T2-weighted MR Research Articles

MRI-based prostate cancer classification using 3D efficient capsule network.

Prostate cancer (PCa) is the most common cancer in men and the second leading cause of male cancer-related death. Gleason score (GS) is the primary driver of PCa risk-stratification and medical decision-making, but can only be assessed at present via biopsy under anesthesia. Magnetic resonance imaging (MRI) is a promising non-invasive method to further characterize PCa, providing additional anatomical and functional information. Meanwhile, the diagnostic power of MRI is limited by qualitative or, at best, semi-quantitative interpretation criteria, leading to inter-reader variability. Computer-aided diagnosis employing quantitative MRI analysis has yielded promising results in non-invasive prediction of GS. However, convolutional neural networks (CNNs) do not implicitly impose a frame of reference to the objects. Thus, CNNs do not encode the positional information properly, limiting method robustness against simple image variations such as flipping, scaling, or rotation. Capsule network (CapsNet) has been proposed to address this limitation and achieves promising results in this domain. In this study, we develop a 3D Efficient CapsNet to stratify GS-derived PCa risk using T2-weighted (T2W) MRI images. In our method, we used 3D CNN modules to extract spatial features and primary capsule layers to encode vector features. We then propose to integrate fully-connected capsule layers (FC Caps) to create a deeper hierarchy for PCa grading prediction. FC Caps comprises a secondary capsule layer which routes active primary capsules and a final capsule layer which outputs PCa risk. To account for data imbalance, we propose a novel dynamic weighted margin loss. We evaluate our method on a public PCa T2W MRI dataset from the Cancer Imaging Archive containing data from 976 patients. Two groups of experiments were performed: (1) we first identified high-risk disease by classifying low + medium risk versus high risk; (2) we then stratified disease in one-versus-one fashion: low versus high risk, medium versus high risk, and low versus medium risk. Five-fold cross validation was performed. Our model achieved an area under receiver operating characteristic curve (AUC) of 0.83 and 0.64 F1-score for low versus high grade, 0.79 AUC and 0.75 F1-score for low + medium versus high grade, 0.75 AUC and 0.69 F1-score for medium versus high grade and 0.59 AUC and 0.57 F1-score for low versus medium grade. Our method outperformed state-of-the-art radiomics-based classification and deep learning methods with the highest metrics for each experiment. Our divide-and-conquer strategy achieved weighted Cohen's Kappa score of 0.41, suggesting moderate agreement with ground truth PCa risks. In this study, we proposed a novel 3D Efficient CapsNet for PCa risk stratification and demonstrated its feasibility. This developed tool provided a non-invasive approach to assess PCa risk from T2W MR images, which might have potential to personalize the treatment of PCa and reduce the number of unnecessary biopsies.

Comprehensive deep learning-based framework for automatic organs-at-risk segmentation in head-and-neck and pelvis for MR-guided radiation therapy planning

Introduction: The excellent soft-tissue contrast of magnetic resonance imaging (MRI) is appealing for delineation of organs-at-risk (OARs) as it is required for radiation therapy planning (RTP). In the last decade there has been an increasing interest in using deep-learning (DL) techniques to shorten the labor-intensive manual work and increase reproducibility. This paper focuses on the automatic segmentation of 27 head-and-neck and 10 male pelvis OARs with deep-learning methods based on T2-weighted MR images.Method: The proposed method uses 2D U-Nets for localization and 3D U-Net for segmentation of the various structures. The models were trained using public and private datasets and evaluated on private datasets only.Results and discussion: Evaluation with ground-truth contours demonstrated that the proposed method can accurately segment the majority of OARs and indicated similar or superior performance to state-of-the-art models. Furthermore, the auto-contours were visually rated by clinicians using Likert score and on average, 81% of them was found clinically acceptable.

PP10 Presentation Time: 11:00 AM: Assessment of NASHA Spacer Symmetry for Prostate Radiation Therapy

