Metabolic profiling of galectin-1 and galectin-3: a cross-sectional, multi-omics, association study.

OBJECTIVES: Experimental studies indicate a role for galectin-1 and galectin-3 in metabolic disease, but clinical evidence from larger populations is limited. METHODS: We measured circulating levels of galectin-1 and galectin-3 in the Prospective investigation of Obesity, Energy and Metabolism (POEM) study, participants (n = 502, all aged 50 years) and characterized the individual association profiles with metabolic markers, including clinical measures, metabolomics, adipose tissue distribution (Imiomics) and proteomics. RESULTS: Galectin-1 and galectin-3 were associated with fatty acids, lipoproteins and triglycerides including lipid measurements in the metabolomics analysis adjusted for body mass index (BMI). Galectin-1 was associated with several measurements of adiposity, insulin secretion and insulin sensitivity, while galectin-3 was associated with triglyceride-glucose index (TyG) and fasting insulin levels. Both galectins were associated with inflammatory pathways and fatty acid binding protein (FABP)4 and -5-regulated triglyceride metabolic pathways. Galectin-1 was also associated with several proteins related to adipose tissue differentiation. CONCLUSIONS: The association profiles for galectin-1 and galectin-3 indicate overlapping metabolic effects in humans, while the distinctly different associations seen with fat mass, fat distribution, and adipose tissue differentiation markers may suggest a functional role of galectin-1 in obesity.

Voxel-wise body composition analysis using image registration of a three-slice CT imaging protocol: methodology and proof-of-concept studies.

BACKGROUND: Computed tomography (CT) is an imaging modality commonly used for studies of internal body structures and very useful for detailed studies of body composition. The aim of this study was to develop and evaluate a fully automatic image registration framework for inter-subject CT slice registration. The aim was also to use the results, in a set of proof-of-concept studies, for voxel-wise statistical body composition analysis (Imiomics) of correlations between imaging and non-imaging data. METHODS: The current study utilized three single-slice CT images of the liver, abdomen, and thigh from two large cohort studies, SCAPIS and IGT. The image registration method developed and evaluated used both CT images together with image-derived tissue and organ segmentation masks. To evaluate the performance of the registration method, a set of baseline 3-single-slice CT images (from 2780 subjects including 8285 slices) from the SCAPIS and IGT cohorts were registered. Vector magnitude and intensity magnitude error indicating inverse consistency were used for evaluation. Image registration results were further used for voxel-wise analysis of associations between the CT images (as represented by tissue volume from Hounsfield unit and Jacobian determinant) and various explicit measurements of various tissues, fat depots, and organs collected in both cohort studies. RESULTS: Our findings demonstrated that the key organs and anatomical structures were registered appropriately. The evaluation parameters of inverse consistency, such as vector magnitude and intensity magnitude error, were on average less than 3 mm and 50 Hounsfield units. The registration followed by Imiomics analysis enabled the examination of associations between various explicit measurements (liver, spleen, abdominal muscle, visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), thigh SAT, intermuscular adipose tissue (IMAT), and thigh muscle) and the voxel-wise image information. CONCLUSION: The developed and evaluated framework allows accurate image registrations of the collected three single-slice CT images and enables detailed voxel-wise studies of associations between body composition and associated diseases and risk factors.

Automatic segmentation of large-scale CT image datasets for detailed body composition analysis.

