DL)-based method for automatic segmentation and quantification of infection regions as well as the entire lungs from chest CT scans. METHODS: The DL-based segmentation method employs the "VB-Net" neural network to segment COVID-19 infection regions in CT scans. The developed DL-based segmentation system is trained by CT scans from 249 COVID-19 patients, and further validated by CT scans from other 300 COVID-19 patients. To accelerate the manual delineation of CT scans for training, a human-involved-model-iterations (HIMI) strategy is also adopted to assist radiologists to refine automatic annotation of each training case. To evaluate the performance of the DL-based segmentation system, three metrics, that is, Dice…

PURPOSE: Current clinical application of cone-beam CT (CBCT) is limited to patient setup. Imaging artifacts and Hounsfield unit (HU) inaccuracy make the process of CBCT-based adaptive planning presently impractical. In this study, we developed a deep-learning-based approach to improve CBCT image quality and HU accuracy for potential extended clinical use in CBCT-guided pancreatic adaptive radiotherapy. METHODS: Thirty patients previously treated with pancreas SBRT were included. The CBCT acquired prior to the first fraction of treatment was registered to the planning CT for training and generation of synthetic CT (sCT). A self-attention cycle generative adversarial network (cycleGAN) was used to generate CBCT-based sCT. For the cohort of 30 patients, the CT-based contours and treatment plans were transferred to the first fraction CBCTs and sCTs for dosimetric comparison. RESULTS: At the site of abdomen, mean absolute error (MAE) between CT and sCT was 56.89 ± 13.84 HU, comparing to 81.06 ± 15.86 HU between CT and the raw CBCT. No significant differences (P > 0.05) were observed in…

Purpose To characterize the noise and spatial resolution properties of a commercially available deep learning‐based computed tomography (CT) reconstruction algorithm. Methods Two phantom experiments were performed. The first used a multisized image quality phantom (Mercury v3.0, Duke University) imaged at five radiation dose levels (CTDIvol: 0.9, 1.2, 3.6, 7.0, and 22.3 mGy) with a fixed tube current technique on a commercial CT scanner (GE Revolution CT). Images were reconstructed with conventional (FBP), iterative (GE ASiR‐V), and deep learning‐based (GE True Fidelity) reconstruction algorithms. Noise power spectrum (NPS), high‐contrast (air–polyethylene interface), and intermediate‐contrast (water–polyethylene interface) task transfer functions (TTF) were measured for each dose level and phantom size and summarized in terms of average noise frequency (fav) and frequency at which the TTF was reduced to 50% (f50%), respectively. The second experiment used a custom phantom with low‐contrast rods and lung texture sections for the assessment of low‐contrast TTF and noise spatial…

PURPOSE: To develop a method of (a) calculating the dose rate of voxels within a proton field delivered using pencil beam scanning (PBS), and (b) reporting a representative dose rate for the PBS treatment field that enables correspondence between multiple treatment modalities. This method takes into account the unique spatiotemporal delivery patterns of PBS FLASH radiotherapy. METHODS: The dose rate at each voxel of a PBS radiation field is approximately the quotient of the voxel's dose and "effective" irradiation time. Each voxel's "effective" irradiation time starts when the cumulative dose rises above a chosen threshold value, and stops when its cumulative dose reaches its total dose minus the same threshold value. The above calculation yields a distribution of dose rates for the voxels within a PBS treatment field. To report a representative dose rate for the PBS field, we propose a user-selectable parameter of pth percentile of the dose rate distribution, such that (100 - p) % of the field is above the corresponding dose rate. To demonstrate the method described above, we…

BACKGROUND: Geant4 is a Monte Carlo code extensively used in medical physics for a wide range of applications, such as dosimetry, micro- and nanodosimetry, imaging, radiation protection, and nuclear medicine. Geant4 is continuously evolving, so it is crucial to have a system that benchmarks this Monte Carlo code for medical physics against reference data and to perform regression testing. AIMS: To respond to these needs, we developed G4-Med, a benchmarking and regression testing system of Geant4 for medical physics. MATERIALS AND METHODS: G4-Med currently includes 18 tests. They range from the benchmarking of fundamental physics quantities to the testing of Monte Carlo simulation setups typical of medical physics applications. Both electromagnetic and hadronic physics processes and models within the prebuilt Geant4 physics lists are tested. The tests included in G4-Med are executed on the CERN computing infrastructure via the use of the geant-val web application, developed at CERN for Geant4 testing. The physical observables can be compared to reference data for benchmarking and to…