<h3>Purpose</h3> NASHA (non-animal stabilized hyaluronic acid) is a spacer that can increase the separation between the prostate and anterior rectal wall prior to prostate radiation therapy (RT). Since there is no pre-defined time limit for NASHA spacer injection, physicians can precisely distribute the product within the peri-rectal space. Furthermore, the spacer is highly visible on T2-weighted MR, and generally visible on CT with optimized view settings. We conducted a randomized controlled trial evaluating whether a NASHA spacer improves rectal dosimetry among patients receiving hypofractionated RT (NCT 04189913). The purpose of the study is to evaluate the symmetry of a NASHA spacer among all patients who received the device on that study. <h3>Materials and Methods</h3> We analyzed data for 131 of 136 patients, who were randomized to and received a NASHA spacer on trial (5 datasets were unavailable at the time of analysis). All patients had pre-implant and post-implant CT/MR simulation. We developed an automated system for evaluating symmetry based on a previously published method (Fischer-Valuck, PRO, 2017). Briefly, prostate, rectum and spacer contours that had been prospectively validated by a core laboratory were downloaded. From the prostate center-of-mass, we defined axial slices at mid-gland, 1-cm superior, and 1-cm inferior. For each axial slice, we calculated the center of the spacer (spacer width center) along the transverse axis. Axial slice symmetry scores were assigned based on the distance between the spacer width center and the prostate midline (1 = distance within 1 cm of midline (symmetric), 2L/2R = distance between 1-2 cm from midline, 3L/3R = distance more than 2 cm from midline). Patient-level symmetry scores were assigned based on the number of slices with an asymmetric distribution (1 = none, 2 = 1 of 3 slices with axial score of 2L/2R, 3 = 2 of 3 slices with axial score of 2L/2R, 4 = all 3 slices with axial score of 2L/2R). <h3>Results</h3> The percentages of patients with an axial slice symmetry score of 1 (symmetric) at mid-gland, superior, and inferior planes, were 97%, 98%, and 98%, respectively (see Figure). Following the Fischer rubric, planes without gel were omitted (N=2 mid-gland, N=2 superior and N=9 inferior planes). The percentage of patients with patient-level symmetry scores of 1, 2, 3 and 4, were 95.4%, 2.3%, 1.5%, and 0.8% respectively. <h3>Conclusions</h3> Over 95% of NASHA spacers placed in this prospective trial demonstrated patient-level symmetry. These results suggest that NASHA may also improve rectal dosimetry for prostate brachytherapy. Further study is ongoing to correlate the dosimetry and clinical outcomes to implant quality for these patients.

Deep learning-based classification of healthy aging controls, mild cognitive impairment and Alzheimer’s disease using fusion of MRI-PET imaging

Automated detection of dementia stage using multimodal imaging modalities will be helpful for improving the clinical diagnosis. In this study, we develop the Inception-ResNet wrapper model in differentiating the healthy controls (HC), mild cognitive impairment (MCI), and Alzheimer’s disease (AD) using conjoint magnetic resonance imaging (MRI) and positron emission tomography (PET) scans. We use T1-weighted MR and PET images of individuals aged between 42 and 95 years, including HC, MCI and AD patients. We first perform 3D tissue segmentation of MR images after skull striping. The atlas-based segmented MR image tissue is fused with PET image. Then we transform PET images from RGB to HSI color space and apply fusion of MRI with PET images using two-dimensional Fourier and discrete wavelet transform (DWT) and then reconstruct the MR-PET fused image using inverse Fourier and DWT methods. After the fusion of MRI and PET imaging modalities, we used 60 % training, 20 % for validation and the remaining 20 % as a test set using various convolutional neural networks. We found the proposed model as the best classifier with an accuracy of 95.5 %, 94.1 % and 95.9 % in classifying HC vs MCI, MCI vs AD and AD vs HC respectively when compared to the existing methods. We conclude that the proposed deep learning model has potential in automated classification of healthy and dementia stages using combined MRI and PET modalities with good performance.

Predicting IDH subtype of grade 4 astrocytoma and glioblastoma from tumor radiomic patterns extracted from multiparametric magnetic resonance images using a machine learning approach.

Background and purposeSemantic imaging features have been used for molecular subclassification of high-grade gliomas. Radiomics-based prediction of molecular subgroups has the potential to strategize and individualize therapy. Using MRI texture features, we propose to distinguish between IDH wild type and IDH mutant type high grade gliomas.MethodsBetween 2013 and 2020, 100 patients were retrospectively analyzed for the radiomics study. Immunohistochemistry of the pathological specimen was used to initially identify patients for the IDH mutant/wild phenotype and was then confirmed by Sanger’s sequencing. Image texture analysis was performed on contrast-enhanced T1 (T1C) and T2 weighted (T2W) MR images. Manual segmentation was performed on MR image slices followed by single-slice multiple sampling image augmentation. Both whole tumor multislice segmentation and single-slice multiple sampling approaches were used to arrive at the best model. Radiomic features were extracted, which included first-order features, second-order (GLCM—Grey level co-occurrence matrix), and shape features. Feature enrichment was done using LASSO (Least Absolute Shrinkage and Selection Operator) regression, followed by radiomic classification using Support Vector Machine (SVM) and a 10-fold cross-validation strategy for model development. The area under the Receiver Operator Characteristic (ROC) curve and predictive accuracy were used as diagnostic metrics to evaluate the model to classify IDH mutant and wild-type subgroups.ResultsMultislice analysis resulted in a better model compared to the single-slice multiple-sampling approach. A total of 164 MR-based texture features were extracted, out of which LASSO regression identified 14 distinctive GLCM features for the endpoint, which were used for further model development. The best model was achieved by using combined T1C and T2W MR images using a Quadratic Support Vector Machine Classifier and a 10-fold internal cross-validation approach, which demonstrated a predictive accuracy of 89% with an AUC of 0.89 for each IDH mutant and IDH wild subgroup.ConclusionA machine learning classifier of radiomic features extracted from multiparametric MRI images (T1C and T2w) provides important diagnostic information for the non-invasive prediction of the IDH mutant or wild-type phenotype of high-grade gliomas and may have potential use in either escalating or de-escalating adjuvant therapy for gliomas or for using targeted agents in the future.