BACKGROUND: Body composition (BC) is an important factor in determining the risk of type 2-diabetes and cardiovascular disease. Computed tomography (CT) is a useful imaging technique for studying BC, however manual segmentation of CT images is time-consuming and subjective. The purpose of this study is to develop and evaluate fully automated segmentation techniques applicable to a 3-slice CT imaging protocol, consisting of single slices at the level of the liver, abdomen, and thigh, allowing detailed analysis of numerous tissues and organs. METHODS: The study used more than 4000 CT subjects acquired from the large-scale SCAPIS and IGT cohort to train and evaluate four convolutional neural network based architectures: ResUNET, UNET++, Ghost-UNET, and the proposed Ghost-UNET++. The segmentation techniques were developed and evaluated for automated segmentation of the liver, spleen, skeletal muscle, bone marrow, cortical bone, and various adipose tissue depots, including visceral (VAT), intraperitoneal (IPAT), retroperitoneal (RPAT), subcutaneous (SAT), deep (DSAT), and superficial SAT (SSAT), as well as intermuscular adipose tissue (IMAT). The models were trained and validated for each target using tenfold cross-validation and test sets. RESULTS: The Dice scores on cross validation in SCAPIS were: ResUNET 0.964 (0.909-0.996), UNET++ 0.981 (0.927-0.996), Ghost-UNET 0.961 (0.904-0.991), and Ghost-UNET++ 0.968 (0.910-0.994). All four models showed relatively strong results, however UNET++ had the best performance overall. Ghost-UNET++ performed competitively compared to UNET++ and showed a more computationally efficient approach. CONCLUSION: Fully automated segmentation techniques can be successfully applied to a 3-slice CT imaging protocol to analyze multiple tissues and organs related to BC. The overall best performance was achieved by UNET++, against which Ghost-UNET++ showed competitive results based on a more computationally efficient approach. The use of fully automated segmentation methods can reduce analysis time and provide objective results in large-scale studies of BC.

Serum levels of perfluoroalkyl substances (PFAS) and body composition - A cross-sectional study in a middle-aged population.

BACKGROUND: It has been suggested that per- and polyfluoroalkyl substances (PFAS) are endocrine disruptors with a potential to influence fat mass. OBJECTIVE: The primary hypothesis tested was that we would find positive relationships for PFAS vs measures of adiposity. METHODS: In 321 subjects all aged 50 years in the POEM study, five PFAS (perfluorooctane sulfonic acid (PFOS), perfluorooctanoic acid (PFOA), perfluorohexane sulfonic acid (PFHxS), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA)) were measured in serum together with a Dual-energy X-ray absorptiometry (DXA) scan for determination of fat and lean mass. Whole-body magnetic resonance imaging scan was performed and the body was divided into >1 million voxels. Voxel-wise statistical analysis was carried out by a novel method denoted Imiomics. RESULTS: PFOS and PFHxS, did not show any consistent associations with body composition. However, PFOA, and especially PFNA and PFDA, levels were inversely related to most traditional measures reflecting the amount of fat in women, but not in men. In the Imiomics analysis of tissue volume, PFDA and PFNA levels were inversely related to the volume of subcutaneous fat, mainly in the arm, trunk and hip regions in women, while no such clear relationship was seen in men. Also, the visceral fat content of the liver, the pericardium, and the gluteus muscle were inversely related to PFDA and PFNA in women. DISCUSSION: Contrary to our hypothesis, some PFAS showed inverse relationships vs measurements of adiposity. CONCLUSION: PFOS and PFHxS levels in plasma did not show any consistent associations with body composition, but PFOA, and especially PFNA and PFDA were inversely related to multiple measures reflecting the amount of fat, but in women only.

Cardiovascular-related proteins and the abdominal visceral to subcutaneous adipose tissue ratio.

BACKGROUND AND AIMS: An increased amount of visceral adipose tissues has been related to atherosclerosis and future cardiovascular events. The present study aims to investigate how the abdominal fat distribution links to plasma levels of cardiovascular-related proteins. METHOD AND RESULTS: In the Prospective investigation of Obesity, Energy and Metabolism (POEM) study (n = 326, all aged 50 years), abdominal visceral (VAT) and subcutaneous (SAT) adipose tissue volumes were quantified by MRI. Eighty-six cardiovascular-related proteins were measured by the proximity extension assay (PEA). Similar investigations were carried out in the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS) study (n = 400, all aged 75 years). In the discovery dataset (POEM), 10 proteins were related to the VAT/SAT-ratio using false discovery rate <.05. Of those, Cathepsin D (CTSD), Interleukin-1 receptor antagonist protein (IL-1RA) and Growth hormone (GH) (inversely) were related to the VAT/SAT-ratio in the validation in PIVUS following adjustment for sex, BMI, smoking, education level and exercise habits (p < 0.05). In a secondary analysis, a meta-analysis of the two samples suggested that 15 proteins could be linked to the VAT/SAT-ratio following adjustment as above and Bonferroni-correction of the p-value. CONCLUSION: Three cardiovascular-related proteins, cathepsin D, IL-1RA and growth hormone, were being associated with the distribution of abdominal adipose tissue using a discovery/validation approach. A meta-analysis of the two samples suggested that also a number of other cardiovascular-related proteins could be associated with an unfavorable abdominal fat distribution.

