T2w Magnetic Resonance Imaging Research Articles

Multivariable Diagnostic Prediction Model to Detect Hormone Secretion Profile From T2W MRI Radiomics with Artificial Neural Networks in Pituitary Adenomas.

Objective:This study aims to develop neural networks to detect hormone secretion profiles in the pituitary adenomas based on T2 weighted magnetic resonance imaging (MRI) radiomics.Methods:This retrospective model-development study included a cohort of patients with pituitary adenomas (n=130) from January 2015 to January 2020 in one tertiary center. The mean age was 46.49±13.69 years, and 76/130 (58.46%) were women. Three observers segmented lesions on coronal T2 weighted MRI, and an interrater agreement was evaluated using the Dice coefficient. Predictors were determined as radiomics features (n=851). Feature selection was based on intraclass correlation coefficient, coefficient variance, variance inflation factor, and LASSO regression analysis. Outcomes were identified as 7 hormone secretion profiles [nonfunctioning pituitary adenoma, growth hormone-secreting adenomas, prolactinomas, adrenocorticotropic hormone-secreting adenomas, pluri-hormonal secreting adenomas (PHA), follicle-stimulating hormone and luteinizing hormone-secreting adenomas, and thyroid-stimulating hormone adenomas]. A multivariable diagnostic prediction model was developed with artificial neural networks (ANN) for 7 outcomes. ANN performance was presented as an area under the receiver operating characteristic curve (AUC) and accepted as successful if the AUC was >0.85 and p-value was <0.01.Results:The performance of the ANN distinguishing prolactinomas from other adenomas was validated (AUC=0.95, p<0.001, sensitivity: 91%, and specificity: 98%). The model distinguishing PHA had the lowest AUC (AUC=0.74 and p<0.001). The AUC values for the other five ANN were >0.85 and p values were <0.001.Conclusions:This study was successful in training neural networks that could differentiate the hormone secretion profile of pituitary adenomas.

Contrast enhancement: Cross-modal learning approach for medical images

Contrast is an imperative perceptible attribute embodying the image quality. In medical images, the poor quality specifically low contrast inhibits precise interpretation of the image. Contrast enhancement is, therefore, applied not merely to improve the visual quality of images but also enabling them to facilitate further processing tasks. We propose a contrast enhancement approach based on cross-modal learning in this paper. Cycle-GAN (Generative Adversarial Network) is used for this purpose, where UNet augmented with global features acts as a generator. Besides, individual batch normalization has been used to make generators adapt specifically to their input distributions. The proposed method accepts low contrast T2-weighted (T2-w) Magnetic Resonance images (MRI) and uses the corresponding high contrast T1-w MRI to learn the global contrast characteristics. The experiments were conducted on a publicly available IXI dataset. Comparison with recent CE methods and quantitative assessment using two prevalent metrics FSIM and BRISQUE validate the superior performance of the proposed method.

MRI-based automatic segmentation of rectal cancer using 2D U-Net on two independent cohorts

Background Tumor delineation is time- and labor-intensive and prone to inter- and intraobserver variations. Magnetic resonance imaging (MRI) provides good soft tissue contrast, and functional MRI captures tissue properties that may be valuable for tumor delineation. We explored MRI-based automatic segmentation of rectal cancer using a deep learning (DL) approach. We first investigated potential improvements when including both anatomical T2-weighted (T2w) MRI and diffusion-weighted MR images (DWI). Secondly, we investigated generalizability by including a second, independent cohort. Material and methods Two cohorts of rectal cancer patients (C1 and C2) from different hospitals with 109 and 83 patients, respectively, were subject to 1.5 T MRI at baseline. T2w images were acquired for both cohorts and DWI (b-value of 500 s/mm2) for patients in C1. Tumors were manually delineated by three radiologists (two in C1, one in C2). A 2D U-Net was trained on T2w and T2w + DWI. Optimal parameters for image pre-processing and training were identified on C1 using five-fold cross-validation and patient Dice similarity coefficient (DSCp) as performance measure. The optimized models were evaluated on a C1 hold-out test set and the generalizability was investigated using C2. Results For cohort C1, the T2w model resulted in a median DSCp of 0.77 on the test set. Inclusion of DWI did not further improve the performance (DSCp 0.76). The T2w-based model trained on C1 and applied to C2 achieved a DSCp of 0.59. Conclusion T2w MR-based DL models demonstrated high performance for automatic tumor segmentation, at the same level as published data on interobserver variation. DWI did not improve results further. Using DL models on unseen cohorts requires caution, and one cannot expect the same performance.