Free-breathing 3D Stack of Stars GRE (StarVIBE) sequence for detecting pulmonary nodules in 18F-FDG PET/MRI

BackgroundThe free-breathing T1-weighted 3D Stack of Stars GRE (StarVIBE) MR sequence potentially reduces artifacts in chest MRI. The purpose of this study was to evaluate StarVIBE for the detection of pulmonary nodules in 18F-FDG PET/MRI.Material and methodsIn this retrospective analysis, conducted on a prospective clinical trial cohort, 88 consecutive women with newly diagnosed breast cancer underwent both contrast-enhanced whole-body 18F-FDG PET/MRI and computed tomography (CT). Patients’ chests were examined on CT as well as on StarVIBE and conventional T1-weighted VIBE and T2-weighted HASTE MR sequences, with CT serving as the reference standard. Presence, size, and location of all detectable lung nodules were assessed. Wilcoxon test was applied to compare nodule features and Pearson’s, and Spearman’s correlation coefficients were calculated.ResultsOut of 65 lung nodules detected in 36 patients with CT (3.7 ± 1.4 mm), StarVIBE was able to detect 31 (47.7%), VIBE 26 (40%) and HASTE 11 (16.8%), respectively. Overall, CT showed a significantly higher detectability than all MRI sequences combined (65 vs. 36, difference 44.6%, p < 0.001). The VIBE showed a significantly better detection rate than the HASTE (23.1%, p = 0.001). Detection rates between StarVIBE and VIBE did not significantly differ (7.7%, p = 0.27), but the StarVIBE showed a significant advantage detecting centrally located pulmonary nodules (66.7% vs. 16.7%, p = 0.031). There was a strong correlation in nodule size between CT and MRI sequences (HASTE: ρ = 0.80, p = 0.003; VIBE: ρ = 0.77, p < 0.001; StarVIBE: ρ = 0.78, p < 0.001). Mean image quality was rated as good to excellent for CT and MRI sequences.ConclusionThe overall lung nodule detection rate of StarVIBE was slightly, but not significantly, higher than conventional T1w VIBE and significantly higher than T2w HASTE. Detectability of centrally located nodules is better with StarVIBE than with VIBE. Nevertheless, all MRI analyses demonstrated considerably lower detection rates for small lung nodules, when compared to CT.

MRI-based peritumoral radiomics analysis for preoperative prediction of lymph node metastasis in early-stage cervical cancer: A multi-center study

PurposeTo evaluate intra- and preitumoral radiomics on the contrast-enhanced T1-weighted (CE-T1) and T2-weighted (T2W) MRI for predicting the LNM, and develop a nomogram for potential clinical uses. MethodsWe enrolled 169 cervical cancer cases who underwent CE-T1 and T2W MR scans from two hospitals between Dec. 2015 and Sep. 2021. Intra- and peritumoral features were extracted separately and selected by the least absolute shrinkage and selection operator (LASSO) regression. Radiomics signatures were built using the selected features from different regions. Clinical parameters were evaluated by statistical analysis. The nomogram was developed combining the multi-regional radiomics signature and the most predictive clinical parameters. ResultsFive radiomics features were finally selected from the peritumoral regions with 1 and 3 mm distances in the CE-T1 and T2W MRI, respectively. The nomogram incorporating multi-regional combined radiomics signature, MR-reported LN status and tumor diameter achieved the highest AUCs in the training (nomogram vs. combined radiomics signature vs. clinical model, 0.891 vs. 0.830 vs. 0.812), internal validation (nomogram vs. combined radiomics signature vs. clinical model, 0.863 vs. 0.853 vs. 0.816) and external validation (nomogram vs. combined radiomics signature vs. clinical model, 0.804 vs. 0.701 vs. 0.787) cohort. DCA suggested good clinical usefulness of our developed models. ConclusionThe current work suggested clinical potential for intra- and peritumoral radiomics with multi-modal MRI for preoperative predicting LNM.