Voxel-wise Study of Cohort Associations in Whole-Body MRI: Application in Metabolic Syndrome and Its Components.

Background The metabolic syndrome is related to obesity and ectopic fat distribution. Purpose To investigate whether an image analysis approach that uses image registration for whole-body voxel-wise analysis could provide additional information about the relationship between metabolic syndrome and body composition compared with traditional image analysis. Materials and Methods Whole-body quantitative water-fat MRI was performed in a population-based prospective study on obesity, energy, and metabolism between October 2010 and November 2016. Fat mass was measured with dual-energy x-ray absorptiometry (DXA). Whole-body voxel-wise analysis of tissue volume and fat content was applied in more than 2 million voxels from the whole-body examinations by automated interindividual deformable image registration of the water and fat MRI data. Metabolic syndrome was diagnosed by the harmonized National Cholesterol Education Program criteria. Two-tailed t tests were used and P values less than .05 were considered to indicate statistical significance. Results This study evaluated 167 women and 159 men (mean age, 50 years) by using voxel-wise analysis. Metabolic syndrome (13.5%; 44 of 326) was related to traditional measurements of fat distribution, such as total fat mass at DXA, visceral and subcutaneous adipose tissue, and liver and pancreatic fat at MRI. Voxel-wise analysis found metabolic syndrome related to liver, heart, and perirenal fat volume; fat content in subcutaneous fat in the hip region in both sexes; fatty infiltration of leg muscles in men, especially in gluteus maximus; and pericardial and aortic perivascular fat mainly in women. Sex differences in associations with subcutaneous adipose tissue were identified. In women, metabolic syndrome diagnosis was linked to regional differences in associations to adipose tissue volumes in upper versus lower body, and dorsal versus ventral abdominal depots. In men similar gradients were only seen in individual components. Conclusion In addition to showing the relationships between metabolic syndrome and body composition in a detailed and intuitive fashion in the whole body, the voxel-wise analysis provided additional information compared with traditional image analysis. © RSNA, 2020 Online supplemental material is available for this article.

Region-by-region analysis of PET, MRI, and histology in en bloc-resected oligodendrogliomas reveals intra-tumoral heterogeneity.

PURPOSE: Oligodendrogliomas are heterogeneous tumors in terms of imaging appearance, and a deeper understanding of the histopathological tumor characteristics in correlation to imaging parameters is needed. We used PET-to-MRI-to-histology co-registration with the aim of studying intra-tumoral 11C-methionine (MET) uptake in relation to tumor perfusion and the protein expression of histological cell markers in corresponding areas. METHODS: Consecutive histological sections of four tumors covering the entire en bloc-removed tumor were immunostained with antibodies against IDH1-mutated protein (tumor cells), Ki67 (proliferating cells), and CD34 (blood vessels). Software was developed for anatomical landmarks-based co-registration of subsequent histological images, which were overlaid on corresponding MET PET scans and MRI perfusion maps. Regions of interest (ROIs) on PET were selected throughout the entire tumor volume, covering hot spot areas, areas adjacent to hot spots, and tumor borders with infiltrating zone. Tumor-to-normal tissue (T/N) ratios of MET uptake and mean relative cerebral blood volume (rCBV) were measured in the ROIs and protein expression of histological cell markers was quantified in corresponding regions. Statistical correlations were calculated between MET uptake, rCBV, and quantified protein expression. RESULTS: A total of 84 ROIs were selected in four oligodendrogliomas. A significant correlation (p < 0.05) between MET uptake and tumor cell density was demonstrated in all tumors separately. In two tumors, MET correlated with the density of proliferating cells and vessel cell density. There were no significant correlations between MET uptake and rCBV, and between rCBV and histological cell markers. CONCLUSIONS: The MET uptake in hot spots, outside hotspots, and in infiltrating tumor edges unanimously reflects tumor cell density. The correlation between MET uptake and vessel density and density of proliferating cells is less stringent in infiltrating tumor edges and is probably more susceptible to artifacts caused by larger blood vessels surrounding the tumor. Although based on a limited number of samples, this study provides histological proof for MET as an indicator of tumor cell density and for the lack of statistically significant correlations between rCBV and histological cell markers in oligodendrogliomas.