PURPOSE: Gliomas are the most common primary tumor of the brain and are classified into grades I-IV of the World Health Organization (WHO), based on their invasively histological appearance. Gliomas grading plays an important role to determine the treatment plan and prognosis prediction. In this study we propose two novel methods for automatic, non-invasively distinguishing low-grade (Grades II and III) glioma (LGG) and high-grade (grade IV) glioma (HGG) on conventional MRI images by using deep convolutional neural networks (CNNs). METHODS: All MRI images have been preprocessed first by rigid image registration and intensity inhomogeneity correction. Both proposed methods consist of two steps: (a) three-dimensional (3D) brain tumor segmentation based on a modification of the popular U-Net model; (b) tumor classification on segmented brain tumor. In the first method, the slice with largest area of tumor is determined and the state-of-the-art mask R-CNN model is employed for tumor grading. To improve the performance of the grading model, a two-dimensional (2D) data augmentation has…

Radiotherapy (RT) is one of the basic treatment modalities for cancer of the head and neck (H&N), which requires a precise spatial description of the target volumes and organs at risk (OARs) to deliver a highly conformal radiation dose to the tumor cells while sparing the healthy tissues. For this purpose, target volumes and OARs have to be delineated and segmented from medical images. As manual delineation is a tedious and time-consuming task subjected to intra/interobserver variability, computerized auto-segmentation has been developed as an alternative. The field of medical imaging and RT planning has experienced an increased interest in the past decade, with new emerging trends that shifted the field of H&N OAR auto-segmentation from atlas-based to deep learning-based approaches. In this review, we systematically analyzed 78 relevant publications on auto-segmentation of OARs in the H&N region from 2008 to date, and provided critical discussions and recommendations from various perspectives: image modality - both computed tomography and magnetic resonance image modalities are…

BACKGROUND: While the review of radiotherapy treatment plans and charts by a medical physicist is a key component of safe, high-quality care, very few specific recommendations currently exist for this task. AIMS: The goal of TG-275 is to provide practical, evidence-based recommendations on physics plan and chart review for radiation therapy. While this report is aimed mainly at medical physicists, others may benefit including dosimetrists, radiation therapists, physicians and other professionals interested in quality management. METHODS: The scope of the report includes photon/electron external beam radiotherapy (EBRT), proton radiotherapy, as well as high-dose rate (HDR) brachytherapy for gynecological applications (currently the highest volume brachytherapy service in most practices). The following review time points are considered: initial review prior to treatment, weekly review, and end-of-treatment review. The Task Group takes a risk-informed approach to developing recommendations. A failure mode and effects analysis was performed to determine the highest-risk aspects of each…

PURPOSE: To improve image quality and computed tomography (CT) number accuracy of daily cone beam CT (CBCT) through a deep learning methodology with generative adversarial network. METHODS: One hundred and fifty paired pelvic CT and CBCT scans were used for model training and validation. An unsupervised deep learning method, 2.5D pixel-to-pixel generative adversarial network (GAN) model with feature mapping was proposed. A total of 12 000 slice pairs of CT and CBCT were used for model training, while ten-fold cross validation was applied to verify model robustness. Paired CT-CBCT scans from an additional 15 pelvic patients and 10 head-and-neck (HN) patients with CBCT images collected at a different machine were used for independent testing purpose. Besides the proposed method above, other network architectures were also tested as: 2D vs 2.5D; GAN model with vs without feature mapping; GAN model with vs without additional perceptual loss; and previously reported models as U-net and cycleGAN with or without identity loss. Image quality of deep-learning generated synthetic CT (sCT)…

Recent years have witnessed tremendous growth in the application of machine learning (ML) and deep learning (DL) techniques in medical physics. Embracing the current big data era, medical physicists equipped with these state-of-the-art tools should be able to solve pressing problems in modern radiation oncology. Here, a review of the basic aspects involved in ML/DL model building, including data processing, model training, and validation for medical physics applications is presented and discussed. Machine learning can be categorized based on the underlying task into supervised learning, unsupervised learning, or reinforcement learning; each of these categories has its own input/output dataset characteristics and aims to solve different classes of problems in medical physics ranging from automation of processes to predictive analytics. It is recognized that data size requirements may vary depending on the specific medical physics application and the nature of the algorithms applied. Data processing, which is a crucial step for model stability and precision, should be performed before…

PURPOSE: To develop an accurate and fast deformable image registration (DIR) method for four-dimensional computed tomography (4D-CT) lung images. Deep learning-based methods have the potential to quickly predict the deformation vector field (DVF) in a few forward predictions. We have developed an unsupervised deep learning method for 4D-CT lung DIR with excellent performances in terms of registration accuracies, robustness, and computational speed. METHODS: A fast and accurate 4D-CT lung DIR method, namely LungRegNet, was proposed using deep learning. LungRegNet consists of two subnetworks which are CoarseNet and FineNet. As the name suggests, CoarseNet predicts large lung motion on a coarse scale image while FineNet predicts local lung motion on a fine scale image. Both the CoarseNet and FineNet include a generator and a discriminator. The generator was trained to directly predict the DVF to deform the moving image. The discriminator was trained to distinguish the deformed images from the original images. CoarseNet was first trained to deform the moving images. The deformed images…

