T2-weighted Magnetic Resonance Imaging Images Research Articles

Axillary lymph node response to neoadjuvant systemic therapy with dedicated axillary hybrid 18F-FDG PET/MRI in clinically node-positive breast cancer patients: a pilot study

To investigate the diagnostic performance of dedicated axillary hybrid 18F-2-[18F]-fluoro-2-deoxy-d-glucose (FDG) positron emission tomography (PET)/magnetic resonance imaging (MRI) in detecting axillary pathological complete response (pCR) following neoadjuvant systemic therapy (NST) in clinically node-positive breast cancer patients. Ten prospectively included clinically node-positive breast cancer patients underwent dedicated axillary hybrid 18F-FDG PET/MRI after completing NST followed by axillary surgery. PET images were reviewed by a nuclear medicine physician and coronal T1-weighted and T2-weighted MRI images by a radiologist. All axillary lymph nodes visible on PET/MRI were matched with those removed during axillary surgery. Diagnostic performance parameters were calculated based on patient-by-patient and node-by-node validation with histopathology of the axillary surgical specimen as the reference standard. Six patients achieved axillary pCR at final histopathology. A total of 84 surgically harvested axillary lymph nodes were matched with axillary lymph nodes depicted on PET/MRI. Histopathological examination of the matched axillary lymph nodes resulted in 10 lymph nodes with residual axillary disease of which eight contained macrometastases and two micrometastases. The patient-by-patient analysis yielded a sensitivity, specificity, positive predictive value, and negative predictive value of 25%, 100%, 100%, and 67%, respectively. The diagnostic performance parameters of the node-by-node analysis were 0%, 96%, 0%, and 88%, respectively. Excluding micrometastases from the node-by-node analysis increased the negative predictive value to 90%. This pilot study suggests that the negative predictive value and sensitivity of dedicated axillary 18F-FDG PET/MRI are insufficiently accurate to detect axillary pCR or exclude residual axillary disease following NST in clinically node-positive breast cancer patients.

152-OR: Evidence Linking Loss of Hypothalamic Perineuronal Nets to Obesity Pathogenesis

Defective neurocircuit activity within the hypothalamic arcuate nucleus (ARC) , a crucial brain area for the regulation of metabolism, is implicated in obesity pathogenesis. Perineuronal nets (PNNs) are extracellular matrix structures that regulate neurocircuit activity, and PNN loss can adversely influence the activity of the enmeshed neurons. To determine if defective ARC PNNs might contribute to obesity, we first quantified PNN abundance in obese mice induced by consuming a high-fat high-sucrose diet (HFHS; Bio-Serv #F4997) 12 wk. We found that compared to chow-fed controls, HFHS diet feeding caused an 84.3% decrease (p<0.0001) in the abundance of PNNs enmeshing ARC neurons. This loss was also associated with increases of ARC microgliosis (1.5-fold; p=0.004) and astrogliosis (2.9-fold; p=0.006) , suggesting a link between ARC PNN loss and glial activation. To translate these findings to humans, we analyzed postmortem hypothalamic samples from 9 subjects (Sex, 7M/2F; Age, 22-71y) that had previously undergone quantitative T2 magnetic resonance imaging (MRI) to assess obesity-associated hypothalamic gliosis. Consistent with our findings in mice, ARC PNN content was strongly and inversely correlated with the hypothalamic gliosis score measured by T2-weighted MRI imaging (R2=0.63, p=0.01) . To assess whether this depletion of ARC PNNs might contribute to obesity, we microinjected either the PNN-digesting enzyme, ChABC, or heat-inactivated ChABC control, into the rat ARC at the onset of a 4 wk HFHS dietary intervention. We found that compared to controls, the magnitude of diet-induced hyperphagia (17%; p<0.0001) , weight gain (41%; p<0.0001) , and excess body fat accumulation (28%; p=0.01) increased following PNN digestion, driven largely by increased dark cycle food intake (24%; p=0.006) . These findings collectively identify ARC PNNs as novel participants in the central control of energy homeostasis, and implicate their loss as a potential contributor to obesity pathogenesis. Disclosure K.M.Alonge: None. G.J.Morton: Research Support; Novo Nordisk A/S. E.Schur: None. J.Thaler: None. M.W.Schwartz: None. M.D.Dorfman: None. J.M.Scarlett: None. S.J.Melhorn: None. Y.Ye: None. S.J.W.Hu: None. K.L.Francis: None. A.Hendrickson: None. K.Cui: None. Funding AG066509DK017047AG074152DK101997DK720269