Relationship between endothelium-dependent vasodilation and fat distribution using the new "imiomics" image analysis technique.

BACKGROUND AND AIMS: We investigated how vasoreactivity in the brachial artery and the forearm resistance vessels were related to fat distribution and tissue volume, using both traditional imaging analysis and a new technique, called "Imiomics", whereby vasoreactivity was related to each of the >2M 3D image elements included in the whole-body magnetic resonance imaging (MRI). METHODS AND RESULTS: In 326 subjects in the Prospective investigation of Obesity, Energy and Metabolism (POEM) study (all aged 50 years), endothelium-dependent vasodilation was measured by acetylcholine infusion in the brachial artery (EDV) and flow-mediated vasodilation (FMD). Fat distribution was evaluated by dual-energy X-ray absorptiometry (DXA) and magnetic resonance imaging (MRI). EDV, but not FMD, was significantly related to total fat mass, liver fat, subcutaneous (SAT) and visceral (VAT) adipose tissue in a negative fashion in women, but not in men. Using Imiomics, an inverse relationship was seen between EDV and a local tissue volume of SAT in both the upper part of the body, as well as the gluteo-femoral part and the medial parts of the legs in women. Also the size of the liver, heart and VAT was inversely related to EDV. In men, less pronounced relationships were seen. FMD was also significantly related to local tissue volume of upper-body SAT and liver fat in women, but less so in men. CONCLUSION: EDV, and to a lesser degree also FMD, were related to liver fat, SAT and VAT in women, but less so in men. Imiomics both confirmed findings from traditional methods and resulted in new, more detailed results.

Deep learning-based dose prediction for magnetic resonance-guided prostate radiotherapy.

BACKGROUND: Daily adaptive radiotherapy, as performed with the Elekta Unity MR-Linac, requires choosing between different adaptation methods, namely ATP (Adapt to Position) and ATS (Adapt to Shape), where the latter requires daily re-contouring to obtain a dose plan tailored to the daily anatomy. These steps are inherently resource-intensive, and quickly predicting the dose distribution and the dosimetric evaluation criteria while the patient is on the table could facilitate a fast selection of adaptation method and decrease the treatment times. PURPOSE: In this work, we aimed to develop a deep-learning-based dose-prediction pipeline for prostate MR-Linac treatments. METHODS: Two hundred twelve MR-images, structure sets, and dose distributions from 35 prostate patients treated with 6.1 Gy for 7 or 6 fractions at our MR-Linac were included, split into train/test partitions of 152/60 images, respectively. A deep-learning segmentation network was trained to segment the CTV (prostate), bladder, and rectum. A second network was trained to predict the dose distribution based on manually delineated structures. At inference, the predicted segmentations acted as input to the dose prediction network, and the predicted dose was compared to the true (optimized in the treatment planning system) dose distribution. RESULTS: Median DSC values from the segmentation network were 0.90/0.94/0.87 for CTV/bladder/rectum. Predicted segmentations as input to the dose prediction resulted in mean differences between predicted and true doses of 0.7%/0.7%/1.7% (relative to the prescription dose) for D98%/D95%/D2% for the CTV. For the bladder, the difference was 0.7%/0.3% for Dmean/D2% and for the rectum 0.1/0.2/0.2 pp (percentage points) for V33Gy/V38Gy/V41Gy. In comparison, true segmentations as input resulted in differences of 1.1%/0.9%/1.6% for CTV, 0.5%/0.4% for bladder, and 0.7/0.4/0.3 pp for the rectum. Only D2% for CTV and Dmean/D2% for bladder were found to be statistically significantly better when using true structures instead of predicted structures as input to the dose prediction. CONCLUSIONS: Small differences in the fulfillment of clinical dose-volume constraints are seen between utilizing deep-learning predicted structures as input to a dose prediction network and manual structures. Overall mean differences <2% indicate that the dose-prediction pipeline is useful as a decision support tool where differences are >2%.