PURPOSE: Despite its increasing application, radiomics has not yet demonstrated a solid reliability, due to the difficulty in replicating analyses. The extraction of radiomic features from clinical MRI (T1w/T2w) presents even more challenges because of the absence of well-defined units (e.g. HU). Some preprocessing steps are required before the estimation of radiomic features and one of this is the intensity normalization, that can be performed using different methods. The aim of this work was to evaluate the effect of three different normalization techniques, applied on T2w-MRI images of the pelvic region, on radiomic features reproducibility. METHODS: T2w-MRI acquired before (MRI1) and 12 months after radiotherapy (MRI2) from 14 patients treated for prostate cancer were considered. Four different conditions were analyzed: (a) the original MRI (No_Norm); (b) MRI normalized by the mean image value (Norm_Mean); (c) MRI normalized by the mean value of the urine in the bladder (Norm_ROI); (d) MRI normalized by the histogram-matching method (Norm_HM). Ninety-one radiomic features were…

PURPOSE: The purpose of this study is to investigate the effect of different magnetic resonance (MR) sequences on the accuracy of deep learning-based synthetic computed tomography (sCT) generation in the complex head and neck region. METHODS: Four sequences of MR images (T1, T2, T1C, and T1DixonC-water) were collected from 45 patients with nasopharyngeal carcinoma. Seven conditional generative adversarial network (cGAN) models were trained with different sequences (single channel) and different combinations (multi-channel) as inputs. To further verify the cGAN performance, we also used a U-net network as a comparison. Mean absolute error, structural similarity index, peak signal-to-noise ratio, dice similarity coefficient, and dose distribution were evaluated between the actual CTs and sCTs generated from different models. RESULTS: The results show that the cGAN model with multi-channel (i.e., T1 + T2 + T1C + T1DixonC-water) as input to predict sCT achieves higher accuracy than any single MR sequence model. The T1-weighted MR model achieves better results than T2, T1C, and…

PURPOSE: Automatic breast ultrasound (ABUS) imaging has become an essential tool in breast cancer diagnosis since it provides complementary information to other imaging modalities. Lesion segmentation on ABUS is a prerequisite step of breast cancer computer-aided diagnosis (CAD). This work aims to develop a deep learning-based method for breast tumor segmentation using three-dimensional (3D) ABUS automatically. METHODS: For breast tumor segmentation in ABUS, we developed a Mask scoring region-based convolutional neural network (R-CNN) that consists of five subnetworks, that is, a backbone, a regional proposal network, a region convolutional neural network head, a mask head, and a mask score head. A network block building direct correlation between mask quality and region class was integrated into a Mask scoring R-CNN based framework for the segmentation of new ABUS images with ambiguous regions of interest (ROIs). For segmentation accuracy evaluation, we retrospectively investigated 70 patients with breast tumor confirmed with needle biopsy and manually delineated on ABUS, of which…

BACKGROUND: Recent advances in machine and deep learning based on an increased availability of clinical data have fueled renewed interest in computerized clinical decision support systems (CDSSs). CDSSs have shown great potential to improve healthcare, increase patient safety and reduce costs. However, the use of CDSSs is not without pitfalls, as an inadequate or faulty CDSS can potentially deteriorate the quality of healthcare and put patients at risk. In addition, the adoption of a CDSS might fail because its intended users ignore the output of the CDSS due to lack of trust, relevancy or actionability. AIM: In this article, we provide guidance based on literature for the different aspects involved in the adoption of a CDSS with a special focus on machine and deep learning based systems: selection, acceptance testing, commissioning, implementation and quality assurance. RESULTS: A rigorous selection process will help identify the CDSS that best fits the preferences and requirements of the local site. Acceptance testing will make sure that the selected CDSS fulfills the defined…

PURPOSE: Despite the proven utility of multiparametric magnetic resonance imaging (MRI) in radiation therapy, MRI-guided radiation treatment planning is limited by the fact that MRI does not directly provide the electron density map required for absorbed dose calculation. In this work, a new deep convolutional neural network model with efficient learning capability, suitable for applications where the number of training subjects is limited, is proposed to generate accurate synthetic computed tomography (sCT) images from MRI. METHODS: This efficient convolutional neural network (eCNN) is built upon a combination of the SegNet architecture (a 13-layer encoder-decoder structure similar to the U-Net network) without softmax layers and the residual network. Moreover, maxpooling indices and high resolution features from the encoding network were incorporated into the corresponding decoding layers. A dataset containing 15 co-registered MRI-CT pairs of male pelvis (1861 two-dimensional images) were used for training and evaluation of MRI to CT synthesis process using a fivefold…