Effects of diet and social status on white matter integrity in female cynomolgus monkeys

AbstractBackgroundMounting evidence suggests that diet may influence risk for cognitive impairment and neurodegenerative disease, including Alzheimer’s disease. Nonhuman primates are important models of cognitive aging and Alzheimer’s disease‐like neuropathology given their complex central nervous systems and susceptibility to diet‐induced diseases. Diffusion magnetic resonance imaging (MRI) studies in humans report age‐related increases in mean diffusivity (MD) and decreases in fractional anisotropy (FA). Whether nonhuman primate models of Alzheimer’s disease‐like neuropathology show similar changes in diffusivity, and whether diet affects these changers is unknown. As such, the purpose of this study was to determine the longitudinal effects of diet (Mediterranean vs Western) and social subordination stress on diffusion MRI measures of white matter structural integrity in a nonhuman primate model of aging.MethodIn this IACUC approved study, thirty‐seven middle‐aged cynomolgus monkeys (Macaca fascicularis) were evaluated longitudinally with diffusion MRI measures (FA, MD, axial diffusivity [AxD], and radial diffusivity [RD]). Monkeys were scanned at baseline and after 31 months of diet consumption (∼ equivalent to a 9‐year follow up in humans). Diffusion MRIs were preprocessed (corrected for eddy current and geometric distortion) and co‐registered to high resolution T1w MRI to derive diffusion measures. For statistical analyses, cerebral white matter and 9 bilateral regions of interest (ROIs) were selected from NeuroMaps nonhuman primate atlas. Statistical analyses were performed with repeated measures analysis of variance.ResultWe observed social status differences in frontal (p=0.021) and occipital (p=0.029) white mater ROIs, with greater longitudinal increase or smaller longitudinal decrease in FA among subordinates compared to dominants. Diet and social status interacted to impact AxD in the genu of the corpus callosum (p=0.028). Finally, subordinates in the Mediterranean group demonstrated greater longitudinal diffusivity increases compared to 1) dominants in the Mediterranean group, 2) dominants in the Western group, and 3) subordinates in the Western group (Figure below).ConclusionWhile the biological mechanisms underlying these findings require future investigation, our observations suggest that diet and social subordination stress illicit changes in structural white matter integrity that could modify vulnerability to Alzheimer’s disease‐like neuropathology in middle‐aged nonhuman primates.

Multisite reproducibility and test-retest reliability of the T1w/T2w-ratio: A comparison of processing methods

BackgroundThe ratio of T1-weighted (T1w) and T2-weighted (T2w) magnetic resonance imaging (MRI) images is often used as a proxy measure of cortical myelin. However, the T1w/T2w-ratio is based on signal intensities that are inherently non-quantitative and known to be affected by extrinsic factors. To account for this a variety of processing methods have been proposed, but a systematic evaluation of their efficacy is lacking. Given the dependence of the T1w/T2w-ratio on scanner hardware and T1w and T2w protocols, it is important to ensure that processing pipelines perform well also across different sites. MethodsWe assessed a variety of processing methods for computing cortical T1w/T2w-ratio maps, including correction methods for nonlinear field inhomogeneities, local outliers, and partial volume effects as well as intensity normalisation. These were implemented in 33 processing pipelines which were applied to four test-retest datasets, with a total of 170 pairs of T1w and T2w images acquired on four different MRI scanners. We assessed processing pipelines across datasets in terms of their reproducibility of expected regional distributions of cortical myelin, lateral intensity biases, and test-retest reliability regionally and across the cortex. Regional distributions were compared both qualitatively with histology and quantitatively with two reference datasets, YA-BC and YA-B1+, from the Human Connectome Project. ResultsReproducibility of raw T1w/T2w-ratio distributions was overall high with the exception of one dataset. For this dataset, Spearman rank correlations increased from 0.27 to 0.70 after N3 bias correction relative to the YA-BC reference and from -0.04 to 0.66 after N4ITK bias correction relative to the YA-B1+ reference. Partial volume and outlier corrections had only marginal effects on the reproducibility of T1w/T2w-ratio maps and test-retest reliability. Before intensity normalisation, we found large coefficients of variation (CVs) and low intraclass correlation coefficients (ICCs), with total whole-cortex CV of 10.13% and whole-cortex ICC of 0.58 for the raw T1w/T2w-ratio. Intensity normalisation with WhiteStripe, RAVEL, and Z-Score improved total whole-cortex CVs to 5.91%, 5.68%, and 5.19% respectively, whereas Z-Score and Least Squares improved whole-cortex ICCs to 0.96 and 0.97 respectively. ConclusionsIn the presence of large intensity nonuniformities, bias field correction is necessary to achieve acceptable correspondence with known distributions of cortical myelin, but it can be detrimental in datasets with less intensity inhomogeneity. Intensity normalisation can improve test-retest reliability and inter-subject comparability. However, both bias field correction and intensity normalisation methods vary greatly in their efficacy and may affect the interpretation of results. The choice of T1w/T2w-ratio processing method must therefore be informed by both scanner and acquisition protocol as well as the given study objective. Our results highlight limitations of the T1w/T2w-ratio, but also suggest concrete ways to enhance its usefulness in future studies.