Dedicated Axillary MRI-Based Radiomics Analysis for the Prediction of Axillary Lymph Node Metastasis in Breast Cancer.

Simple SummaryThe presence of axillary lymph node metastases in breast cancer patients is an essential factor in axillary surgery and possible additional treatment. This study aimed to investigate the potential of dedicated axillary MRI-based radiomics analysis for the prediction of axillary lymph node metastases. Dedicated axillary MRI examinations provide a very specific and complete field of view of the axilla. Accurate preoperative prediction of axillary lymph node metastases in breast cancer patients using radiomics analysis can aid in clinical decision-making for the type of treatment.Radiomics features may contribute to increased diagnostic performance of MRI in the prediction of axillary lymph node metastasis. The objective of the study was to predict preoperative axillary lymph node metastasis in breast cancer using clinical models and radiomics models based on T2-weighted (T2W) dedicated axillary MRI features with node-by-node analysis. From August 2012 until October 2014, all women who had undergone dedicated axillary 3.0T T2W MRI, followed by axillary surgery, were retrospectively identified, and available clinical data were collected. All axillary lymph nodes were manually delineated on the T2W MR images, and quantitative radiomics features were extracted from the delineated regions. Data were partitioned patient-wise to train 100 models using different splits for the training and validation cohorts to account for multiple lymph nodes per patient and class imbalance. Features were selected in the training cohorts using recursive feature elimination with repeated 5-fold cross-validation, followed by the development of random forest models. The performance of the models was assessed using the area under the curve (AUC). A total of 75 women (median age, 61 years; interquartile range, 51–68 years) with 511 axillary lymph nodes were included. On final pathology, 36 (7%) of the lymph nodes had metastasis. A total of 105 original radiomics features were extracted from the T2W MR images. Each cohort split resulted in a different number of lymph nodes in the training cohorts and a different set of selected features. Performance of the 100 clinical and radiomics models showed a wide range of AUC values between 0.41–0.74 and 0.48–0.89 in the training cohorts, respectively, and between 0.30–0.98 and 0.37–0.99 in the validation cohorts, respectively. With these results, it was not possible to obtain a final prediction model. Clinical characteristics and dedicated axillary MRI-based radiomics with node-by-node analysis did not contribute to the prediction of axillary lymph node metastasis in breast cancer based on data where variations in acquisition and reconstruction parameters were not addressed.

Prostate lesion segmentation in MR images using radiomics based deeply supervised U-Net

Prostate lesion detection in an axial T2 weighted (T2W) MR images is a very challenging task due to heterogeneous and inconsistent pixel representation surrounding the prostate boundary. In this paper, a radiomics based deeply supervised U-Net is proposed for both prostate gland and prostate lesion segmentation. The proposed pipeline is trained and validated on 1174 and 2071 T2W MR images of 40 patients and tested on 250 and 415 T2W MR images of 10 patients for prostate capsule segmentation and prostate lesion segmentation, respectively. Effective segmentation of prostate lesions in various stages of prostate cancer (namely T1, T2, T3, and T4) is achieved using the proposed framework. The mean Dice Similarity Coefficient (DSC) for actual prostate capsule segmentation and prostate lesion segmentation is 0.8958 and 0.9176, respectively. The proposed framework is also tested on Promise12 public dataset for performance analysis in segmenting prostate gland. The segmentation results using proposed architecture are promising compared to state-of-the-art techniques. It also improves the accuracy of the prostate cancer diagnosis.

Extracting and Selecting Robust Radiomic Features from PET/MR Images in Nasopharyngeal Carcinoma.

This work aims to study the variation, robustness, and feature redundancy of PET/MR radiomic features in the primary tumor of nasopharyngeal carcinoma (NPC). PET/MR scans of 21 NPC patients were used in this study. The primary tumor volumes were defined using PET, T2-weighted-MR (T2-MR), and diffusion-weighted MR (DW-MR) images. A random-dilation-erosion method was used to simulate 10 sets of tumor volumes for identifying features invariant with manual segmentation uncertainties. Feature robustness was evaluated against imaging modalities, pixel sizes, slice thickness, and grey-level bin sizes using intraclass correlation coefficient (ICC) and spearman correlation coefficient. Feature redundancy was analyzed using the hierarchical cluster analysis. Voxel size of 0.5×0.5×1.0mm3 was found optimal for robust feature extraction from PET and MR. Normalized grey level of 64 and 128 was suggested for PET and MR, respectively. The features from wavelet-transformed images were less stable than those from the original images. The robustness analysis and volume correlation analysis identified 335 (62.04%) PET features, 240 (44.44%) T2-MR features, and 366 (67.78%) DW-MR features. The cluster analysis grouped PET, T2-MR, and DW-MR features into 106, 83, and 133 representative features, respectively. The present study analyzed and identified robust features extracted from tumor volumes on PET/MR, which can provide guidance and promote standardization for PET/MR radiomic studies in NPC.