PURPOSE: The finite element method (FEM) is the preferred method to simulate phenomena in anatomical structures. However, purely FEM-based mechanical simulations require considerable time, limiting their use in clinical applications that require real-time responses, such as haptics simulators. Machine learning (ML) approaches have been proposed to help with the reduction of the required time. The present paper reviews cases where ML could help to generate faster simulations, without considerably affecting the performance results. METHODS: This review details the ML approaches used, considering the anatomical structures involved, the data collection strategies, the selected ML algorithms, with corresponding features, the metrics used for validation, and the resulting time gains. RESULTS: A total of 41 references were found. ML algorithms are mainly trained with FEM-based simulations in 32 publications. The preferred ML approach is neural networks, including deep learning in 35 publications. Tissue deformation is simulated in 18 applications, but other features are also considered.…

PURPOSE: It has been recently shown that radiotherapy at ultrahigh dose rates (>40 Gy/s, FLASH) has a potential advantage in sparing healthy organs compared to that at conventional dose rates. The purpose of this work is to show the feasibility of proton FLASH irradiation using a gantry-mounted synchrocyclotron as a first step toward implementing an experimental setup for preclinical studies. METHODS: synchrocyclotron was modified to deliver ultrahigh dose rates. Pulse widths of protons with 230 MeV energy were manipulated from 1 to 20 μs to deliver in conventional and ultrahigh dose rate. A boron carbide absorber was placed in the beam for range modulation. A Faraday cup was used to determine the number of protons per pulse at various dose rates. Dose rate was determined by the dose measured with a plane-parallel ionization chamber with respect to the actual delivery time. The integral depth dose (IDD) was measured with a Bragg ionization chamber. Monte Carlo simulation was performed in TOPAS as the secondary check for the measurements. RESULTS: Maximum protons charge per pulse,…

PURPOSE: FLASH radiotherapy (RT) can potentially reduce normal tissue toxicity while preserving tumoricidal effectiveness to improve the therapeutic ratio. The key of FLASH for sparing normal tissues is to irradiate tissues with an ultra-high dose rate (i.e., ≥40 Gy/s), for which proton RT can be used. However, currently available treatment plan optimization method only optimizes the dose distribution and does not directly optimize the dose rate. The contribution of this work to FLASH proton RT is the development of a novel treatment optimization method, that is, simultaneous dose and dose rate optimization (SDDRO), to optimize tissue-receiving dose rate distribution as well as dose distribution. METHODS: Distinguished from existing methods, SDDRO accounts for dose rate constraint and optimizes dose rate distribution. In terms of mathematical formulation, SDDRO is a constrained optimization problem with dose-volume constraint on dose distribution, minimum dose rate constraint on dose-averaged tissue-receiving dose rates, minimum monitor unit constraint on spot weight, and maximum…

PURPOSE: In the treatment planning process of intensity-modulated radiation therapy (IMRT), a human planner operates the treatment planning system (TPS) to adjust treatment planning parameters, for example, dose volume histogram (DVH) constraints' locations and weights, to achieve a satisfactory plan for each patient. This process is usually time-consuming, and the plan quality depends on planer's experience and available planning time. In this study, we proposed to model the behaviors of human planners in treatment planning by a deep reinforcement learning (DRL)-based virtual treatment planner network (VTPN), such that it can operate the TPS in a human-like manner for treatment planning. METHODS AND MATERIALS: Using prostate cancer IMRT as an example, we established the VTPN using a deep neural network developed. We considered an in-house optimization engine with a weighted quadratic objective function. Virtual treatment planner network was designed to observe an intermediate plan DVHs and decide the action to improve the plan by changing weights and threshold dose in the objective…

The era of real-time radiotherapy is upon us. Robotic and gimbaled linac tracking are clinically established technologies with the clinical realization of couch tracking in development. Multileaf collimators (MLCs) are a standard equipment for most cancer radiotherapy systems, and therefore MLC tracking is a potentially widely available technology. MLC tracking has been the subject of theoretical and experimental research for decades and was first implemented for patient treatments in 2013. The AAPM Task Group 264 Safe Clinical Implementation of MLC Tracking in Radiotherapy Report was charged to proactively provide the broader radiation oncology community with (a) clinical implementation guidelines including hardware, software, and clinical indications for use, (b) commissioning and quality assurance recommendations based on early user experience, as well as guidelines on Failure Mode and Effects Analysis, and (c) a discussion of potential future developments. The deliverables from this report include: an explanation of MLC tracking and its historical development; terms and…

PURPOSE: Complementary information obtained from multiple contrasts of tissue facilitates physicians assessing, diagnosing and planning treatment of a variety of diseases. However, acquiring multiple contrasts magnetic resonance images (MRI) for every patient using multiple pulse sequences is time-consuming and expensive, where, medical image synthesis has been demonstrated as an effective alternative. The purpose of this study is to develop a unified framework for multimodal MR image synthesis. METHODS: A unified generative adversarial network consisting of only a single generator and a single discriminator was developed to learn the mappings among images of four different modalities. The generator took an image and its modality label as inputs and learned to synthesize the image in the target modality, while the discriminator was trained to distinguish between real and synthesized images and classify them to their corresponding modalities. The network was trained and tested using multimodal brain MRI consisting of four different contrasts which are T1-weighted (T1), T1-weighted…