Classification of rotation-invariant biomedical images using equivariant neural networks.

Transmission electron microscopy (TEM) is an imaging technique used to visualize and analyze nano-sized structures and objects such as virus particles. Light microscopy can be used to diagnose diseases or characterize e.g. blood cells. Since samples under microscopes exhibit certain symmetries, such as global rotation invariance, equivariant neural networks are presumed to be useful. In this study, a baseline convolutional neural network is constructed in the form of the commonly used VGG16 classifier. Thereafter, it is modified to be equivariant to the p4 symmetry group of rotations of multiples of 90° using group convolutions. This yields a number of benefits on a TEM virus dataset, including higher top validation set accuracy by on average 7.6% and faster convergence during training by on average 23.1% of that of the baseline. Similarly, when training and testing on images of blood cells, the convergence time for the equivariant neural network is 7.9% of that of the baseline. From this it is concluded that augmentation strategies for rotation can be skipped. Furthermore, when modelling the accuracy versus amount of TEM virus training data with a power law, the equivariant network has a slope of - 0.43 compared to - 0.26 of the baseline. Thus the equivariant network learns faster than the baseline when more training data is added. This study extends previous research on equivariant neural networks applied to images which exhibit symmetries to isometric transformations.

Multi-Level Residual Dual Attention Network for Major Cerebral Arteries Segmentation in MRA Toward Diagnosis of Cerebrovascular Disorders.

Segmentation of major brain vessels is very important for the diagnosis of cerebrovascular disorders and subsequent surgical planning. Vessel segmentation is an important preprocessing step for a wide range of algorithms for the automatic diagnosis or treatment of several vascular pathologies and as such, it is valuable to have a well-performing vascular segmentation pipeline. In this article, we propose an end-to-end multiscale residual dual attention deep neural network for resilient major brain vessel segmentation. In the proposed network, the encoder and decoder blocks of the U-Net are replaced with the multi-level atrous residual blocks to enhance the learning capability by increasing the receptive field to extract the various semantic coarse- and fine-grained features. Dual attention block is incorporated in the bottleneck to perform effective multiscale information fusion to obtain detailed structure of blood vessels. The methods were evaluated on the publicly available TubeTK data set. The proposed method outperforms the state-of-the-art techniques with dice of 0.79 on the whole-brain prediction. The statistical and visual assessments indicate that proposed network is robust to outliers and maintains higher consistency in vessel continuity than the traditional U-Net and its variations.

Streamlining neuroradiology workflow with AI for improved cerebrovascular structure monitoring.

Radiological imaging to examine intracranial blood vessels is critical for preoperative planning and postoperative follow-up. Automated segmentation of cerebrovascular anatomy from Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) can provide radiologists with a more detailed and precise view of these vessels. This paper introduces a domain generalized artificial intelligence (AI) solution for volumetric monitoring of cerebrovascular structures from multi-center MRAs. Our approach utilizes a multi-task deep convolutional neural network (CNN) with a topology-aware loss function to learn voxel-wise segmentation of the cerebrovascular tree. We use Decorrelation Loss to achieve domain regularization for the encoder network and auxiliary tasks to provide additional regularization and enable the encoder to learn higher-level intermediate representations for improved performance. We compare our method to six state-of-the-art 3D vessel segmentation methods using retrospective TOF-MRA datasets from multiple private and public data sources scanned at six hospitals, with and without vascular pathologies. The proposed model achieved the best scores in all the qualitative performance measures. Furthermore, we have developed an AI-assisted Graphical User Interface (GUI) based on our research to assist radiologists in their daily work and establish a more efficient work process that saves time.

