Standard Clinical Imaging Protocols Research Articles

RADCURE: An open-source head and neck cancer CT dataset for clinical radiation therapy insights.

This manuscript presents RADCURE, one of the most extensive head and neck cancer (HNC) imaging datasets accessible to the public. Initially collected for clinical radiation therapy (RT) treatment planning, this dataset has been retrospectively reconstructed for use in imaging research. RADCURE encompasses data from 3346 patients, featuring computed tomography (CT) RT simulation images with corresponding target and organ-at-risk contours. These CT scans were collected using systems from three different manufacturers. Standard clinical imaging protocols were followed, and contours were manually generated and reviewed at weekly RT quality assurance rounds. RADCURE imaging and structure set data was extracted from our institution's radiation treatment planning and oncology information systems using a custom-built data mining and processing system. Furthermore, images were linked to our clinical anthology of outcomes data for each patient and includes demographic, clinical and treatment information based on the 7th edition TNM staging system (Tumor-Node-Metastasis Classification System of Malignant Tumors). The median patient age is 63, with the final dataset including 80% males. Half of the cohort is diagnosed with oropharyngeal cancer, while laryngeal, nasopharyngeal, and hypopharyngeal cancers account for 25%, 12%, and 5% of cases, respectively. The median duration of follow-up is five years, with 60% of the cohort surviving until the last follow-up point. The dataset provides images and contours in DICOM CT and RT-STRUCT formats, respectively. We have standardized the nomenclature for individual contours-such as the gross primary tumor, gross nodal volumes, and 19 organs-at-risk-to enhance the RT-STRUCT files' utility. Accompanying demographic, clinical, and treatment data are supplied in a comma-separated values (CSV) file format. This comprehensive dataset is publicly accessible via The Cancer Imaging Archive. RADCURE's amalgamation of imaging, clinical, demographic, and treatment data renders it an invaluable resource for a broad spectrum of radiomics image analysis research endeavors. Researchers can utilize this dataset to advance routine clinical procedures using machine learning or artificial intelligence, to identify new non-invasive biomarkers, or to forge prognostic models.

Carotid disease, cognition, and aging: time to redefine asymptomatic disease?

There is an increasing appreciation of the vascular contributions in the development of age-related cognitive impairment and dementia1,2. Identifying risk and maintaining cognitive health for successful aging is ever relevant in our aging population. Carotid disease, a well-established risk factor for stroke and often a harbinger of other vascular disease states, is also emerging as another vascular risk factor for age-related cognitive decline. When combined with vascular risk factors, the incidence of age-related carotid disease can be as high as 70%3,4. Historically, carotid disease has been dichotomized into two large groups in trial design, outcome measurements, and treatment decisions: symptomatic and asymptomatic carotid artery stenosis. The dichotomous distinction between asymptomatic and symptomatic carotid stenosis based on existing definitions may be limiting the care we are able to provide for patients classified as "asymptomatic" from their carotid disease. Medically, we now know that these patients should be treated with the same intensive medical therapy as those with "symptomatic" carotid disease. Emerging data also shows that hypoperfusion from asymptomatic disease may lead to significant cognitive impairment in the aging population, and it is plausible that most "age-related" cognitive changes may be reflective of vascular impairment and neurovascular dysfunction. While over the past 30years medical, surgical, and radiological advances have pushed the field of neurovascular disease to significantly reduce the number of ischemic strokes, we are far from any meaningful interventions to prevent vascular cognitive impairment. In addition to including cognitive outcome measures, future studies of carotid disease will also benefit from including advanced neuroimaging modalities not currently utilized in standard clinical imaging protocols, such as perfusion imaging and/or functional connectivity mapping, which may provide novel data to better assess for hypoxic-ischemic changes and neurovascular dysfunction across diffuse cognitive networks. While current recommendations advise against widespread population screening for asymptomatic carotid stenosis, emerging evidence linking carotid stenosis to cognitive impairment prompts us to re-consider our approach for older patients with vascular risk factors who are at risk for cognitive decline.

T1 Mapping From MPRAGE Acquisitions: Application to the Measurement of the Concentration of Nanoparticles in Tumors for Theranostic Use.