PURPOSE: Segmentation of multiple organs-at-risk (OARs) is essential for magnetic resonance (MR)-only radiation therapy treatment planning and MR-guided adaptive radiotherapy of abdominal cancers. Current practice requires manual delineation that is labor-intensive, time-consuming, and prone to intra- and interobserver variations. We developed a deep learning (DL) technique for fully automated segmentation of multiple OARs on clinical abdominal MR images with high accuracy, reliability, and efficiency. METHODS: We developed Automated deep Learning-based abdominal multiorgan segmentation (ALAMO) technique based on two-dimensional U-net and a densely connected network structure with tailored design in data augmentation and training procedures such as deep connection, auxiliary supervision, and multiview. The model takes in multislice MR images and generates the output of segmentation results. 3.0-Tesla T1 VIBE (Volumetric Interpolated Breath-hold Examination) images of 102 subjects were used in our study and split into 66 for training, 16 for validation, and 20 for testing. Ten OARs…

PURPOSE: Needle-based procedures for diagnosing and treating prostate cancer, such as biopsy and brachytherapy, have incorporated three-dimensional (3D) transrectal ultrasound (TRUS) imaging to improve needle guidance. Using these images effectively typically requires the physician to manually segment the prostate to define the margins used for accurate registration, targeting, and other guidance techniques. However, manual prostate segmentation is a time-consuming and difficult intraoperative process, often occurring while the patient is under sedation (biopsy) or anesthetic (brachytherapy). Minimizing procedure time with a 3D TRUS prostate segmentation method could provide physicians with a quick and accurate prostate segmentation, and allow for an efficient workflow with improved patient throughput to enable faster patient access to care. The purpose of this study was to develop a supervised deep learning-based method to segment the prostate in 3D TRUS images from different facilities, generated using multiple acquisition methods and commercial ultrasound machine models to create…

Purpose We present a first‐principles molecular dynamics (MD) simulation and expound upon a mechanism of oxygen depletion hypothesis to explain the mitigation of normal tissue injury observed in ultra‐high‐dose‐rate (UHDR) FLASH radiotherapy. Methods We simulated damage to a segment of DNA (also representing other biomolecules such as RNA and proteins) in a simulation box filled with O and molecules. Attoseconds physical interactions (ionizations, electronic, and vibrational excitations) were simulated by using the Monte Carlo track structure code Geant4‐DNA. Immediately after ionization, ab initio Car–Parrinello molecular dynamics (CPMD) simulation was used to identify which O and molecules surrounding the DNA molecule were converted into reactive oxygen species (ROS). Subsequently, the femto‐ to nanosecond reactions of ROS were simulated by using MD with reactive force field (ReaxFF), to illustrate ROS merging into new types of non‐reactive oxygen species (NROS) due to strong coupling among ROS. A coarse‐grained model was constructed to describe the relevant collective phenomenon…

PURPOSE: To develop and evaluate a novel method for pseudo-CT generation from multi-parametric MR images using multi-channel multi-path generative adversarial network (MCMP-GAN). METHODS: Pre- and post-contrast T1-weighted (T1-w), T2-weighted (T2-w) MRI, and treatment planning CT images of 32 nasopharyngeal carcinoma (NPC) patients were employed to train a pixel-to-pixel MCMP-GAN. The network was developed based on a 5-level Residual U-Net (ResU-Net) with the channel-based independent feature extraction network to generate pseudo-CT images from multi-parametric MR images. The discriminator with five convolutional layers was added to distinguish between the real CT and pseudo-CT images, improving the nonlinearity and prediction accuracy of the model. Eightfold cross validation was implemented to validate the proposed MCMP-GAN. The pseudo-CT images were evaluated against the corresponding planning CT images based on mean absolute error (MAE), peak signal-to-noise ratio (PSNR), Dice similarity coefficient (DSC), and Structural similarity index (SSIM). Similar comparisons were also…

PURPOSE: One technical barrier to patient-specific computed tomography (CT) dosimetry has been the lack of computational tools for the automatic patient-specific multi-organ segmentation of CT images and rapid organ dose quantification. When previous CT images are available for the same body region of the patient, the ability to obtain patient-specific organ doses for CT - in a similar manner as radiation therapy treatment planning - will open the door to personalized and prospective CT scan protocols. This study aims to demonstrate the feasibility of combining deep-learning algorithms for automatic segmentation of multiple radiosensitive organs from CT images with the GPU-based Monte Carlo rapid organ dose calculation. METHODS: A deep convolutional neural network (CNN) based on the U-Net for organ segmentation is developed and trained to automatically delineate multiple radiosensitive organs from CT images. Two databases are used: The lung CT segmentation challenge 2017 (LCTSC) dataset that contains 60 thoracic CT scan patients, each consisting of five segmented organs, and the…