Deep Learning-Based Detection of Malformed Optic Chiasms From MRI Images.

Convolutional neural network (CNN) models are of great promise to aid the segmentation and analysis of brain structures. Here, we tested whether CNN trained to segment normal optic chiasms from the T1w magnetic resonance imaging (MRI) image can be also applied to abnormal chiasms, specifically with optic nerve misrouting as typical for human albinism. We performed supervised training of the CNN on the T1w images of control participants (n = 1049) from the Human Connectome Project (HCP) repository and automatically generated algorithm-based optic chiasm masks. The trained CNN was subsequently tested on data of persons with albinism (PWA; n = 9) and controls (n = 8) from the CHIASM repository. The quality of outcome segmentation was assessed via the comparison to manually defined optic chiasm masks using the Dice similarity coefficient (DSC). The results revealed contrasting quality of masks obtained for control (mean DSC ± SEM = 0.75 ± 0.03) and PWA data (0.43 ± 0.8, few-corrected p = 0.04). The fact that the CNN recognition of the optic chiasm fails for chiasm abnormalities in PWA underlines the fundamental differences in their spatial features. This finding provides proof of concept for a novel deep-learning-based diagnostics approach of chiasmal misrouting from T1w images, as well as further analyses on chiasmal misrouting and their impact on the structure and function of the visual system.

Ultrashort echo time MRI of the lung in children and adolescents: comparison with non-enhanced computed tomography and standard post-contrast T1w MRI sequences

ObjectivesTo compare the diagnostic value of ultrashort echo time (UTE) magnetic resonance imaging (MRI) for the lung versus the gold standard computed tomography (CT) and two T1-weighted MRI sequences in children.MethodsTwenty-three patients with proven oncologic disease (14 male, 9 female; mean age 9.0 + / − 5.4 years) received 35 low-dose CT and MRI examinations of the lung. The MRI protocol (1.5-T) included the following post-contrast sequences: two-dimensional (2D) incoherent gradient echo (GRE; acquisition with breath-hold), 3D volume interpolated GRE (breath-hold), and 3D high-resolution radial UTE sequences (performed during free-breathing). Images were evaluated by considering image quality as well as distinct diagnosis of pulmonary nodules and parenchymal areal opacities with consideration of sizes and characterisations.ResultsThe UTE technique showed significantly higher overall image quality, better sharpness, and fewer artefacts than both other sequences. On CT, 110 pulmonary nodules with a mean diameter of 4.9 + / − 2.9 mm were detected. UTE imaging resulted in a significantly higher detection rate compared to both other sequences (p < 0.01): 76.4% (84 of 110 nodules) for UTE versus 60.9% (67 of 110) for incoherent GRE and 62.7% (69 of 110) for volume interpolated GRE sequences. The detection of parenchymal areal opacities by the UTE technique was also significantly higher with a rate of 93.3% (42 of 45 opacities) versus 77.8% (35 of 45) for 2D GRE and 80.0% (36 of 45) for 3D GRE sequences (p < 0.05).ConclusionThe UTE technique for lung MRI is favourable in children with generally high diagnostic performance compared to standard T1-weighted sequences as well as CT.Key Points• Due to the possible acquisition during free-breathing of the patients, the UTE MRI sequence for the lung is favourable in children.• The UTE technique reaches higher overall image quality, better sharpness, and lower artefacts, but not higher contrast compared to standard post-contrast T1-weighted sequences.• In comparison to the gold standard chest CT, the detection rate of small pulmonary nodules small nodules ≤ 4 mm and subtle parenchymal areal opacities is higher with the UTE imaging than standard T1-weighted sequences.