The measurement of the concentration of theranostic agents in vivo is essential for the assessment of their therapeutic efficacy and their safety regarding healthy tissue. To this end, there is a need for quantitative T1 measurements that can be obtained as part of a standard clinical imaging protocol applied to tumor patients. To generate T1 maps from MR images obtained with the magnetization-prepared rapid gradient echo (MPRAGE) sequence. To evaluate the feasibility of the proposed approach on phantoms, animal and patients with brain metastases. Pilot. Solutions containing contrast agents (chelated Gd3+ and iron nanoparticles), male rat of Wistar strain, three patients with brain metastases. A 3-T and 7-T, saturation recovery (SR), and MPRAGE sequences. The MPRAGE T1 measurement was compared to the reference SR method on phantoms and rat brain at 7-T. The robustness of the in vivo method was evaluated by studying the impact of misestimates of tissue proton density. Concentrations of Gd-based theranostic agents were measured at 3-T in gray matter and metastases in patients recruited in NanoRad clinical trial. A linear model was used to characterize the relation between T1 measurements from the MPRAGE and the SR acquisitions obtained in vitro at 7-T. The slope of the linear model was 0.966 (R2 =0.9934). MPRAGE-based T1 values measured in the rat brain were 1723 msec in the thalamus. MPRAGE-based T1 values measured in patients in white matter and gray matter amounted to 747 msec and 1690 msec. Mean concentration values of Gd3+ in metastases were 61.47 μmol. The T1 values obtained in vitro and in vivo support the validity of the proposed approach. The concentrations of Gd-based theranostic agents may be assessed in patients with metastases within a standard clinical imaging protocol using the MPRAGE sequence. 2. Stage 1.

Skull phantom-based methodology to validate MRI co-registration accuracy for Gamma Knife radiosurgery.

Target localization, for stereotactic radiosurgery (SRS) treatment with Gamma Knife, has become increasingly reliant on the co-registration between the planning MRI and the stereotactic cone-beam computed tomography (CBCT). Validating image registration between modalities would be particularly beneficial when considering the emergence of novel functional and metabolic MRI pulse sequences for target delineation. This study aimed to develop a phantom-based methodology to quantitatively compare the co-registration accuracy of the standard clinical imaging protocol to a representative MRI sequence that was likely to fail co-registration. The comparative methodology presented in this study may serve as a useful tool to evaluate the clinical translatability of novel MRI sequences. A realistic human skull phantom with fiducial marker columns was designed and manufactured to fit into a typical MRI head coil and the Gamma Knife patient positioning system. A series of "optimized" 3D MRI sequences-T1 -weighted Dixon, T1 -weighted fast field echo (FFE), and T2 -weighted fluid-attenuated inversion recovery (FLAIR)-were acquired and co-registered to the CBCT. The same sequences were "compromised" by reconstructing without geometric distortion correction and re-collecting with lower signal-to-noise-ratio (SNR) to simulate a novel MRI sequence with poor co-registration accuracy. Image similarity metrics-structural similarity (SSIM) index, mean squared error (MSE), and peak SNR (PSNR)-were used to quantitatively compare the co-registration of the optimized and compromised MR images. The ground truth fiducial positions were compared to positions measured from each optimized image volume revealing a maximum median geometric uncertainty of 0.39mm (LR), 0.92mm (AP), and 0.13mm (SI) between the CT and CBCT, 0.60mm (LR), 0.36mm (AP), and 0.07mm (SI) between the CT and T1 -weighted Dixon, 0.42mm (LR), 0.23mm (AP), and 0.08mm (SI) between the CT and T1 -weighted FFE, and 0.45mm (LR), 0.19mm (AP), and 1.04mm (SI) between the CT and T2 -weighted FLAIR. Qualitatively, pairs of optimized and compromised image slices were compared using a fusion image where separable colors were used to differentiate between images. Quantitatively, MSE was the most predictive and SSIM the second most predictive metric for evaluating co-registration similarity. A clinically relevant threshold of MSE, SSIM, and/or PSNR may be defined beyond which point an MRI sequence should be rejected for target delineation based on its dissimilarity to an optimized sequence co-registration. All dissimilarity thresholds calculated using correlation coefficients with in-plane geometric uncertainty would need to be defined on a sequence-by-sequence basis and validated with patient data. This study utilized a realistic skull phantom and image similarity metrics to develop a methodology capable of quantitatively assessing whether a modern research-based MRI sequence can be co-registered to the Gamma Knife CBCT with equal or less than equal accuracy when compared to a clinically accepted protocol.