PURPOSE: To develop a tool for the automatic contouring of clinical treatment volumes (CTVs) and normal tissues for radiotherapy treatment planning in cervical cancer patients. METHODS: An auto-contouring tool based on convolutional neural networks (CNN) was developed to delineate three cervical CTVs and 11 normal structures (seven OARs, four bony structures) in cervical cancer treatment for use with the Radiation Planning Assistant, a web-based automatic plan generation system. A total of 2254 retrospective clinical computed tomography (CT) scans from a single cancer center and 210 CT scans from a segmentation challenge were used to train and validate the CNN-based auto-contouring tool. The accuracy of the tool was evaluated by calculating the Sørensen-dice similarity coefficient (DSC) and mean surface and Hausdorff distances between the automatically generated contours and physician-drawn contours on 140 internal CT scans. A radiation oncologist scored the automatically generated contours on 30 external CT scans from three South African hospitals. RESULTS: The average DSC, mean…

Purpose Positron emission tomography (PET) is an essential technique in many clinical applications that allows for quantitative imaging at the molecular level. This study aims to develop a denoising method using a novel dilated convolutional neural network (CNN) to recover full‐count images from low‐count images. Methods We adopted similar hierarchical structures as the conventional U‐Net and incorporated dilated kernels in each convolution to allow the network to observe larger, more robust features within the image without the requirement of downsampling and upsampling internal representations. Our dNet was trained alongside a U‐Net for comparison. Both models were evaluated using a leave‐one‐out cross‐validation procedure on a dataset of 35 subjects (~3500 slabs), which were obtained from an ongoing 18 F‐Fluorodeoxyglucose (FDG) study. Low‐count PET data (10% count) were generated by randomly selecting one‐tenth of all events in the associated listmode file. Analysis was done on the static image from the last 10 minutes of emission data. Both low‐count PET and full‐count PET were…

PURPOSE: The ion collection efficiency of vented ionization chambers has been investigated in an ultra-high dose-per-pulse (DPP) electron beam. The role of the chamber design and the electric field strength in the sensitive air volume have been evaluated. METHODS: An advanced Markus chamber and three specially designed parallel plate air-filled ionization chambers (EWC: End Window Chamber) with varying electrode distance of 0.5, 1, and 2 mm have been investigated. Their ion collection efficiencies were determined experimentally using two methods: extrapolation of Jaffé plots and comparison against a DPP-independent reference detector. The latter was achieved by calibrating a current transformer against alanine dosimeters. All measurements were performed in a 24 MeV electron beam with DPP values between 0.01 and 3 Gy. Additionally, the numerical approach introduced by Gotz et al. was implemented taking into account space charge effects at these ultra-high DPPs. The method has been extended to obtain time-resolved and position-dependent electric field distortions within the air…

INTRODUCTION: This work describes the development of a novel radiomics phantom designed for magnetic resonance imaging (MRI) that can be used in a multicenter setting. The purpose of this study is to assess the stability and reproducibility of MRI-based radiomic features using this phantom across different MRI scanners. METHODS & MATERIALS: A set of phantoms were three-dimensional (3D) printed using MRI visible materials. One set of phantoms were imaged on seven MRI scanners and one was imaged on one MRI scanner. Radiomics analysis of the phantoms, which included first-order features, shape and texture features was performed. Intraclass correlation coefficient (ICC) was used to assess the stability of radiomic features across eight scanners and the reproducibility of two printed models on one scanner. Coefficient of variation (COV) was used to assess the reproducibility of radiomics measurements in the phantom on a single scanner. RESULTS: The phantom models provide sufficient signal-to-noise and contrast in all the tumor models permitting robust automatic segmentation. During a…

PURPOSE: To assess the repeatability of radiomic features in magnetic resonance (MR) imaging of glioblastoma (GBM) tumors with respect to test-retest, different image registration approaches and inhomogeneity bias field correction. METHODS: We analyzed MR images of 17 GBM patients including T1- and T2-weighted images (performed within the same imaging unit on two consecutive days). For image segmentation, we used a comprehensive segmentation approach including entire tumor, active area of tumor, necrotic regions in T1-weighted images, and edema regions in T2-weighted images (test studies only; registration to retest studies is discussed next). Analysis included N3, N4 as well as no bias correction performed on raw MR images. We evaluated 20 image registration approaches, generated by cross-combination of four transformation and five cost function methods. In total, 714 images (17 patients × 2 images × ((4 transformations × 5 cost functions) + 1 test image) and 2856 segmentations (714 images × 4 segmentations) were prepared for feature extraction. Various radiomic features were…