Anatomical atlas of the upper part of the human head for electroencephalography and bioimpedance applications

Objective. The objective of this work is to develop a 4D (3D+T) statistical anatomical atlas of the electrical properties of the upper part of the human head for cerebral electrophysiology and bioimpedance applications. Approach. The atlas was constructed based on 3D magnetic resonance images (MRI) of 107 human individuals and comprises the electrical properties of the main internal structures and can be adjusted for specific electrical frequencies. T1w+T2w MRI images were used to segment the main structures of the head while angiography MRI was used to segment the main arteries. The proposed atlas also comprises a time-varying model of arterial brain circulation, based on the solution of the Navier–Stokes equation in the main arteries and their vascular territories. Main results. High-resolution, multi-frequency and time-varying anatomical atlases of resistivity, conductivity and relative permittivity were created and evaluated using a forward problem solver for EIT. The atlas was successfully used to simulate electrical impedance tomography measurements indicating the necessity of signal-to-noise between 100 and 125 dB to identify vascular changes due to the cardiac cycle, corroborating previous studies. The source code of the atlas and solver are freely available to download. Significance. Volume conductor problems in cerebral electrophysiology and bioimpedance do not have analytical solutions for nontrivial geometries and require a 3D model of the head and its electrical properties for solving the associated PDEs numerically. Ideally, the model should be made with patient-specific information. In clinical practice, this is not always the case and an average head model is often used. Also, the electrical properties of the tissues might not be completely known due to natural variability. Anatomical atlases are important tools for in silico studies on cerebral circulation and electrophysiology that require statistically consistent data, e.g. machine learning, sensitivity analyses, and as a benchmark to test inverse problem solvers.

CEREBRUM-7T: Fast and Fully Volumetric Brain Segmentation of 7 Tesla MR Volumes.

Ultra‐high‐field magnetic resonance imaging (MRI) enables sub‐millimetre resolution imaging of the human brain, allowing the study of functional circuits of cortical layers at the meso‐scale. An essential step in many functional and structural neuroimaging studies is segmentation, the operation of partitioning the MR images in anatomical structures. Despite recent efforts in brain imaging analysis, the literature lacks in accurate and fast methods for segmenting 7‐tesla (7T) brain MRI. We here present CEREBRUM‐7T, an optimised end‐to‐end convolutional neural network, which allows fully automatic segmentation of a whole 7T T1w MRI brain volume at once, without partitioning the volume, pre‐processing, nor aligning it to an atlas. The trained model is able to produce accurate multi‐structure segmentation masks on six different classes plus background in only a few seconds. The experimental part, a combination of objective numerical evaluations and subjective analysis, confirms that the proposed solution outperforms the training labels it was trained on and is suitable for neuroimaging studies, such as layer functional MRI studies. Taking advantage of a fine‐tuning operation on a reduced set of volumes, we also show how it is possible to effectively apply CEREBRUM‐7T to different sites data. Furthermore, we release the code, 7T data, and other materials, including the training labels and the Turing test.

A Pipeline for Predicting the Treatment Response of Neoadjuvant Chemoradiotherapy for Locally Advanced Rectal Cancer Using Single MRI Modality: Combining Deep Segmentation Network and Radiomics Analysis Based on "Suspicious Region".

Patients with locally advanced rectal cancer (LARC) who achieve a pathologic complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT) typically have a good prognosis. An early and accurate prediction of the treatment response, i.e., whether a patient achieves pCR, could significantly help doctors make tailored plans for LARC patients. This study proposes a pipeline of pCR prediction using a combination of deep learning and radiomics analysis. Taking into consideration missing pre-nCRT magnetic resonance imaging (MRI), as well as aiming to improve the efficiency for clinical application, the pipeline only included a post-nCRT T2-weighted (T2-w) MRI. Unlike other studies that attempted to carefully find the region of interest (ROI) using a pre-nCRT MRI as a reference, we placed the ROI on a “suspicious region”, which is a continuous area that has a high possibility to contain a tumor or fibrosis as assessed by radiologists. A deep segmentation network, termed the two-stage rectum-aware U-Net (tsraU-Net), is designed to segment the ROI to substitute for a time-consuming manual delineation. This is followed by a radiomics analysis model based on the ROI to extract the hidden information and predict the pCR status. The data from a total of 275 patients were collected from two hospitals and partitioned into four datasets: Seg-T (N = 88) for training the tsraUNet, Rad-T (N = 107) for building the radiomics model, In-V (N = 46) for internal validation, and Ex-V (N = 34) for external validation. The proposed method achieved an area under the curve (AUC) of 0.829 (95% confidence interval [CI]: 0.821, 0.837) on In-V and 0.815 (95% CI, 0.801, 0.830) on Ex-V. The performance of the method was considerable and stable in two validation sets, indicating that the well-designed pipeline has the potential to be used in real clinical procedures.