A detailed analysis of body composition in relation to cardiopulmonary exercise test indices.

A cardiopulmonary exercise test (CPET) is a test assessing an individual's physiological response during exercise. Results may be affected by body composition, which is best evaluated through imaging techniques like magnetic resonance imaging (MRI). The aim of this study was to assess relationships between body composition and indices obtained from CPET. A total of 234 participants (112 female), all aged 50 years, underwent CPETs and whole-body MRI scans (> 1 million voxels). Voxel-wise statistical analysis of tissue volume and fat content was carried out with a method called Imiomics and related to the CPET indices peak oxygen consumption (V̇O2peak), V̇O2peak scaled by body weight (V̇O2kg) and by total lean mass (V̇O2lean), ventilatory efficiency (V̇E/V̇CO2-slope), work efficiency (ΔV̇O2/ΔWR) and peak exercise respiratory exchange ratio (RERpeak). V̇O2peak showed the highest positive correlation with volume of skeletal muscle. V̇O2kg negatively correlated with tissue volume in subcutaneous fat, particularly gluteal fat. RERpeak negatively correlated with tissue volume in skeletal muscle, subcutaneous fat, visceral fat and liver. Some associations differed between sexes: in females ΔV̇O2/ΔWR correlated positively with tissue volume of subcutaneous fat and V̇E/V̇CO2-slope with tissue volume of visceral fat, and, in males, V̇O2peak correlated positively to lung volume. In conclusion, voxel-based Imiomics provided detailed insights into how CPET indices were related to the tissue volume and fat content of different body structures.

Improved automated tumor segmentation in whole-body 3D scans using multi-directional 2D projection-based priors.

Early cancer detection, guided by whole-body imaging, is important for the overall survival and well-being of the patients. While various computer-assisted systems have been developed to expedite and enhance cancer diagnostics and longitudinal monitoring, the detection and segmentation of tumors, especially from whole-body scans, remain challenging. To address this, we propose a novel end-to-end automated framework that first generates a tumor probability distribution map (TPDM), incorporating prior information about the tumor characteristics (e.g. size, shape, location). Subsequently, the TPDM is integrated with a state-of-the-art 3D segmentation network along with the original PET/CT or PET/MR images. This aims to produce more meaningful tumor segmentation masks compared to using the baseline 3D segmentation network alone. The proposed method was evaluated on three independent cohorts (autoPET, CAR-T, cHL) of images containing different cancer forms, obtained with different imaging modalities, and acquisition parameters and lesions annotated by different experts. The evaluation demonstrated the superiority of our proposed method over the baseline model by significant margins in terms of Dice coefficient, and lesion-wise sensitivity and precision. Many of the extremely small tumor lesions (i.e. the most difficult to segment) were missed by the baseline model but detected by the proposed model without additional false positives, resulting in clinically more relevant assessments. On average, an improvement of 0.0251 (autoPET), 0.144 (CAR-T), and 0.0528 (cHL) in overall Dice was observed. In conclusion, the proposed TPDM-based approach can be integrated with any state-of-the-art 3D UNET with potentially more accurate and robust segmentation results.

A whole-body diffusion MRI normal atlas: development, evaluation and initial use.