Design Considerations to Facilitate Clinical Radiological Evaluation of Implantable Biomedical Structures.

Clinical effectiveness of implantable medical devices would be improved with in situ monitoring to ensure device positioning, determine subsequent damage, measure biodegradation, and follow healing. While standard clinical imaging protocols are appropriate for diagnosing disease and injury, these protocols have not been vetted for imaging devices. This study investigated how radiologists use clinical imaging to detect the location and integrity of implanted devices and whether embedding nanoparticle contrast agents into devices can improve assessment. To mimic the variety of devices available, phantoms from hydrophobic polymer films and hydrophilic gels were constructed, with and without computed tomography (CT)-visible TaOx and magnetic resonance imaging (MRI)-visible Fe3O4 nanoparticles. Some phantoms were purposely damaged by nick or transection. Phantoms were implanted in vitro into tissue and imaged with clinical CT, MRI, and ultrasound. In a blinded study, radiologists independently evaluated whether phantoms were present, assessed the type, and diagnosed whether phantoms were damaged or intact. Radiologists identified the location of phantoms 80% of the time. However, without incorporated nanoparticles, radiologists correctly assessed damage in only 54% of cases. With an incorporated imaging agent, the percentage jumped to 86%. The imaging technique which was most useful to radiologists varied with the properties of phantoms. With benefits and drawbacks to all three imaging modalities, future implanted devices should be engineered for visibility in the modality which best fits the treated tissue, the implanted device's physical location, and the type of required information. Imaging protocols should also be tailored to best exploit the properties of the imaging agents.

Conventional and Deep Learning Methods for Skull Stripping in Brain MRI

Skull stripping in brain magnetic resonance volume has recently been attracting attention due to an increased demand to develop an efficient, accurate, and general algorithm for diverse datasets of the brain. Accurate skull stripping is a critical step for neuroimaging diagnostic systems because neither the inclusion of non-brain tissues nor removal of brain parts can be corrected in subsequent steps, which results in unfixed error through subsequent analysis. The objective of this review article is to give a comprehensive overview of skull stripping approaches, including recent deep learning-based approaches. In this paper, the current methods of skull stripping have been divided into two distinct groups—conventional or classical approaches, and convolutional neural networks or deep learning approaches. The potentials of several methods are emphasized because they can be applied to standard clinical imaging protocols. Finally, current trends and future developments are addressed giving special attention to recent deep learning algorithms.

An artificial intelligence framework for automatic segmentation and volumetry of vestibular schwannomas from contrast-enhanced T1-weighted and high-resolution T2-weighted MRI.

Automatic segmentation of vestibular schwannomas (VSs) from MRI could significantly improve clinical workflow and assist in patient management. Accurate tumor segmentation and volumetric measurements provide the best indicators to detect subtle VS growth, but current techniques are labor intensive and dedicated software is not readily available within the clinical setting. The authors aim to develop a novel artificial intelligence (AI) framework to be embedded in the clinical routine for automatic delineation and volumetry of VS. Imaging data (contrast-enhanced T1-weighted [ceT1] and high-resolution T2-weighted [hrT2] MR images) from all patients meeting the study's inclusion/exclusion criteria who had a single sporadic VS treated with Gamma Knife stereotactic radiosurgery were used to create a model. The authors developed a novel AI framework based on a 2.5D convolutional neural network (CNN) to exploit the different in-plane and through-plane resolutions encountered in standard clinical imaging protocols. They used a computational attention module to enable the CNN to focus on the small VS target and propose a supervision on the attention map for more accurate segmentation. The manually segmented target tumor volume (also tested for interobserver variability) was used as the ground truth for training and evaluation of the CNN. We quantitatively measured the Dice score, average symmetric surface distance (ASSD), and relative volume error (RVE) of the automatic segmentation results in comparison to manual segmentations to assess the model's accuracy. Imaging data from all eligible patients (n = 243) were randomly split into 3 nonoverlapping groups for training (n = 177), hyperparameter tuning (n = 20), and testing (n = 46). Dice, ASSD, and RVE scores were measured on the testing set for the respective input data types as follows: ceT1 93.43%, 0.203 mm, 6.96%; hrT2 88.25%, 0.416 mm, 9.77%; combined ceT1/hrT2 93.68%, 0.199 mm, 7.03%. Given a margin of 5% for the Dice score, the automated method was shown to achieve statistically equivalent performance in comparison to an annotator using ceT1 images alone (p = 4e-13) and combined ceT1/hrT2 images (p = 7e-18) as inputs. The authors developed a robust AI framework for automatically delineating and calculating VS tumor volume and have achieved excellent results, equivalent to those achieved by an independent human annotator. This promising AI technology has the potential to improve the management of patients with VS and potentially other brain tumors.