A sum-modified-Laplacian and sparse representation based multimodal medical image fusion in Laplacian pyramid domain.

Fusion of multimodal medical images provides complementary information for diagnosis, surgical planning, and clinical outcome evaluation. Although the multiscale decomposition-based fusion methods have attracted much attention among researchers, the challenges of determining the decomposition levels and the loss of contrast hindered their applications. Here, we present a multimodal medical images fusion method combining the sum-modified-Laplacian (SML) with sparse representation (SR) in the Laplacian pyramid domain. In this method, we first transformed the original images into the high-pass and low-pass bands by the Laplacian pyramid (LP). Then, we use SML and SR to fuse the high- and low-pass bands, respectively. The proposed method has been compared with different methods including NSST_VGG_MAX, DWT_ARV_BURTS, CVT_MAX_LIS, and NSCT_SR_MAX. We also conducted multiple experiments on four groups of medical images, including CT and MR, T1-weighted MR and T2-weighted MR, PET and MR, as well as SPECT and MR, to demonstrate the advantages of our method. Visual and quantitative results illustrate that our method can produce the fused images with better brightness contrast and retain more image details than other evaluated methods on the basis of MI, LAB/F, QAB/F, and Qw. Furthermore, our method could preserve more fine and useful functional information with better image contrast, which is highly relevant in the assessment of lesion shapes and positions. Graphical abstract The basic framework of the proposed fusion method based on sum-modified-Laplacian and sparse representation based in Laplacian pyramid domain using CT and MR images. In the proposed method, the source images are transformed into the low-pass bands and the high-pass bands using the Laplacian pyramid (LP). The low-pass bands are fused by SR, while SML is used to fuse the high-pass bands. Visual and quantitative results show that the proposed method can produce fused images with better brightness contrast and retain more image details than the other compared methods.

RETROSPECTIVE COGNITIVE DECLINE IS ASSOCIATED WITH NEUROFIBRILLARY TANGLES AND BETA-AMYLOID PLAQUES MEASURED BY [F-18]MK-6240 AND [C-11]PIB IN LATE-MIDDLE AGED PERSONS CLINICALLY UNIMPAIRED AT BASELINE

Characterizing contributions of neurofibrillary tangles (NFTs) and beta-amyloid (AB) plaques to cognitive decline, and relationships with other biomarkers during the preclinical phase of Alzheimer's disease (AD) may improve diagnostic and clinical trial outcomes. This work investigated if NFTs and AB plaques measured by PET explained differences in retrospective longitudinal cognitive trajectories of persons recruited from the Wisconsin Registry for Alzheimer's Prevention (WRAP) who were clinically unimpaired at their baseline cognitive assessment. Individuals (N=159; 59±6 years at baseline composite) recruited from WRAP underwent longitudinal cognitive assessments and T1-weighted MR, MK-6240 and PiB PET imaging. A preclinical Alzheimer's cognitive composite (PACC-3: RAVLT Total, WMS-R Logical Memory Delayed Recall, and WAIS-R Digit Symbol) was calculated (mean of Z-scored tests) for each assessment. Four biomarker groups (A+/-, T+/-) were established based on PiB (Global DVR >1.19; Reference Logan, cerebellum GM) and MK-6240 (Entorhinal cortex SUVR>1.27, mean+2SD of PIB(-) group; cerebellum GM, 70–90 min) PET. Differences in PACC-3 trajectories by group were interrogated using a linear mixed effects model with random person-level slopes and intercepts (model in Figure 1). Group differences in demographics, select neuropsychological tests, and biomarkers (PET, hippocampal volume, systolic blood pressure) were also investigated. The linear model indicated a significant group×time interaction (Figure 1); post-hoc pairwise comparisons (Tukey-adjusted) indicated the A+/T+ group had steeper PACC-3 decline compared to all other groups, but other groups did not differ significantly. Groups did not differ by sex, race, or family history of dementia, but did differ in the frequency of APOE-e4 carriers (Table 1). Systolic blood pressure and hippocampal volume did not differ by group at baseline or last assessments. RAVLT Total, WMS-R Logical Memory Delayed Recall, self-reported severity of memory problems, PACC-3 and the frequency of MCI differed by group at last assessment, but MMSE, self- and informant-reported memory problems, and WAIS-R Digit Symbol did not (Table 2).

Diagnosis of Alzheimer's Disease via Multi-Modality 3D Convolutional Neural Network.