Abstract 3007: Imaging characterization of NCI patient derived model repository (PDMR) xenografts

Abstract The purpose of this work is to provide imaging data on patient derived xenografts models that are available from the National Cancer Institute Patient-Derived Models Repository (https://pdmr.cancer.gov) that may be useful to investigators in deciding which models to use in testing therapeutics or designing co-clinical trials. Using imaging to monitor progress of a treatment or evaluate development or prevention of metastases has advantages in minimizing the number of mice required as well as providing more information about metabolic status and tumor heterogeneity. NSG mice were implanted with each model and monitored with non-invasive imaging. The core imaging methodologies were T2W magnetic resonance imaging (MRI), to visualize heterogeneity and determine the usefulness of MRI to detect metastatic disease; 18F fluorodeoxyglucose PET/CT, to evaluate aerobic glycolysis; and 3'-Deoxy-3'-(18)F-fluorothymidine PET/CT, to evaluate proliferation. This work is ongoing and to date we have performed MRI on over 80 models with diverse tumor histology with a subset also undergoing FDG PET (77 models) and FLT PET (36 models). These models and their imaging characteristics will be presented. Nineteen models have demonstrated metastatic disease that could be detected by non-contrast MRI imaging with pathological confirmation, including ones derived from melanoma, lung, colon, pancreatic, uterine, anal, and head &amp; neck. Four of these (two colon adenocarcinoma, one melanoma and one pancreatic adenocarcinoma) have undergone more extensive characterization evaluating cohorts to determine location, timing of appearance, and penetrance of metastasis. These experiments evaluated cohorts of animals who were monitored as the primary tumors grew and others for whom the primary tumor was excised to encourage metastatic spread. The results of this metastatic characterization will be summarized. Additionally, the image data and the appropriate Standard Operating Procedures (SOP's) will be posted in The Cancer Imagining Archive (TCIA) https://www.cancerimagingarchive.net/ for investigators to access. The goal of this poster is to make investigators aware of the availability of this complementary imaging data as they consider research on models accessible from the NCI PDMR. Citation Format: Paula M. Jacobs, James L. Tatum, Joseph D. Kalen, Lilia V. Ileva, Lisa A. Riffle, Keita Saito, Nimit L. Patel, Jessica Phillips, Melinda Hollingshead, Yvonne A. Evrard, Michelle Gottholm-Ahalt, Chelsea Sanders, Amy James, Simone Difilippantonio, Elijah F. Edmondson, James H. Doroshow. Imaging characterization of NCI patient derived model repository (PDMR) xenografts [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 3007.

Can pre-procedural MRI signal intensity ratio predict the success of uterine artery embolization in treatment of myomas?

Background/aim Magnetic resonance (MR) images, signal intensity ratios calculated using region of interests (ROI) in T2W images by proportioning the dominant myoma to iliac muscle can aid patient selection and, thus, in achieving better outcomes with the uterine artery embolization (UAE) procedure. The present study investigates the association between the success of UAE treatment with signal intensity (SI) ratio of the dominant myoma to the iliac muscle in MR imaging performed prior to the procedure.Materials and methods This is a retrospective study and included 30 patients who admitted to our clinic between February 2017 and July 2019 due to symptoms associated with myoma and who underwent MR imaging before and after UAE treatment. All patients, MR images obtained before UAE treatment and at the 12th month after the procedure were evaluated. In MRI, SI values were calculated by proportioning the dominant myoma to the iliac muscle using circular ROI in T1 weighted (W), T2W, and post-contrast T1W images. In the present study, 50% or more volumetric regression of the myoma with infarction of fibroids (loss of enhancement) at the 12-month follow-up MRI after the procedure was considered a successful procedure. Results Myoma volumes calculated in MR images showed significant differences between the MRI performed before UAE procedure and the MRI performed at the 12th month after the procedure (p < 0.0001). SI ratio calculated from pre-procedure T2W MR images was found to be a significant determinant of 50% or more volumetric regression in the myoma after UAE procedure (p = 0.017), T1W, post-contrast T1W images were not statistically significant (p = 0.211).Conclusion Our results indicate that SI ratio of the dominant myoma to the iliac muscle calculated using ROI in T2W images of MR studies performed before UAE procedure can predict the success of the procedure.

Semi-Supervised Learning in Medical MRI Segmentation: Brain Tissue with White Matter Hyperintensity Segmentation Using FLAIR MRI.