BACKGROUND: Statistical atlases can provide population-based descriptions of healthy volunteers and/or patients and can be used for region- and voxel-based analysis. This work aims to develop whole-body diffusion atlases of healthy volunteers scanned at 1.5T and 3T. Further aims include evaluating the atlases by establishing whole-body Apparent Diffusion Coefficient (ADC) values of healthy tissues and including healthy tissue deviations in an automated tumour segmentation task. METHODS: Multi-station whole-body Diffusion Weighted Imaging (DWI) and water-fat Magnetic Resonance Imaging (MRI) of healthy volunteers (n = 45) were acquired at 1.5T (n = 38) and/or 3T (n = 29), with test-retest imaging for five subjects per scanner. Using deformable image registration, whole-body MRI data was registered and composed into normal atlases. Healthy tissue ADCmean was manually measured for ten tissues, with test-retest percentage Repeatability Coefficient (%RC), and effect of age, sex and scanner assessed. Voxel-wise whole-body analyses using the normal atlases were studied with ADC correlation analyses and an automated tumour segmentation task. For the latter, lymphoma patient MRI scans (n = 40) with and without information about healthy tissue deviations were entered into a 3D U-Net architecture. RESULTS: Sex- and Body Mass Index (BMI)-stratified whole-body high b-value DWI and ADC normal atlases were created at 1.5T and 3T. %RC of healthy tissue ADCmean varied depending on tissue assessed (4-48% at 1.5T, 6-70% at 3T). Scanner differences in ADCmean were visualised in Bland-Altman analyses of dually scanned subjects. Sex differences were measurable for liver, muscle and bone at 1.5T, and muscle at 3T. Volume of Interest (VOI)-based multiple linear regression, and voxel-based correlations in normal atlas space, showed that age and ADC were negatively associated for liver and bone at 1.5T, and positively associated with brain tissue at 1.5T and 3T. Adding voxel-wise information about healthy tissue deviations in an automated tumour segmentation task gave numerical improvements in the segmentation metrics Dice score, sensitivity and precision. CONCLUSIONS: Whole-body DWI and ADC normal atlases were created at 1.5T and 3T, and applied in whole-body voxel-wise analyses.

Patient specific deep learning based segmentation for magnetic resonance guided prostate radiotherapy.

Background and Purpose: Treatments on combined Magnetic Resonance (MR) scanners and Linear Accelerators (Linacs) for radiotherapy, called MR-Linacs, often require daily contouring. Currently, deformable image registration (DIR) algorithms propagate contours from reference scans, however large shape and size changes can be troublesome. Artificial neural network (ANN) based contouring may alleviate this issue, however generally requires large datasets for training. Mitigating the problem of scarcity of data, we propose patient specific networks trained on a single dataset for each patient, for contouring onto the following datasets in an adaptive MR-Linacworkflow. Materials and Methods: MR-scans from 17 prostate patients treated on an MR-Linac with contours of Clinical Target Volume (CTV), bladder and rectum were utilized. U-net shaped models were trained based on the image from the first fraction of each patient, and subsequently applied onto the following treatment images. Results were compared with manual contours in terms of the Dice coefficient and Added Path Length (APL). As benchmark, contours propagated through the clinical DIR algorithm were similarly evaluated. Results: In Dice coefficient the ANN output was 0.92 ± 0.03, 0.93 ± 0.07 and 0.84 ± 0.10 while for DIR 0.95 ± 0.03, 0.93 ± 0.08, 0.88 ± 0.06 for CTV, bladder and rectum respectively. Similarly, APL where 3109 ± 1642, 7250 ± 4234 and 5041 ± 2666 for ANN and 1835 ± 1621, 7236 ± 4287 and 4170 ± 2920 voxels for DIR. Conclusions: Patient specific ANN models trained on images from the first fraction of a prostate MR-Linac treatment showed similar accuracy when applied to the subsequent fraction images as a clinically implemented DIR method.

MIMIR: Deep Regression for Automated Analysis of UK Biobank MRI Scans.