Alzheimer’s disease (AD) is one of the most common neurodegenerative diseases. In the last decade, studies on AD diagnosis has attached great significance to artificial intelligence-based diagnostic algorithms. Among the diverse modalities of imaging data, T1-weighted MR and FDG-PET are widely used for this task. In this paper, we propose a convolutional neural network (CNN) to integrate all the multi-modality information included in both T1-MR and FDG-PET images of the hippocampal area, for the diagnosis of AD. Different from the traditional machine learning algorithms, this method does not require manually extracted features, instead, it utilizes 3D image-processing CNNs to learn features for the diagnosis or prognosis of AD. To test the performance of the proposed network, we trained the classifier with paired T1-MR and FDG-PET images in the ADNI datasets, including 731 cognitively unimpaired (labeled as CN) subjects, 647 subjects with AD, 441 subjects with stable mild cognitive impairment (sMCI) and 326 subjects with progressive mild cognitive impairment (pMCI). We obtained higher accuracies of 90.10% for CN vs. AD task, 87.46% for CN vs. pMCI task, and 76.90% for sMCI vs. pMCI task. The proposed framework yields a state-of-the-art performance. Finally, the results have demonstrated that (1) segmentation is not a prerequisite when using a CNN for the classification, (2) the combination of two modality imaging data generates better results.

A novel Tungsten-based fiducial marker for multi-modal brain imaging

BackgroundMulti-modal brain image registration is a prerequisite for accurate mapping of brain structure and function in neuroscience. Image registration is commonly performed using automated software; however, its accuracy decreases when images differ in modality, contrast, uniformity, and resolution. This limitation could be overcome by using an external reference point; however, high-contrast agents in multi-modal imaging have not been previously reported. New methodsHere, we propose a novel multi-modal fiducial marker that contains Tungsten solution and provides high contrast in magnetic resonance imaging (MRI), computed tomography (CT), and positron emission tomography (PET). The basic characteristics of this multi-modal marker were investigated by assessing major sources of image contrast in the following modalities: density and T1-, T2-relaxivity in comparison with conventional contrast agents. ResultsTungsten solution had lower T1- and T2-relaxivity and high solubility, and showed high contrast in T1- and T2-weighted MR and CT images at a high-density concentration (˜3.0 g/mL), whereas other conventional solutions did not show sufficient contrast in either CT or MRI. Comparison with existing methodsThe use of this Tungsten-based multi-modal marker allowed more accurate registration than a software-only method in phantom and animal experiments. Application of this method demonstrated accurate cortical surface mapping of neurotransmitter function (dopamine transporter, DAT) using PET and MRI, and provided a neurobiologically relevant cortical distribution consistent with previous literature on histology-based DAT immunoreactivity. ConclusionsThe Tungsten-based multi-modal fiducial marker is non-radioactive, easy to handle, and aids precise registration across different modalities of brain imaging.

Early Detection of Central Nervous System Relapse of Pediatric Leukemia with Measurement of Optic Nerve Sheath Diameter on MRI

Leukemia is the most common pediatric malignancy. Central Nervous System (CNS) is the most frequently involved extramedullary location at diagnosis and at relapse. To determine if Magnetic Resonance Imaging (MRI) findings of optic nerves should contribute to early detection of CNS relapse in pediatric leukemia. Twenty patients (10 boys, 10 girls; mean age 8,3 years, range 4-16 years) with proven CNS relapse of leukemia followed up between 2009 and 2017 in our institution were included. Orbital MRI exams performed before and during CNS relapse were reviewed retrospectively. Forty optic nerves with Optic Nerve Sheaths (ONS) and Optic Nerve Heads (ONH) were evaluated on fat-suppressed T2-weighted TSE axial MR images. ONS diameter was measured from the point 10 mm posterior to the globe. ONS distension and ONH configuration were graded as 0, 1 and 2. Before CNS relapse, right mean ONS diameter was 4.52 mm and left was 4.61 mm which were 5.68 mm and 5.66 mm respectively during CNS relapse showing a mean increase of 25% on right and 22% on left. During CNS relapse, ONS showed grade 0 distension in 15%, grade 1 in 60%, grade 2 in 25% and ONH demonstrated grade 0 configuration in 70%, grade 1 in 25% and grade 2 in 5% of the patients. MRI findings of optic nerves may contribute to diagnose CNS relapse by demonstrating elevated intracranial pressure in children with leukemia.

Multi-observer contouring of male pelvic anatomy: Highly variable agreement across conventional and emerging structures of interest.