BACKGROUND: In oncology, lesion characterization is essential for tumor grading, treatment planning, and follow-up of cancer patients. Hybrid imaging systems, such as Single Photon Emission Computed Tomography (SPECT)/CT, Positron Emission Tomography (PET)/CT, or PET/magnetic resonance imaging (MRI), play an essential role for the noninvasive quantification of tumor characteristics. However, most of the existing approaches are challenged by intra- and intertumor heterogeneity. Novel quantitative imaging parameters that can be derived from textural feature analysis (as part of radiomics) are promising complements for improved characterization of tumor heterogeneity, thus, supporting clinically relevant implementations of personalized medicine concepts. Nevertheless, establishing new quantitative parameters for tumor characterization requires the use of standardized imaging objects to test the reliability of results prior to their implementation in patient studies. METHODS: In this review, we summarize existing reports on heterogeneous phantoms with a focus on simulating tumor…

PURPOSE: Pancreas segmentation is a difficult task because of the high intrapatient variability in the shape, size, and location of the organ, as well as the low contrast and small footprint of the CT scan. At present, the U-Net model is likely to lead to the problems of intraclass inconsistency and interclass indistinction in pancreas segmentation. To solve this problem, we improved the contextual and semantic feature information acquisition method of the biomedical image segmentation model (U-Net) based on a convolutional network and proposed an improved segmentation model called the multiscale attention dense residual U-shaped network (MAD-UNet). METHODS: There are two aspects considered in this method. First, we adopted dense residual blocks and weighted binary cross-entropy to enhance the semantic features to learn the details of the pancreas. Using such an approach can reduce the effects of intraclass inconsistency. Second, we used an attention mechanism and multiscale convolution to enrich the contextual information and suppress learning in unrelated areas. We let the model…

Purpose Dual‐energy (DE) x‐ray imaging has many clinical applications in radiography, fluoroscopy, and CT. This work characterizes a prototype dual‐layer (DL) flat‐panel detector (FPD) and investigates its DE imaging capabilities for applications in two‐dimensional (2D) radiography/fluoroscopy and quantitative three‐dimensional (3D) cone‐beam CT. Unlike other DE methods like kV switching, a DL FPD obtains DE images from a single exposure, making it robust against patient and system motion. Methods The DL FPD consists of a top layer with a 200 µm‐thick CsI scintillator coupled to an amorphous silicon (aSi) FPD of 150 µm pixel size and a bottom layer with a 550 µm thick CsI scintillator coupled to an identical aSi FPD. The two layers are separated by a 1‐mm Cu filter to increase spectral separation. Images (43 × 43 cm2 active area) can be readout in 2 × 2 binning mode (300 µm pixels) at up to 15 frames per second. Detector performance was first characterized by measuring the MTF, NPS, and DQE for the top and bottom layers. For 2D applications, a qualitative study was conducted using…

PURPOSE: To perform an in-depth evaluation of current state of the art techniques in training neural networks to identify appropriate approaches in small datasets. METHOD: In total, 112,120 frontal-view X-ray images from the NIH ChestXray14 dataset were used in our analysis. Two tasks were studied: unbalanced multi-label classification of 14 diseases, and binary classification of pneumonia vs non-pneumonia. All datasets were randomly split into training, validation, and testing (70%, 10%, and 20%). Two popular convolution neural networks (CNNs), DensNet121 and ResNet50, were trained using PyTorch. We performed several experiments to test: (a) whether transfer learning using pretrained networks on ImageNet are of value to medical imaging/physics tasks (e.g., predicting toxicity from radiographic images after training on images from the internet), (b) whether using pretrained networks trained on problems that are similar to the target task helps transfer learning (e.g., using X-ray pretrained networks for X-ray target tasks), (c) whether freeze deep layers or change all weights…

PURPOSE: Real-time high soft-tissue contrast magnetic resonance imaging (MRI) from the MR-Linac offers the best opportunity for accurate motion tracking during radiation therapy delivery via high-frequency two-dimensional (2D) cine imaging. This work investigates the efficacy of real-time organ motion tracking based on the registration of MRI acquired on MR-Linac. METHODS: Algorithms based on image intensity were developed to determine the three-dimensional (3D) translation of abdominal targets. 2D and 3D abdominal MRIs were acquired for 10 healthy volunteers using a high-field MR-Linac. For each volunteer, 3D respiration-gated T2 and 2D T2/T1-weighted cine in sagittal, coronal, and axial planes with a planar temporal resolution of 0.6 for 60 s was captured. Datasets were also collected on MR-compatible physical and virtual four-dimensional (4D) motion phantoms. Target contours for the liver and pancreas from the 3D T2 were populated to the cine and assumed as the ground-truth motion. We performed image registration using a research software to track the target 3D motion. Standard…

PURPOSE: Sparse-data computed tomography (CT) frequently occurs, such as breast tomosynthesis, C-arm CT, on-board four-dimensional cone-beam CT (4D CBCT), and industrial CT. However, sparse-data image reconstruction remains challenging due to highly undersampled data. This work develops a data-driven image reconstruction method for sparse-data CT using deep neural networks (DNN). METHODS: The new method so-called AirNet is designed to incorporate the benefits from analytical reconstruction method (AR), iterative reconstruction method (IR), and DNN. It is built upon fused analytical and iterative reconstruction (AIR) that synergizes AR and IR via the optimization framework of modified proximal forward-backward splitting (PFBS). By unrolling PFBS into IR updates of CT data fidelity and DNN regularization with residual learning, AirNet utilizes AR such as FBP during the data fidelity, introduces dense connectivity into DNN regularization, and learns PFBS coefficients and DNN parameters that minimize the loss function during the training stage; and then AirNet with trained parameters…