Co-occurring Acromioclavicular Joint Cyst and Hemarthrosis of the Shoulder Associated With Rotator Cuff Tear Arthropathy

Both acromioclavicular joint (ACJ) cysts and hemarthrosis of the shoulder are rare conditions of massive rotator cuff tear that eventually lead to cuff tear arthropathy. We herein report the first case of a patient with co-occurring ACJ cyst and hemarthrosis of the shoulder. An 80-year-old right-hand-dominant man presented to our outpatient department with a six-month history of repeatable right shoulder pain and swelling. Clinical examination revealed a 5 x 5 x 5 cm elastic hard or hard shoulder lump overlying the ACJ on skin with subcutaneous bleeding and swelling of the shoulder. Shoulder pain at rest and a fully reduced active range of motion (ROM), particularly in flexion and abduction, were also noted. Radiographs demonstrated moderate degeneration of the glenohumeral joint including a bone cyst of the humeral head. Magnetic resonance imaging (MRI) revealed a massive rotator cuff tear with atrophy of the supraspinatus, infraspinatus, and subscapularis muscles. The T2-weighted MRI images showed that the cyst was in direct contact with the markedly degenerated glenohumeral joint. Based on these findings, the patient was diagnosed with massive rotator cuff tear with ACJ cyst and hemarthrosis of the shoulder. The patient underwent distal clavicle resection and reverse total shoulder arthroplasty (RSA). At 12-month follow-up, the patient showed no pain symptoms, no recurrence of the cyst, and excellent ROM. We experienced a very rare case of ACJ cyst and hemarthrosis of the shoulder occurring simultaneously with rotator cuff tear arthropathy. This report is very valuable in that it suggests that RSA is useful for both ACJ cysts and hemarthrosis of the shoulders associated with rotator cuff tear arthropathy.

RtNet: a deep hybrid neural network for the identification of acute rejection and chronic allograft nephropathy after renal transplantation using multiparametric MRI.

Reliable diagnosis of the cause of renal allograft dysfunction is of clinical importance. The aim of this study is to develop a hybrid deep-learning approach for determining acute rejection (AR), chronic allograft nephropathy (CAN) and renal function in kidney-allografted patients by multimodality integration. Clinical and magnetic resonance imaging (MRI) data of 252 kidney-allografted patients who underwent post-transplantation MRI between December 2014 and November 2019 were retrospectively collected. An end-to-end convolutional neural network, namely RtNet, was designed to discriminate between AR, CAN and stable renal allograft recipient (SR), and secondarily, to predict the impaired renal graft function [estimated glomerular filtration rate (eGFR) ≤50 mL/min/1.73 m2]. Specially, clinical variables and MRI radiomics features were integrated into the RtNet, resulting in a hybrid network (RtNet+). The performance of the conventional radiomics model RtRad, RtNet and RtNet+ was compared to test the effect of multimodality interaction. Out of 252 patients, AR, CAN and SR was diagnosed in 20/252 (7.9%), 92/252 (36.5%) and 140/252 (55.6%) patients, respectively. Of all MRI sequences, T2-weighted imaging and diffusion-weighted imaging with stretched exponential analysis showed better performance than other sequences. On pairwise comparison of resulting prediction models, RtNet+ produced significantly higher macro-area-under-curve (macro-AUC) (0.733 versus 0.745; P = 0.047) than RtNet in discriminating between AR, CAN and SR. RtNet+ performed similarly to the RtNet (macro-AUC, 0.762 versus 0.756; P > 0.05) in discriminating between eGFR ≤50 mL/min/1.73 m2 and >50 mL/min/1.73 m2. With decision curve analysis, adding RtRad and RtNet to clinical variables resulted in more net benefits in diagnostic performance. Our study revealed that the proposed RtNet+ model owned a stable performance in revealing the cause of renal allograft dysfunction, and thus might offer important references for individualized diagnostics and treatment strategy.

Worse Physical Disability Is Associated With the Expression of PD-1 on Inflammatory T-Cells in Multiple Sclerosis Patients With Older Appearing Brains.