White-matter hyperintensity (WMH) is a primary biomarker for small-vessel cerebrovascular disease, Alzheimer’s disease (AD), and others. The association of WMH with brain structural changes has also recently been reported. Although fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging (MRI) provide valuable information about WMH, FLAIR does not provide other normal tissue information. The multi-modal analysis of FLAIR and T1-weighted (T1w) MRI is thus desirable for WMH-related brain aging studies. In clinical settings, however, FLAIR is often the only available modality. In this study, we thus propose a semi-supervised learning method for full brain segmentation using FLAIR. The results of our proposed method were compared with the reference labels, which were obtained by FreeSurfer segmentation on T1w MRI. The relative volume difference between the two sets of results shows that our proposed method has high reliability. We further evaluated our proposed WMH segmentation by comparing the Dice similarity coefficients of the reference and the results of our proposed method. We believe our semi-supervised learning method has a great potential for use for other MRI sequences and will encourage others to perform brain tissue segmentation using MRI modalities other than T1w.

The role of the iliotibial band cross-sectional area as a morphological parameter of the iliotibial band friction syndrome: a retrospective pilot study.

BackgroundIliotibial band friction syndrome (ITBFS) is a common disorder of the lateral knee. Previous research has reported that the iliotibial band (ITB) thickness (ITBT) is correlated with ITBFS, and ITBT has been considered to be a key morphologic parameter of ITBFS. However, the thickness is different from inflammatory hypertrophy. Thus, we made the ITB cross-sectional area (ITBCSA) a new morphological parameter to assess ITBFS.MethodsForty-three patients with ITBFS group and from 43 normal group who underwent T1W magnetic resonance imaging were enrolled. The ITBCSA was measured as the cross-sectional area of the ITB that was most hypertrophied in the magnetic resonance axial images. The ITBT was measured as the thickest site of ITB.ResultsThe mean ITBCSA was 25.24 ± 6.59 mm2 in the normal group and 38.75 ± 9.11 mm2 in the ITBFS group. The mean ITBT was 1.94 ± 0.41 mm in the normal group and 2.62 ± 0.46 mm in the ITBFS group. Patients in ITBFS group had significantly higher ITBCSA (P < 0.001) and ITBT (P < 0.001) than the normal group. A receiver operator characteristic curve analysis demonstrated that the best cut-off value of the ITBT was 2.29 mm, with 76.7% sensitivity, 79.1% specificity, and area under the curve (AUC) 0.88. The optimal cut-off score of the ITBCSA was 30.66 mm2, with 79.1% sensitivity, 79.1% specificity, and AUC 0.87.ConclusionsITBCSA is a new and sensitive morphological parameter for diagnosing ITBFS, and may even be more accurate than ITBT.

16461 Comparison of voxel intensity standardization methods in head and neck cancer magnetic resonance imaging

ABSTRACT IMPACT: This work will standardize necessary image pre-processing for diagnostic and prognostic clinical workflows dependent on quantitative analysis of conventional magnetic resonance imaging. OBJECTIVES/GOALS: Conventional magnetic resonance imaging (MRI) poses challenges for quantitative analysis due to a lack of uniform inter-scanner voxel intensity values. Head and neck cancer (HNC) applications in particular have not been well investigated. This project aims to systematically evaluate voxel intensity standardization (VIS) methods for HNC MRI. METHODS/STUDY POPULATION: We utilize two separate cohorts of HNC patients, where T2-weighted (T2-w) MRI sequences were acquired before beginning radiotherapy for five patients in each cohort. The first cohort corresponds to patients with images taken at various institutions with a variety of non-uniform acquisition scanners and parameters. The second cohort corresponds to patients from a prospective clinical trial with uniformity in both scanner and acquisition parameters. Regions of interest from a variety of healthy tissues assumed to have minimal interpatient variation were manually contoured for each image and used to compare differences between a variety of VIS methods for each cohort. Towards this end, we implement a new metric for cohort intensity distributional overlap to compare region of interest similarity in a given cohort. RESULTS/ANTICIPATED RESULTS: Using a simple and interpretable metric, we have systematically investigated the effects of various commonly implementable VIS methods on T2-w sequences for two independent cohorts of HNC patients based on region of interest intensity similarity. We demonstrate VIS has a substantial effect on T2-w images where non-uniform acquisition parameters and scanners are utilized. Oppositely, it has a modest to minimal impact on T2-w images generated from the same scanner with the same acquisition parameters. Moreover, with a few notable exceptions, there does not seem to be a clear advantage or disadvantage to using one VIS method over another for T2-w images with non-uniform acquisition parameters. DISCUSSION/SIGNIFICANCE OF FINDINGS: Our results inform which VIS methods should be favored in HNC MRI and may indicate VIS is not a critical factor to consider in circumstances where similar acquisition parameters can be utilized. Moreover, our results can help guide downstream quantitative imaging tasks that may one day be implemented in clinical workflows.

Generalizing deep whole-brain segmentation for post-contrast MRI with transfer learning.