Purpose Anatomical variations occur during head and neck (H&N) radiotherapy treatment. kV cone‐beam computed tomography (CBCT) images can be used for daily dose monitoring to assess dose variations owing to anatomic changes. Deep learning methods (DLMs) have recently been proposed to generate pseudo‐CT (pCT) from CBCT to perform dose calculation. This study aims to evaluate the accuracy of a DLM and to compare this method with three existing methods of dose calculation from CBCT in H&N cancer radiotherapy. Methods Forty‐four patients received VMAT for H&N cancer (70‐63‐56 Gy). For each patient, reference CT (Bigbore, Philips) and CBCT images (XVI, Elekta) were acquired. The DLM was based on a generative adversarial network. The three compared methods were: (a) a method using a density to Hounsfield Unit (HU) relation from phantom CBCT image (HU‐D curve method), (b) a water–air‐bone density assignment method (DAM), and iii) a method using deformable image registration (DIR). The imaging endpoints were the mean absolute error (MAE) and mean error (ME) of HU from pCT and reference CT…

PURPOSE: , a recently developed patient-specific quality assurance (PSQA) tool for end-to-end testing of stereotactic radiosurgery (SRS) and stereotactic body radiation therapy (SBRT), was evaluated in a multi-institution study and compared with radiochromic film. METHODS: The SRS MapCHECK was used to collect data on 84 SBRT or SRS PSQA plans/fields at nine institutions on treatment delivery devices (TDD) manufactured by Varian and Elekta. PSQA plans from five different treatment planning software (TPS) were selected and executed on TDDs operating at beam energies of 6 and 10 MV with and without a flattening filter. The patient plans were all VMAT except for ten conformal arc therapy fields. The plans were selected to encompass a range of size and tumor sites including brain, lung, spine, abdomen, ear, pancreas, and liver. Corresponding radiochromic film data was acquired in 50 plans/fields. Results were evaluated using gamma analysis with absolute dose criterion of 3% global dose-difference (DD) and 1 mm distance-to-agreement (DTA). RESULTS: The mean 3% DD/1 mm DTA Gamma pass rate…

Proton therapy is an expanding radiotherapy modality in the United States and worldwide. With the number of proton therapy centers treating patients increasing, so does the need for consistent, high-quality clinical commissioning practices. Clinical commissioning encompasses the entire proton therapy system's multiple components, including the treatment delivery system, the patient positioning system, and the image-guided radiotherapy components. Also included in the commissioning process are the x-ray computed tomography scanner calibration for proton stopping power, the radiotherapy treatment planning system, and corresponding portions of the treatment management system. This commissioning report focuses exclusively on intensity-modulated scanning systems, presenting details of how to perform the commissioning of the proton therapy and ancillary systems, including the required proton beam measurements, treatment planning system dose modeling, and the equipment needed.

PURPOSE: To apply an imaging metric of the structural SIMilarity (SSIM) index to the radiotherapy dose verification field and evaluate its capability to reveal the different types of errors between two dose distributions. METHOD: The SSIM index consists of three sub-indices: luminance, contrast, and structure. Given two images, luminance analysis compares the local mean result, contrast analysis compares the local standard deviation, and the structure index represents the local Pearson correlation. Three test error patterns (absolute dose error, dose gradient error, and dose structure error) were designed to characterize the response of SSIM and its sub-indices and establish the correlation between the indices and different dose error types. After establishing the correlation, four radiotherapy plans (one MLC picket-fence test plan, one brain stereotactic radiotherapy plan, and two head-and-neck plans) were tested by computing each index and compared with the gamma analysis results to determine their similarities and differences. RESULTS: Among the three test error patterns, the…

PURPOSE: The accurate segmentation of liver and liver tumors from CT images can assist radiologists in decision-making and treatment planning. The contours of liver and liver tumors are currently obtained by manual labeling, which is time-consuming and subjective. Computer-aided segmentation methods have been widely used in the segmentation of liver and liver tumors. However, due to the diversity of shape, volume, and image intensity, the segmentation is still a difficult task. In this study, we present a Spatial Feature Fusion Convolutional Network (SFF-Net) to automatically segment liver and liver tumors from CT images. METHODS: First, we extract side-outputs at each convolutional block in SFF-Net to make full use of multiscale features. Second, skip-connections are added in the down-sampling phase, therefore, the spatial information can be efficiently transferred to later layers. Third, we present feature fusion blocks (FFBs) to merge spatial features and high-level semantic features from early layers and later layers, respectively. Finally, a fully connected 3D conditional…