Background: Magnetic Resonance Imaging (MRI) analysis method “brain-age” paradigm could offer an intuitive prognostic metric (brain-predicted age difference: brain-PAD) for disability in Multiple Sclerosis (MS), reflecting structural brain health adjusted for aging. Equally, cellular senescence has been reported in MS using T-cell biomarker CD8+CD57+.Objective: Here we explored links between MRI-derived brain-age and blood-derived cellular senescence. We examined the value of combining brain-PAD with CD8+CD57+(ILT2+PD-1+) T-cells when predicting disability score in MS and considered whether age-related biological mechanisms drive disability.Methods: Brain-age analysis was applied to T1-weighted MRI images. Disability was assessed and peripheral blood was examined for CD8+CD57+ T-cell phenotypes. Linear regression models were used, adjusted for sex, age and normalized brain volume.Results: We included 179 mainly relapsing-remitting MS patients. A high brain-PAD was associated with high physical disability (mean brain-PAD = +6.54 [5.12–7.95]). CD8+CD57+(ILT2+PD-1+) T-cell frequency was neither associated with disability nor with brain-PAD. Physical disability was predicted by the interaction between brain-PAD and CD8+CD57+ILT2+PD-1+ T-cell frequency (AR2 = 0.196), yet without improvement compared to brain-PAD alone (AR2 = 0.206; AICc = 1.8).Conclusion: Higher frequency of CD8+CD57+ILT2+PD-1+ T-cells in the peripheral blood in patients with an older appearing brain was associated with worse disability scores, suggesting a role of these cells in the development of disability in MS patients with poorer brain health.

STHarDNet: Swin Transformer with HarDNet for MRI Segmentation

In magnetic resonance imaging (MRI) segmentation, conventional approaches utilize U-Net models with encoder–decoder structures, segmentation models using vision transformers, or models that combine a vision transformer with an encoder–decoder model structure. However, conventional models have large sizes and slow computation speed and, in vision transformer models, the computation amount sharply increases with the image size. To overcome these problems, this paper proposes a model that combines Swin transformer blocks and a lightweight U-Net type model that has an HarDNet blocks-based encoder–decoder structure. To maintain the features of the hierarchical transformer and shifted-windows approach of the Swin transformer model, the Swin transformer is used in the first skip connection layer of the encoder instead of in the encoder–decoder bottleneck. The proposed model, called STHarDNet, was evaluated by separating the anatomical tracings of lesions after stroke (ATLAS) dataset, which comprises 229 T1-weighted MRI images, into training and validation datasets. It achieved Dice, IoU, precision, and recall values of 0.5547, 0.4185, 0.6764, and 0.5286, respectively, which are better than those of the state-of-the-art models U-Net, SegNet, PSPNet, FCHarDNet, TransHarDNet, Swin Transformer, Swin UNet, X-Net, and D-UNet. Thus, STHarDNet improves the accuracy and speed of MRI image-based stroke diagnosis.

Are magnetic resonance imaging findings adequate for differentiating head-and-neck masses as malignant or benign?

Purpose: This work aims to determine the magnetic resonance imaging (MRI) findings of the masses seen in the head-and-neck region and the effectiveness of the MRI in separation of malignant from benign head-and-neck masses through histopathological data. Materials and Methods: This retrospective study was conducted on 86 patients who were referred with prediagnosed as mass in oral, head and neck regions for evaluation with magnetic resonance imaging. MRI images were analyzed according to homogeneity and signal intensity in T1- and T2-weighted images, contrast enhancement pattern, peripheral edge characteristic of the lesion, invasion condition, presence of cystic/necrotic content, and presence of bone involvement of head-and-neck masses. Results: The MRI features of benign and malignant head and neck masses in which signal intensity on T1-weighted MRI images, signal homogeneity on T2-weighted MRI images, contrast enhancement pattern, edge feature and presence of invasion showed a statistically significant difference to determine benignity and malignancy. However, the signal homogeneity in T1-weighted images, signal intensity in T2-weighted images, presence of cystic/necrotic content, and bone involvement did not reveal a statistically significant difference. Conclusions: Overall, our findings were consistent with the literature data. As a result of our study, we recommend that certain MRI features of head-and-neck masses (signal intensity in T1-weighted images, signal homogeneity in T2-weighted images, contrast enhancement pattern, edge feature, and presence of invasion) can be used to differentiate between benign and malignant masses.

Preoperative Templating of Bone-Patellar Tendon-Bone Graft for Anterior Cruciate Ligament Reconstruction: A Morphometry-Based Graft Harvest Method.