This study quantified inter-observer contouring variations for multiple male pelvic structures, many of which are of emerging relevance for prostate cancer radiotherapy progression and toxicity response studies. Five prostate cancer patient datasets (CT and T2-weighted MR) were distributed to 13 observers for contouring. CT structures contoured included the clinical target volume (CTV), seminal vesicles, rectum, colon, bowel bag, bladder and peri-rectal space (PRS). MR contours included CTV, trigone, membranous urethra, penile bulb, neurovascular bundle and multiple pelvic floor muscles. Contouring variations were assessed using the intraclass correlation coefficient (ICC), Dice similarity coefficient (DSC), and multiple additional metrics. Clinical target volume (CT and MR), bladder, rectum and PRS contours showed excellent inter-observer agreement (median ICC=0.97; 0.99; 1.00; 0.95; 0.90, DSC=0.83±0.05; 0.88±0.05; 0.93±0.03; 0.81±0.07; 0.80±0.06, respectively). Seminal vesicle contours were more variable (ICC=0.75, DSC=0.73±0.14), while colon and bowel bag contoured volumes were consistent (ICC=0.97; 0.97), but displayed poor overlap (DSC=0.58±0.22; 0.67±0.21). Smaller MR structures showed significant inter-observer variations, with poor overlap for trigone, membranous urethra, penile bulb, and left and right neurovascular bundles (DSC=0.44±0.22; 0.41±0.21; 0.66±0.21; 0.16±0.17; 0.15±0.15). Pelvic floor muscles recorded moderate to strong inter-observer agreement (ICC=0.50-0.97), although large outlier variations were observed. Inter-observer contouring variation was significant for multiple pelvic structures contoured on MR.

Feature-Based Deformable Registration Using Minimal Spanning Tree for Prostate MR Segmentation

Automatic and accurate segmentation of the prostate is still a challenging task due to intensity inhomogeneity and complicated deformation of MR images. To tackle these problems with multi-atlas segmentation, in this paper, we propose a new metric for image registration and new descriptor for label fusion. First, to reduce the amount of edges in entropic graph, a modified $\alpha $ -mutual information ( $\alpha $ -MI) based on fast minimal spanning tree (MST) is implemented for deformable registration. Second, localized $\alpha $ -MI allowing for the spatial information is proposed with the stochastic gradient optimization, and the feature space is encoded by a sparse auto-encoder. Finally, a multi-scale descriptor utilizing local self-similarity is integrated into the patch-based label fusion to obtain final segmentation. Experiments were performed on two subsets of totally 46 T2-weighted prostate MR images from 46 patients. Compared to $\alpha $ -MI based on ${k}$ -nearest neighbor graph, the registration time of $\alpha $ -MI based on fast MST can be reduced by almost half. The median Dice overlap of registration using localized $\alpha $ -MI on one subset is shown to improve significantly from 0.725 to 0.764 ( $p=1.14\times 10^{-5}$ ), compared to using $\alpha $ -MI without the spatial information. The median Dice overlap of prostate segmentation using the proposed method on 20 testing images of another subset is 0.871, and the median Hausdorff distance is 8.013 mm, which demonstrate a comparable accuracy to state-of-the-art methods.

Radiomics modelling of IMRT induced acute rectal toxicity using clinical and magnetic resonance imaging features

Introduction: Rectal toxicity is a dose limiting issue in prostate cancer radiotherapy. Prediction of these effects may be used to tailor the therapy. The purpose of this work was to develop predictive radiomic models based on clinical, dosimetric and radiomic features extracted from rectal wall magnetic resonance image (MRI). Materials and Methods: This study was conducted on 30 prostate cancer patients underwent intensity modulated radiation therapy (IMRT). Patient’s clinical and dosimetric parameters were collected and proctitis was assessed. All patients underwent MRI before and after IMRT with same protocol. Several radiomic features were extracted from rectal wall MR images including T2 weighted (T2W) and apparent diffusion coefficient (ADC) scans and robust features were found. Lasso regularization was used to create multivariable Elastic Net logistic regression model, based on clinical, dosimetric and robust features for prediction and feature selection, simultaneously. All models were cross-validated using repeated five-fold cross-validation and their performance were evaluated by area under the curve of the ROC curve (AUC), sensitivity and specificity. Results: Eleven predictive models including one clinical and ten radiomic models were built. All models were found as high predictive performance (AUC≥0.83). T2 based radiomic models were more predictive rather than ADC models. Post-IMRT T2W radiomic model also showed a good predictive performance. Image features enhanced the all model’s predictive performance. Conclusion: Rectal wall MR radiomic features integrating with clinical and dose-volume parameters could improve the prediction performance of radiotherapy induced proctitis. The authors have demonstrated the performance of radiomic features extracted from pre IMRT T2W and ADC MR Images for prediction of IMRT induced rectal toxicity. These results may be used to tailor the therapy in terms of patient and.