Purpose: Generalizability is an important problem in deep neural networks, especially with variability of data acquisition in clinical magnetic resonance imaging (MRI). Recently, the spatially localized atlas network tiles (SLANT) can effectively segment whole brain, non-contrast T1w MRI with 132 volumetric labels. Transfer learning (TL) is a commonly used domain adaptation tool to update the neural network weights for local factors, yet risks degradation of performance on the original validation/test cohorts. Approach: We explore TL using unlabeled clinical data to address these concerns in the context of adapting SLANT to scanning protocol variations. We optimize whole-brain segmentation on heterogeneous clinical data by leveraging 480 unlabeled pairs of clinically acquired T1w MRI with and without intravenous contrast. We use labels generated on the pre-contrast image to train on the post-contrast image in a five-fold cross-validation framework. We further validated on a withheld test set of 29 paired scans over a different acquisition domain. Results: Using TL, we improve reproducibility across imaging pairs measured by the reproducibility Dice coefficient (rDSC) between the pre- and post-contrast image. We showed an increase over the original SLANT algorithm (rDSC 0.82 versus 0.72) and the FreeSurfer v6.0.1 segmentation pipeline ( ). We demonstrate the impact of this work decreasing the root-mean-squared error of volumetric estimates of the hippocampus between paired images of the same subject by 67%. Conclusion: This work demonstrates a pipeline for unlabeled clinical data to translate algorithms optimized for research data to generalize toward heterogeneous clinical acquisitions.

An unusual case of oedematous prostate volumetric changes observed over the course of radiotherapy on the MR linear accelerator

IntroductionThe integration of magnetic resonance (MR) imaging into radiotherapy through new technology, including the MR -linear accelerator (MRL), has allowed further advancements into image guided radiotherapy (IGRT). Better soft tissue visualisation has led to some unusual findings. Case and outcomesA patient with T1c N0 M0 prostate adenocarcinoma received 60Gy in 20# radiotherapy on the MRL. Radiotherapy planning (RTP) scans were completed on both CT and MR (using T2 and T1 weighted three-dimensional turbo spin echo sequences, reconstructed transaxially (TRA). The MR scans revealed atypical oedema in the right peripheral zone, visualised on T2-weighted (T2w) MR Images as an accumulation of high signal intensity fluid. Daily MRL treatment includes a (T2w 3D Tra) sequence with which oedematous changes could be monitored. The images demonstrated an increase in oedematous volume over fractions 1–10 causing the prostate contour variations from the initial planning scans. Despite the prostate volume variations PTV coverage was never breached and dose constraints were always met for both PTV and surrounding organs at risk (OAR's), excluding the need for oncologist input.A single Therapeutic Radiographer (RTT) experienced in MRL delivery, contoured the prostate and oedematous volumes on the radiotherapy plan (RTP) MR and all on-treatment MR images to assess change over the radiotherapy course. The initial volumes were 53.4 cm3 and 8.3 cm3 for the prostate plus oedema and oedema alone respectively. The most significant change was seen for both the prostate and oedema on fraction nine (68.0 cm3 and 10.1 cm3, respectively). Reductions were noted after this with final (fraction 20) volumes of 55.2 cm3 and 0.58 cm3 respectively. DiscussionThe ability to visualise prostatic oedema was new to the radiotherapy treatment team due to better soft tissue visualisation than standard radiotherapy. The results from contouring the prostate and oedema volumes confirmed radiographer observations and demonstrated how oedema impacted the overall prostate volume by quantifying the oedematous variations over time. The changes in oedema volume are presumed to be in response to radiotherapy. ConclusionFurther adaptive radiotherapy work-flow developments, utilising an “Adapt to Shape” model will allow real-time re-contouring of the prostate to ensure tumour control is not compromised. Further work investigating the frequency and impact of oedemotous changes to external beam prostate patients will help to inform practice.

Intensive Teenage Activity Is Associated With Greater Muscle Hyperintensity on T1W Magnetic Resonance Imaging in Adults With Dysferlinopathy.