Knowledge of anatomy and morphometry of the patella and patellar tendon is crucial for the selection of bone-patellar tendon-bone (BTB) graft for anterior cruciate ligament reconstruction. Graft tunnel mismatch in BTB graft especially in patients with patella alta or baja can result in compromised fixation for the bone-to-bone healing. This complication can be avoided by proper templating of graft using parameters measured from magnetic resonance imaging (MRI). The study aimed to derive morphometric data from MRI and predict the suitability of BTB graft preoperatively. MRI of 1,002 knees was chosen from database after applying the eligibility criteria, which included individuals in the age group of 18-50 years (both sexes) with the intact patella and patellar tendon. Individuals with pathologies of the knee joint and associated structures such as patellar fracture/dislocations, fractures of the distal femur and proximal tibia, and avulsion of the quadriceps tendon or patellar tendon were excluded. For analysis, 1.5 Tesla, proton density, and fat-suppressed sequences of sagittal and axial sections of T2-weighted MRI images were used. Mean age of the 1,002 patients was 35.45 years and there were 290 women and 712 men. Respective measurements were as follows: patella length, width, and thickness, 40.3 mm, 40.2 mm, and 18.6 mm, respectively; patellar tendon length, width, and insertional thickness, 45.2 mm, 27.2 mm, and 5.7 mm, respectively; Insall-Salvati ratio, 1.13; overall graft length, 90.2 mm; and effective tendon length, 26.1 mm. A simple MRI analysis can give us valuable inputs on BTB graft morphometry. The values can also help us with the near-perfect graft harvest. The intraoperative complication of graft tunnel mismatch can be avoided by predicting the overall graft length, effective tendon length, tibial tunnel length, and patellar position using the measured parameters on MRI.

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.

Analysis of the paraspinal muscle morphology of the lumbar spine using a convolutional neural network (CNN)

PurposeThis single-center study aimed to develop a convolutional neural network to segment multiple consecutive axial magnetic resonance imaging (MRI) slices of the lumbar spinal muscles of patients with lower back pain and automatically classify fatty muscle degeneration.MethodsWe developed a fully connected deep convolutional neural network (CNN) with a pre-trained U-Net model trained on a dataset of 3,650 axial T2-weighted MRI images from 100 patients with lower back pain. We included all qualities of MRI; the exclusion criteria were fractures, tumors, infection, or spine implants. The training was performed using k-fold cross-validation (k = 10), and performance was evaluated using the dice similarity coefficient (DSC) and cross-sectional area error (CSA error). For clinical correlation, we used a simplified Goutallier classification (SGC) system with three classes.ResultsThe mean DSC was high for overall muscle (0.91) and muscle tissue segmentation (0.83) but showed deficiencies in fatty tissue segmentation (0.51). The CSA error was small for the overall muscle area of 8.42%, and fatty tissue segmentation showed a high mean CSA error of 40.74%. The SGC classification was correctly predicted in 75% of the patients.ConclusionOur fully connected CNN segmented overall muscle and muscle tissue with high precision and recall, as well as good DSC values. The mean predicted SGC values of all available patient axial slices showed promising results. With an overall Error of 25%, further development is needed for clinical implementation. Larger datasets and training of other model architectures are required to segment fatty tissue more accurately.

No patella resurfacing total knee arthroplasty leads to reduction in the thickness of patellar cartilage to less than half within 5\u2009years: a quantitative longitudinal evaluation using MRI

PurposePatellar resurfacing in total knee arthroplasty (TKA) remains controversial as recent meta-analyses have not shown its clear superiority; however, most authors recommend it because it is associated with less frequent anterior knee pain and need for reoperation. We aimed to clarify the changes in patellar cartilage thickness in no patellar resurfacing TKA using a ceramic femoral component on magnetic resonance imaging (MRI).MethodsBetween 2009 and 2014, 40 consecutive patients (59 knees) were included in this study. All patients underwent TKA using zirconia ceramic femoral implants without patellar resurfacing. Indications for no patellar resurfacing TKA were absence of anterior knee pain, patellar compression pain, and osteoarthritic changes in the patellofemoral joint on plain radiography. The mean postoperative follow-up duration was 81.5 months (range, 25–131 months). Clinical and radiological evaluations were performed preoperatively and 5 years after TKA. Patellar cartilage thickness was evaluated preoperatively and every year for 5 years after TKA using MRI T2-weighted imaging. The patellar cartilage was divided into three regions of interest: medial, central, and lateral. To standardise the variation in patellar thickness among patients, the percent cartilage thickness was calculated.ResultsThe implant’s position was appropriate in all cases. Compared to preoperative scores, 5 years postoperatively, the Japanese Orthopedic Association score and Oxford knee score significantly improved from 52.1 to 84.7; mean tilting angle and congruence angle did not change significantly; mean lateral shift ratio significantly increased from 7.1% to 14.6%; cartilage thickness significantly decreased (P < 0.05); and the percentage cartilage thickness of the central, medial, and lateral cartilage zones gradually thinned to less than half. Four patients underwent conversion to patellar resurfacing due to anterior knee pain, without loosening the femoral and tibial implants.ConclusionThe patellar cartilage thickness decreased to less than half its preoperative level within 5 years after no patellar resurfacing TKA; this would led to clinical problems and conversion to patellar resurfacing.Level of evidenceLevel III.