MRI based medical image analysis: Survey on brain tumor grade classification

Abstract A review on the recent segmentation and tumor grade classification techniques of brain Magnetic Resonance (MR) Images is the objective of this paper. The requisite for early detection of a brain tumor and its grade is the motivation for this study. In Magnetic Resonance Imaging (MRI), the tumor might appear clear but physicians need quantification of the tumor area for further treatment. This is where the digital image processing methodologies along with machine learning aid further diagnosis, treatment, prior and post-surgical procedures, synergizing between the radiologist and computer. These hybrid techniques provide a second opinion and assistance to radiologists in understanding medical images hence improving diagnostic accuracy. This article aims to retrospect the current trends in segmentation and classification relevant to tumor infected human brain MR images with a target on gliomas which include astrocytoma. The methodologies used for extraction and grading of tumors which can be integrated into the standard clinical imaging protocols are elucidated. Lastly, a crucial assessment of the state of the art, future developments and trends are dissertated.

EP-1455: IGRT kV-CBCT dose calculations using Virtual Source Models and validated in phantoms using OSL

Purpose or Objective: With the growing use of X-ray imaging equipment in Image-Guided RadioTherapy (IGRT), the need to evaluate the dose-to-organs delivered by kV-CBCT imaging acquisition increases. This study aims to propose accurate Monte Carlo (MC) calculations of the patient dose-to-organs delivered by two commercially available kV-CBCT systems: the XVI from Elekta’s VERSA HD accelerator and the OBI from Varian’s TrueBeam system. Simulations are to be validated using in phantom OSL measurements. Material and Methods: For both kV-CBCT systems, the kV irradiation head geometry was implemented in the MC simulation code Penelope. As a first step, the resulting photon distributions were expressed as Virtual Source Models (VSM) for every standard irradiation condition (kVp, filtration, collimation); it was then validated and adjusted using in water-phantom measurements performed with a calibrated Farmer-type ionization chamber. In a second step, the validated VSMs were used to simulate the dose delivered by both the XVI and OBI systems in anthropomorphic phantoms, using standard clinical imaging protocols. Simulated dose-to-organs were then confronted to dose measurements performed using OSL inserted into the same phantoms, following a dosimetric protocol for OSLs previously established [1]. In addition, VSM results were confronted to their direct MC counterparts in order to evaluate the benefit of using such technique. Results: The current study highlights the possibility to reproduce OSL dose-to-organ measurements using VSM-driven Monte Carlo simulation with an overall agreement better than 20 %. In addition, the use of VSM in the MC simulation enables to speed-up the calculation time by a factor better than two (for the same statistical uncertainty) compared to direct MC simulation. Nevertheless, if direct and VSM calculations are in agreement inside the irradiation field, outside, VSM results tend to be significantly lower (10-30%). Conclusion: The use of a VSM was demonstrated to simplify and fasten MC simulations for personalized kV-CBCT MC dose estimation. In addition, OSLs enable to perform the low dose measurement in the kV range needed for in phantom X-ray imaging equipment dose QA. This study is to be completed in the near future by the addition of other standard X-ray imaging equipment dedicated to IGRT.