Practice of sports during childhood or adolescence correlates with an earlier onset and more rapidly progressing phenotype in dysferlinopathies. To determine if this correlation relates to greater muscle pathology that persists into adulthood, we investigated the effect of exercise on the degree of muscle fatty replacement measured using muscle MRI. We reviewed pelvic, thigh and leg T1W MRI scans from 160 patients with genetically confirmed dysferlinopathy from the Jain Foundation International clinical outcomes study in dysferlinopathy. Two independent assessors used the Lamminen-Mercuri visual scale to score degree of fat replacement in each muscle. Exercise intensity for each individual was defined as no activity, minimal, moderate, or intensive activity by using metabolic equivalents and patient reported frequency of sports undertaken between the ages of 10 and 18. We used ANCOVA and linear modeling to compare the mean Lamminen-Mercuri score for the pelvis, thigh, and leg between exercise groups, controlling for age at assessment and symptom duration. Intensive exercisers showed greater fatty replacement in the muscles of the pelvis than moderate exercisers, but no significant differences of the thigh or leg. Within the pelvis, Psoas was the muscle most strongly associated with this exercise effect. In patients with a short symptom duration of <15 years there was a trend toward greater fatty replacement in the muscles of the thigh. These findings define key muscles involved in the exercise-phenotype effect that has previously been observed only clinically in dysferlinopathy and support recommendations that pre-symptomatic patients should avoid very intensive exercise.

Evaluation of patient-specific MR distortion correction schemes for improved target localization accuracy in SRS.

This work aims at promoting target localization accuracy in cranial stereotactic radiosurgery (SRS) applications by focusing on the correction of sequence-dependent (also patient induced) magnetic resonance (MR) distortions at the lesion locations. A phantom-based quality assurance (QA) methodology was developed and implemented for the evaluation of three distortion correction techniques. The same approach was also adapted to cranial MR images used for SRS treatment planning purposes in single or multiple brain metastases cases. A three-dimensional (3D)-printed head phantom was filled with a 3D polymer gel dosimeter. Following treatment planning and dose delivery, volumes of radiation-induced polymerization served as hypothetical lesions, offering adequate MR contrast with respect to the surrounding unirradiated areas. T1-weighted (T1w) MR imaging was performed at 1.5T using the clinical scanning protocol for SRS. Additional images were acquired to implement three distortion correction methods; the field mapping (FM), mean image (MI) and signal integration (SI) techniques. Reference lesion locations were calculated as the averaged centroid positions of each target identified in the forward and reverse read gradient polarity MRI scans. The same techniques and workflows were implemented for the correction of contrast-enhanced T1w MR images of 10 patients with a total of 27 brain metastases. All methods employed in the phantom study diminished spatial distortion. Median and maximum distortion magnitude decreased from 0.7mm (2.10ppm) and 0.8mm (2.36ppm), respectively, to <0.2mm (0.61ppm) at all target locations, using any of the three techniques. Image quality of the corrected images was acceptable, while contrast-to-noise ratio slightly increased. Results of the patient study were in accordance with the findings of the phantom study. Residual distortion in corrected patient images was found to be <0.3mm in the vast majority of targets. Overall, the MI approach appears to be the most efficient correction method from the three investigated. In cranial SRS applications, patient-specific distortion correction at the target location(s) is feasible and effective, despite the expense of longer imaging time since additional MRI scan(s) need to be performed. A phantom-based QA methodology was developed and presented to reassure efficient implementation of correction techniques for sequence-dependent spatial distortion.

Increased signal intensity of spinal cord on T2W magnetic resonance imaging for cervical spondylotic myelopathy patients: Risk factors and prognosis (a STROBE-compliant article).

To investigate the risk factors for progression of increased signal intensity (ISI) on T2W magnetic resonance imaging (MRI) and its prognostic value in patients with cervical spondylotic myelopathy (CSM).A total of 109 patients with CSM were included in this study. All the patients were treated with anterior cervical discectomy and fusion. MRI was performed for all 109 patients preoperatively and at the final follow-up. Radiological evaluation included ISI, anterior compression (AC) of dural and spinal cord, hyperintensity region (HR) at the involved level. Clinical data including Japanese Orthopedic Association (JOA) score, Neck Disability Index (NDI) score, and Visual Analogue Scale were collected and evaluated. Patients were divided into 2 groups according to ISI grades (Group A: no hyper-intensity; Group B: presence of ISI). Then all patients presented with ISI were divided into 2 subgroups based on the range of HR (Group B1: hyper-intensity diameter accounts for less than half of the spinal cord diameter at the involved level; Group B2, hyper-intensity diameter accounts for more than half of the spinal cord diameter at the involved level). AC, disease duration, age, and gender were analyzed as potential risk factors.Significantly better JOA and NDI scores were observed in Group A preoperatively and at the final follow-up, compared to Group B (P < .05). Disease duration was found significantly longer in patients with ISI (P < .05). Notably better JOA and NDI scores were noticed in Group B1 rather than Group B2 (P < .05). Logistical regression showed that disease duration was the only factor that significantly correlated with the progress of ISI (P < .001).CSM patients with ISI on T2W MR images had poorer surgical outcomes compared to others, while the increased range of HR may deteriorate preoperative neurological function. Moreover, patients with longer disease duration had greater risk of ISI in spinal cord.