Cascaded deep learning-based auto-segmentation for head and neck cancer patients: Organs at risk on T2-weighted magnetic resonance imaging.

To investigate multiple deep learning methods for automated segmentation (auto-segmentation) of the parotid glands, submandibular glands, and level II and level III lymph nodes on magnetic resonance imaging (MRI). Outlining radiosensitive organs on images used to assist radiation therapy (radiotherapy) of patients with head and neck cancer (HNC) is a time-consuming task, in which variability between observers may directly impact on patient treatment outcomes. Auto-segmentation on computed tomography imaging has been shown to result in significant time reductions and more consistent outlines of the organs at risk. Three convolutional neural network (CNN)-based auto-segmentation architectures were developed using manual segmentations and T2-weighted MRI images provided from the American Association of Physicists in Medicine (AAPM) radiotherapy MRI auto-contouring (RT-MAC) challenge dataset (n=31). Auto-segmentation performance was evaluated with segmentation similarity and surface distance metrics on the RT-MAC dataset with institutional manual segmentations (n=10). The generalizability of the auto-segmentation methods was assessed on an institutional MRI dataset (n=10). Auto-segmentation performance on the RT-MAC images with institutional segmentations was higher than previously reported MRI methods for the parotid glands (Dice: 0.860 ± 0.067, mean surface distance [MSD]: 1.33 ± 0.40mm) and the first report of MRI performance for submandibular glands (Dice: 0.830 ± 0.032, MSD: 1.16 ± 0.47mm). We demonstrate that high-resolution auto-segmentations with improved geometric accuracy can be generated for the parotid and submandibular glands by cascading a localizer CNN and a cropped high-resolution CNN. Improved MSDs were observed between automatic and manual segmentations of the submandibular glands when a low-resolution auto-segmentation was used as prior knowledge in the second-stage CNN. Reduced auto-segmentation performance was observed on our institutional MRI dataset when trained on external RT-MAC images; only the parotid gland auto-segmentations were considered clinically feasible for manual correction (Dice: 0.775 ± 0.105, MSD: 1.20 ± 0.60mm). This work demonstrates that CNNs are a suitable method to auto-segment the parotid and submandibular glands on MRI images of patients with HNC, and that cascaded CNNs can generate high-resolution segmentations with improved geometric accuracy. Deep learning methods may be suitable for auto-segmentation of the parotid glands on T2-weighted MRI images from different scanners, but further work is required to improve the performance and generalizability of these methods for auto-segmentation of the submandibular glands and lymph nodes.

Greater radiologic evidence of hypothalamic gliosis predicts adiposity gain in children at risk for obesity.

This study investigated, in a large pediatric population, whether magnetic resonance imaging (MRI) evidence of mediobasal hypothalamic (MBH) gliosis is associated with baseline or change over 1 year in body adiposity. Cross-sectional and prospective cohort analyses were conducted within the Adolescent Brain Cognitive Development Study. Study 1 included 169 children with usable baseline T2-weighted MRI images and anthropometrics from baseline and 1-year follow-up study visits. Signal ratios compared T2 signal intensity in MBH and two reference regions (amygdala [AMY] and putamen) as a measure of MBH gliosis. Study 2 included a distinct group of 238 children with overweight or obesity to confirm initial findings in an independent sample. In Study 1, MBH/AMY signal ratio was positively associated with BMI z score (β = 4.27, p < 0.001). A significant interaction for the association of MBH/AMY signal ratio with change in BMI z score suggested that relationships differed by baseline weight status. Study 2 found that higher MBH/AMY signal ratios associated with an increase in BMI z score for children with overweight (β = 0.58, p = 0.01), but not those with obesity (β = 0.02, p = 0.91). Greater evidence of hypothalamic gliosis by MRI is associated with baseline BMI z score and predicts adiposity gain in young children at risk of obesity.