UK Biobank (UKB) has recruited more than 500 000 volunteers from the United Kingdom, collecting health-related information on genetics, lifestyle, blood biochemistry, and more. Ongoing medical imaging of 100 000 participants with 70 000 follow-up sessions will yield up to 170 000 MRI scans, enabling image analysis of body composition, organs, and muscle. This study presents an experimental inference engine for automated analysis of UKB neck-to-knee body 1.5-T MRI scans. This retrospective cross-validation study includes data from 38 916 participants (52% female; mean age, 64 years) to capture baseline characteristics, such as age, height, weight, and sex, as well as measurements of body composition, organ volumes, and abstract properties, such as grip strength, pulse rate, and type 2 diabetes status. Prediction intervals for each end point were generated based on uncertainty quantification. On a subsequent release of UKB data, the proposed method predicted 12 body composition metrics with a 3% median error and yielded mostly well-calibrated individual prediction intervals. The processing of MRI scans from 1000 participants required 10 minutes. The underlying method used convolutional neural networks for image-based mean-variance regression on two-dimensional representations of the MRI data. An implementation was made publicly available for fast and fully automated estimation of 72 different measurements from future releases of UKB image data. Keywords: MRI, Adipose Tissue, Obesity, Metabolic Disorders, Volume Analysis, Whole-Body Imaging, Quantification, Supervised Learning, Convolutional Neural Network (CNN) © RSNA, 2022.

Erratum for: MIMIR: Deep Regression for Automated Analysis of UK Biobank MRI Scans.

[This corrects the article DOI: 10.1148/ryai.210178.].

An image registration method for voxel-wise analysis of whole-body oncological PET-CT.

Whole-body positron emission tomography-computed tomography (PET-CT) imaging in oncology provides comprehensive information of each patient's disease status. However, image interpretation of volumetric data is a complex and time-consuming task. In this work, an image registration method targeted towards computer-aided voxel-wise analysis of whole-body PET-CT data was developed. The method used both CT images and tissue segmentation masks in parallel to spatially align images step-by-step. To evaluate its performance, a set of baseline PET-CT images of 131 classical Hodgkin lymphoma (cHL) patients and longitudinal image series of 135 head and neck cancer (HNC) patients were registered between and within subjects according to the proposed method. Results showed that major organs and anatomical structures generally were registered correctly. Whole-body inverse consistency vector and intensity magnitude errors were on average less than 5 mm and 45 Hounsfield units respectively in both registration tasks. Image registration was feasible in time and the nearly automatic pipeline enabled efficient image processing. Metabolic tumor volumes of the cHL patients and registration-derived therapy-related tissue volume change of the HNC patients mapped to template spaces confirmed proof-of-concept. In conclusion, the method established a robust point-correspondence and enabled quantitative visualization of group-wise image features on voxel level.

Relationships between carotid artery intima-media thickness and echogenicity and body composition using a new magnetic resonance imaging voxel-based technique.

BACKGROUND: We evaluated how carotid artery intima-media thickness (IMT) and the echogenicity of the intima-media (IM-GSM), measured by ultrasound, were related to body composition, evaluated by both traditional imaging techniques, as well as with a new voxel-based "Imiomics" technique. METHODS: In 321 subjects all aged 50 years in the POEM study, IMT and IM-GSM were measured together with a DXA scan for determination of fat and lean mass. Also a whole-body MRI scan was performed and the body volume was divided into >1 million voxels in a standardized fashion. IMT and IM-GSM were related to each of these voxels to create a 3D-view of how these measurements were related to size of each part of the body. RESULTS: IM-GSM was inversely related to almost all traditional measurements of body composition, like fat and lean mass, liver fat, visceral and subcutaneous fat, but this was not seen for IMT. Using Imiomics, IMT was positively related to the intraabdominal fat volume, as well of the leg skeletal muscle in women. In males, IMT was mainly positively related to the leg skeletal muscle volume. IM-GSM was inversely related to the volume of the SAT in the upper part of the body, leg skeletal muscle, the liver and intraabdominal fat in both men and women. CONCLUSION: The voxel-based Imiomics technique provided a detailed view of how the echogenicity of the carotid artery wall was related to body composition, being inversely related to the volume of the major fat depots, as well as leg skeletal muscle.