Automatic cardiac LV segmentation in MRI using modified graph cuts with smoothness and interslice constraints

Magnetic resonance imaging (MRI), specifically late-enhanced MRI, is the standard clinical imaging protocol to assess cardiac viability. Segmentation of myocardial walls is a prerequisite for this assessment. Automatic and robust multisequence segmentation is required to support processing massive quantities of data. A generic rule-based framework to automatically segment the left ventricle myocardium is presented here. We use intensity information, and include shape and interslice smoothness constraints, providing robustness to subject- and study-specific changes. Our automatic initialization considers the geometrical and appearance properties of the left ventricle, as well as interslice information. The segmentation algorithm uses a decoupled, modified graph cut approach with control points, providing a good balance between flexibility and robustness. The method was evaluated on late-enhanced MRI images from a 20-patient in-house database, and on cine-MRI images from a 15-patient open access database, both using as reference manually delineated contours. Segmentation agreement, measured using the Dice coefficient, was 0.81±0.05 and 0.92±0.04 for late-enhanced MRI and cine-MRI, respectively. The method was also compared favorably to a three-dimensional Active Shape Model approach. The experimental validation with two magnetic resonance sequences demonstrates increased accuracy and versatility.

Semiautomated digital analysis of knee joint space width using MR images

The goal of this study was to (a) develop a semiautomated computer algorithm to measure knee joint space width (JSW) from magnetic resonance (MR) images using standard imaging techniques and (b) evaluate the reproducibility of the algorithm. Using a standard clinical imaging protocol, bilateral knee MR images were obtained twice within a 2-week period from 17 asymptomatic research participants. Images were analyzed to determine the variability of the measurements performed by the program compared with the variability of manual measurements. Measurement variability of the computer algorithm was considerably smaller than the variability of manual measurements. The average difference between two measurements of the same slice performed with the computer algorithm by the same user was 0.004 +/- 0.07 mm for the tibiofemoral joint (TF) and 0.009 +/- 0.11 mm for the patellofemoral joint (PF) compared with an average of 0.12 +/- 0.22 mm TF and 0.13 +/- 0.29 mm PF, respectively, for the manual method. Interuser variability of the computer algorithm was also considerably smaller, with an average difference of 0.004 +/- 0.1 mm TF and 0.0006 +/- 0.1 mm PF compared with 0.38 +/- 0.59 mm TF and 0.31 +/- 0.66 mm PF obtained using a manual method. The between-day reproducibility was larger but still within acceptable limits at 0.09 +/- 0.39 mm TF and 0.09 +/- 0.51 mm PF. This technique has proven consistently reproducible on a same slice base,while the reproducibility comparing different acquisitions of the same subject was larger. Longitudinal reproducibility improvement needs to be addressed through acquisition protocol improvements. A semiautomated method for measuring knee JSW from MR images has been successfully developed.

Thermal effects of MR imaging: worst-case studies on sheep.

The objective of this study was to provide a worst-case estimate of thermal effects of MR imaging by subjecting anesthetized unshorn sheep to power deposition at specific absorption rates (SARs) well above approved standards for periods of time in excess of normal clinical imaging protocols. A control period with no RF power was followed by 20-105 min of RF power application. Afterward, there was a 20-min or longer recovery period with no RF power. Eight sheep were given whole-body RF exposure (1.5- to 4-W/kg SAR) while rectal and skin temperatures were monitored. Four sheep were subjected to 4-W/kg head scans for an average of 75 min while temperatures of the cornea, vitreous humor, head skin, jugular vein, and rectum were measured. In head scanning experiments, skin and eye temperatures increased about 1.5 degrees C. Jugular vein temperature rose a maximum of 0.4 degrees C after an average exposure of 75 min. In whole-body exposures, elevation of rectal temperature was correlated with energy input. Deep-body temperature rises in excess of 2.0 degrees C were attained for 4-W/kg whole-body exposure periods greater than 82 min. Animals exposed for 40 min to 4 W/kg in either body coil (three sheep) or head coil (two sheep) were recovered and observed to be in good health for 10 weeks; no cataracts were found. MR power deposition at SAR levels well above typical clinical imaging protocols caused body temperature to increase. For exposure periods in excess of standard clinical imaging protocols the temperature increase was insufficient to cause adverse thermal effects. Studies in healthy humans are needed to determine whether enhanced heat-loss effector mechanisms are likely to cause deep-body temperatures to plateau at an acceptable level, and to elucidate mechanisms that determine subcutaneous temperature.