Splenic kaposiform hemangioendothelioma presenting as insidious consumptive coagulopathy

Splenic kaposiform hemangioendothelioma presenting as insidious consumptive coagulopathy

A Prediction Model for Para-Aortic Lymph Node Metastasis in Cervical Cancer by Radiomics Analysis Using Pre-Treatment MRI Images of the Primary Tumor

Whole pelvis irradiation (WPI) is the standard radiation technique for locally advanced cervical cancer without para-aortic lymph node (PALN) metastasis. PALN is one of the most common late failure sites in patients treated with WPI. Although there are several reports about the utility of prophylactic extended-field irradiation (EFI), it is still controversial which patients benefit from prophylactic EFI. If we could predict a PALN metastasis from pre-treatment imaging data, it might help patients to select prophylactic EFI. This study aimed to construct a predictive model for the PALN metastasis in patients with cervical cancer by radiomics analysis using pretreatment magnetic resonance imaging (MRI) of the primary tumor.Data of 94 patients with cervical squamous cell carcinoma who underwent radiotherapy between 2003/10 and 2018/2 were split into two sets: 66 patients for the training/validation and 28 patients for testing. The PALN status was classified into two groups (positive or negative). Both the synchronous and metachronous PALN metastasis was classified as PALN positive. A total number of 9394 radiomics features per a patient image were extracted from T1- and T2-weighted MRI images. The set of candidate predictors were selected with the least absolute shrinkage and selection operator (LASSO) logistic regression and build predictive models with neural network classifiers were used. The precision, accuracy, and sensitivity by generating confusion matrices and the areas under the receiver operating characteristic curve (AUC) for each model were evaluated.By the LASSO analysis of the training/validation data, we found 9 radiomics features from T1-weighted MRI image and 61 radiomics features from T2-weighted MRI image for the classification. The accuracy, specificity, sensitivity, and AUC of the prediction model for the dataset in testing group were 67.9 %, 91.0%, 10.0%, and 0.60 with T1-weighted MRI image, 95.7%, 97.0%, 92.5%, and 0.98 with T2-weighted MRI image, 96.4%, 99.0%, 90.7%, and 0.99 with the combination of T1 and T2-weighted MRI images, respectively.We constructed a model to predict the PALN metastasis in patients with cervical cancer using pre-treatment MRI image-based radiomics and machine learning. The model based on T2-weighted image or combination of T1 and T2-wighted MRI image showed promising prediction accuracy. This model may be useful to select patients who benefit from prophylactic EFI.I. Nishibuchi: None. D. Kawahara: None. M. Kawamura: None. K. Kubo: None. N. Imano: None. Y. Takeuchi: None. A. Saito: None. Y. Murakami: None. Y. Nagata: None.

Generating Synthesized Computed Tomography (CT) from Magnetic Resonance Imaging Using Cycle-Consistent Generative Adversarial Network for Brain Tumor Radiation Therapy

<h3>Purpose/Objective(s)</h3> Brain tumor is the most common malignant tumor of the head and neck in China, postoperative radiotherapy is one of the main methods to improve the survival rate of patients. Magnetic resonance imaging (MRI) has the advantage of its soft tissue contrast, for the sake of getting the tumor boundary accurately, it is necessary to combine CT/MR image fusing in gross tumor volume (GTV) delineation. In order to avoid the uncertainty caused by multimodal images registration and reduce unnecessary radiation dose, replacing CT with MRI has become one of the research hotspots in the field of radiotherapy. The aim of this study was to use unregistered MRI and CT images from brain tumor patients to generate pseudo-CT images based on a cycle-consistent generative adversarial network (CycleGAN) framework. <h3>Materials/Methods</h3> T1, T2-weighted MRI and CT-simulation images of the whole brain were collected from 31 brain tumor patients. In this work, we have used a CycleGAN framework to generate pseudo-CT images from MRI, this model is capable of image-to-image translation using unpaired MRI and CT images in an unsupervised learning method. Due to the influence of head frame in CT-simulation images and the different scanning range, imaging resolution and contrast of each patient, preprocessing of MRI and CT images, such as clipping, background removal, resampling and normalization, was required firstly. Secondly, in order to compensate for the insufficient to separate all major tissue types by the single MR sequence, each patient's T1/T2-weighted MR image pair was served as the input of the CycleGAN framework after registering and fusing. Finally, a simple cross-validation study, randomly selecting 70% samples as the training set and the remaining images as the testing set, was performed to compare the quality of synthetic CT and real CT image. <h3>Results</h3> The CycleGAN method produced the overall average MAE below 0.24 ± 0.02 and the 0.79 ± 0.03 SSIM value for the testing set images, the heterogeneity between the synthetic CT image and the actual CT image was acceptable. <h3>Conclusion</h3> We successfully realized the image-to-image translation using unregistered MRI and CT images based on the CycleGAN framework. The evaluation of pseudo-CT image showed the feasibility and accuracy of this method, which can effectively reduce the error caused by multi-mode image registration in gross tumor volume (GTV) delineation for brain tumor patients.