Magnetic hyperthermia of breast cancer cells and MRI relaxometry with dendrimer-coated iron-oxide nanoparticles

BackgroundRecently, some studies have focused on dendrimer nanopolymers as a magnetic resonance imaging (MRI) contrast agent or a vehicle for gene and drug delivery. Considering the suitable properties of these materials, they are appropriate candidates for coating iron-oxide nanoparticles which are applied in magnetic hyperthermia. To the best of our knowledge, the novelty of this study is the investigation of fourth-generation dendrimer-coated iron-oxide nanoparticles (G4@IONPs) in magnetic hyperthermia and MRI.MethodsIONPs were synthesized via co-precipitation and coated with the fourth generation (G4) of polyamidoamine dendrimer. The cytotoxicity of G4@IONPs with different concentrations was assessed in a human breast cancer cell line (MCF7) and human fibroblast cell line (HDF1). Hemolysis and stability of G4@IONPs were investigated, and in addition, the interaction of these particles with MCF7 cells was assessed by Prussian blue staining. Heat generation and specific absorption rate (SAR) were calculated from measurement and simulation results at 200 and 300 kHz. MCF7 and HDF1 cells were incubated with G4@IONPs for 2 h and then put into the magnetic coil for 120 min. Relaxometry experiments were performed with different concentrations of G4@IONPs with T1- and T2-weighted MR images.ResultsThe TEM results showed that G4@IONPs were 10 ± 4 nm. The in vitro toxicity assessments showed that synthesized nanoparticles had low toxicity. The viability of MCF7 cells incubated with G4@IONPs decreased significantly after magnetic hyperthermia. In addition, MR imaging revealed that G4@IONPs improved transverse relaxivity (r2) significantly.ConclusionsOur results encouraged the future application of G4@IONPs in magnetic hyperthermia and MR imaging.

Ferromagnetic Metal Side-Rails on Air-Hover HDR Patient Transport Table Can Cause Severe Skin Burns to Patients During MR Simulation for Brachytherapy

The air-hover system (Zephyr®, Diacor Inc) used for high-dose-rate brachytherapy (HDR BT) is designed for patient transfer between the HDR procedure suite and CT or MR simulation room. A recent case illustrated that the MR-conditional, ferromagnetic metal side-rails of the Zephyr system caused severe skin burns during our 3 Tesla MR simulation scan. Proposed solutions are presented to avoid potential patient skin burns when using a device with ferromagnetic metal side-rails or other components even if MR conditional. Institutional MR-guided, HDR BT program for cervical cancer consists of titanium Fletcher-Suit-Delclos tandem-and-ovoids (Varian Medical System, Inc.) implants under ultrasound guidance. Orthogonal X-rays before 3D simulation images and an AP X-ray after simulation scan were performed to validate applicator displacements during patient transfer between HDR and 3D image scan rooms. Our institutional MR scanning protocol for HDR BT includes 3D isotropically reconstructed T2- and T1-weighted MR, along with 2D sagittal T2-weighted MR (IJROBP 80;3:947-955;2011). The Zephyr system was implemented in August, 2014 after two cases of staff injury following patient transfer involving obese patients. A Zephyr arm support system was implemented in October, 2016 to facilitate intravenous access by anesthesiology staff. The Zephyr arm support system consists of two arm supports and two ferromagnetic metal side-rails. The ferromagnetic metal of these side-rails is a stainless steel alloy. In-house velcro straps are used to secure the arms and legs during transfer when using the Zephyr system. After 65 uneventful patient simulation scans a severe skin burn occurred on one patient’s arm which was in contact with the metal side rail during the 3T MR simulation scan (25 - 30 minutes duration) performed under anesthesia. The skin burn was initially identified as a third-degree burn by our institutional burn unit wound assessment team. This skin burn case was reported to institutional patient-safety-network (PSN). The vendor was immediately notified. A multi-disciplinary team of radiation therapists, MR-certified therapists, nurses, physicians, anesthesia team and medical physicists was created to evaluate and improve the clinical workflow and Zephyr system. New glass-fiber embedded polycarbonate side-rails were developed and built that should mitigate any risk of inductive heating. The glass-fiber embedded polycarbonate has tensile strength of 16,000 psi, along with impact strength of 2.06 f.-lbs per inch which meet the specifications of American Society for Testing Materials (ASTM: D3935 and D6098). Glass-fiber embedded polycarbonate side-rails that are MR-safe were developed and installed on an air-hover based HDR patient transfer system. This should eliminate the root cause of potential skin burns associated with ferromagnetic MR-compatible side rails in patients receiving MR scans for MRI-guided HDR brachytherapy.