Spinal Cord Signal Change on Magnetic Resonance Imaging May Predict Worse Clinical In- and Outpatient Outcomes in Patients with Spinal Cord Injury: A Prospective Multicenter Study in 459 Patients.

Prognostic factors for clinical outcome after spinal cord (SC) injury (SCI) are limited but important in patient management and education. There is a lack of evidence regarding magnetic resonance imaging (MRI) and clinical outcomes in SCI patients. Therefore, we aimed to investigate whether baseline MRI features predicted the clinical course of the disease. This study is an ancillary to the prospective North American Clinical Trials Network (NACTN) registry. Patients were enrolled from 2005–2017. MRI within 72 h of injury and a minimum follow-up of one year were available for 459 patients. Patients with American Spinal Injury Association impairment scale (AIS) E were excluded. Patients were grouped into those with (n = 354) versus without (n = 105) SC signal change on MRI T2-weighted images. Logistic regression analysis adjusted for commonly known a priori confounders (age and baseline AIS). Main outcomes and measures: The primary outcome was any adverse event. Secondary outcomes were AIS at the baseline and final follow-up, length of hospital stay (LOS), and mortality. A regression model adjusted for age and baseline AIS. Patients with intrinsic SC signal change were younger (46.0 (interquartile range (IQR) 29.0 vs. 50.0 (IQR 20.5) years, p = 0.039). There were no significant differences in the other baseline variables, gender, body mass index, comorbidities, and injury location. There were more adverse events in patients with SC signal change (230 (65.0%) vs. 47 (44.8%), p < 0.001; odds ratio (OR) = 2.09 (95% confidence interval (CI) 1.31–3.35), p = 0.002). The most common adverse event was cardiopulmonary (186 (40.5%)). Patients were less likely to be in the AIS D category with SC signal change at baseline (OR = 0.45 (95% CI 0.28–0.72), p = 0.001) and in the AIS D or E category at the final follow-up (OR = 0.36 (95% CI 0.16–0.82), p = 0.015). The length of stay was longer in patients with SC signal change (13.0 (IQR 17.0) vs. 11.0 (IQR 14.0), p = 0.049). There was no difference between the groups in mortality (11 (3.2%) vs. 4 (3.9%)). MRI SC signal change may predict adverse events and overall LOS in the SCI population. If present, patients are more likely to have a worse baseline clinical presentation (i.e., AIS) and in- or outpatient clinical outcome after one year. Patients with SC signal change may benefit from earlier, more aggressive treatment strategies and need to be educated about an unfavorable prognosis.

High-Definition Transcranial Direct Current Stimulation Improves Delayed Memory in Alzheimer's Disease Patients: A Pilot Study Using Computational Modeling to Optimize Electrode Position.

The optimal stimulation parameters when using transcranial direct current stimulation (tDCS) to improve memory performance in patients with Alzheimer's disease (AD) are lacking. In healthy individuals, inter-individual differences in brain anatomy significantly influence current distribution during tDCS, an effect that might be aggravated by variations in cortical atrophy in AD patients. To measure the effect of individualized HD-tDCS in AD patients. Nineteen AD patients were randomly assigned to receive active or sham high-definition tDCS (HD-tDCS). Computational modeling of the HD-tDCS-induced electric field in each patient's brain was analyzed based on magnetic resonance imaging (MRI) scans. The chosen montage provided the highest net anodal electric field in the left dorsolateral prefrontal cortex (DLPFC). An accelerated HD-tDCS design was conducted (2 mA for 3×20 min) on two separate days. Pre- and post-intervention cognitive tests and T1 and T2-weighted MRI and diffusion tensor imaging data at baseline were analyzed. Different montages were optimal for individual patients. The active HD-tDCS group improved significantly in delayed memory and MMSE performance compared to the sham group. Five participants in the active group had higher scores on delayed memory post HD-tDCS, four remained stable and one declined. The active HD-tDCS group had a significant positive correlation between fractional anisotropy in the anterior thalamic radiation and delayed memory score. HD-tDCS significantly improved delayed memory in AD. Our study can be regarded as a proof-of-concept attempt to increase tDCS efficacy. The present findings should be confirmed in larger samples.

Influence of Lumbar Spinal Canal Dimensions on Neurological Claudication Symptomatology - A Case Control Study

Lumbar spinal stenosis (LSS) is abnormal narrowing of spinal canal of lumbar vertebrae causing compression of neural tissue which leads to a neurologic deficit. LSS is diagnosed based on clinical symptoms of patient and it is confirmed using MRI (Magnetic Resonance Imaging). The aim of the study is to determine the association between MRI lumbar spinal canal and foraminal dimensions in on neurological claudication symptomology. A total of 89 patients (45 males ,44 females, age range: 20-60 years) referred for MRI Lumbar spine were included in this prospective study. Patients were categorized into symptomatic (49 cases) and asymptomatic (40 controls) based on Edinburgh claudication questionnaire (ECQ). The Antero-posterior diameter of dural sac, Transverse diameter of dural sac, Cross sectional area of dural sac, Cross sectional area of lateral recess, Lateral recess depth, Angle of lateral recess, Ligamentous interfacet distance was measured from L3 to S1 on Axial T2 weighted MRI images. The strength of association between MRI measurements and clinical symptoms were assessed using logistic regression analysis and Cramer’s V test. The Phi coefficient value for the lumbar spinal canal, foraminal dimensions with neurogenic claudication symptomology based on ECQ was 0.108 &amp; 0.207 respectively, which showed weak positive correlation. Our study concludes that MRI measurements of lumbar spinal canal and foraminal dimensions denoting LSS were found to show weak positive correlation with clinical symptoms assessed based on Edinburgh claudication questionnaire.

Absence of T1 Hyperintensity in the Brain of Thalassemia Patients after Multiple Administrations of Gadobutrol

Gadolinium based contrast agents (GBCA) are used in magnetic resonance imaging (MRI) in order to improve the detection and the characterization of pathophysiologic processes. Many studies have reported an association between increased signal intensities (SI) on unenhanced T1-weighted MRI images in different brain regions and the history of repeated intravenous administrations of GBCAs in patients with normal renal function. We conducted a prospective study aimed to evaluate signal changes in the dentate nucleus (DN), globus pallidus (GP), pons, and thalamus (normalized to the deep cerebellum white matter) in T1-MRI images after serial injections of Gadobutrol for the study of the heart in patients with thalassemia.Among the thalassemic patients who underwent a brain MRI within a pilot study for iron overload assessment, we selected two groups of patients: 15 patients transfused and chelated with≥4 Gadobutrol administrations at a high dose (0.2 mmol/Kg per scan) and 8 thalassemia patients who never received Gadobutrol. Moreover, we included 13 healthy subjects who never received Gadobutrol as control group. Iron overload was assessed by the T2* technique.Signal intensity (SI) ratios in all regions were comparable among the three groups (see Table) and were not correlated with the number of Gadobutol administrations. No correlation was detected between SI ratios and T2* values.It is important to underline that thalassemia patients represent an high-risk population for the GBCA accumulation, due to the fact that, in presence of siderosis, ferric ions can compete with Gadolinium ions for the ligand, destabilizing the gadolinium complex. Nevertheless, our study describes the lack of increased SI after repeated administration of Gadobutrol, a nonionic macrocyclic agent. All our patients were chelated when they received the GBCA. Different studies showed the ability of all three iron chelators to chelate other metals. So, a potential role of chelation therapy may be supposed. [Display omitted] DisclosuresPepe:Chiesi Farmaceutici and ApoPharma Inc.: Other: Alessia Pepe is the PI of the MIOT project, that receives no profit support from Chiesi Farmaceutici S.p.A. and ApoPharma Inc.