Generative Adversarial Networks to Synthesize Missing T1 and FLAIR MRI Sequences for Use in a Multisequence Brain Tumor Segmentation Model.

Background Missing MRI sequences represent an obstacle in the development and use of deep learning (DL) models that require multiple inputs. Purpose To determine if synthesizing brain MRI scans using generative adversarial networks (GANs) allows for the use of a DL model for brain lesion segmentation that requires T1-weighted images, postcontrast T1-weighted images, fluid-attenuated inversion recovery (FLAIR) images, and T2-weighted images. Materials and Methods In this retrospective study, brain MRI scans obtained between 2011 and 2019 were collected, and scenarios were simulated in which the T1-weighted images and FLAIR images were missing. Two GANs were trained, validated, and tested using 210 glioblastomas (GBMs) (Multimodal Brain Tumor Image Segmentation Benchmark [BRATS] 2017) to generate T1-weighted images from postcontrast T1-weighted images and FLAIR images from T2-weighted images. The quality of the generated images was evaluated with mean squared error (MSE) and the structural similarity index (SSI). The segmentations obtained with the generated scans were compared with those obtained with the original MRI scans using the dice similarity coefficient (DSC). The GANs were validated on sets of GBMs and central nervous system lymphomas from the authors' institution to assess their generalizability. Statistical analysis was performed using the Mann-Whitney, Friedman, and Dunn tests. Results Two hundred ten GBMs from the BRATS data set and 46 GBMs (mean patient age, 58 years ± 11 [standard deviation]; 27 men [59%] and 19 women [41%]) and 21 central nervous system lymphomas (mean patient age, 67 years ± 13; 12 men [57%] and nine women [43%]) from the authors' institution were evaluated. The median MSE for the generated T1-weighted images ranged from 0.005 to 0.013, and the median MSE for the generated FLAIR images ranged from 0.004 to 0.103. The median SSI ranged from 0.82 to 0.92 for the generated T1-weighted images and from 0.76 to 0.92 for the generated FLAIR images. The median DSCs for the segmentation of the whole lesion, the FLAIR hyperintensities, and the contrast-enhanced areas using the generated scans were 0.82, 0.71, and 0.92, respectively, when replacing both T1-weighted and FLAIR images; 0.84, 0.74, and 0.97 when replacing only the FLAIR images; and 0.97, 0.95, and 0.92 when replacing only the T1-weighted images. Conclusion Brain MRI scans generated using generative adversarial networks can be used as deep learning model inputs in case MRI sequences are missing. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Zhong in this issue. An earlier incorrect version of this article appeared online. This article was corrected on April 12, 2021.

Determining age and sex-specific distribution of pancreatic whole-gland CT attenuation using artificial intelligence aided image segmentation: Associations with body composition and pancreatic cancer risk.

BACKGROUND & AIMS: Increased intrapancreatic fat is associated with pancreatic diseases; however, there are no established objective diagnostic criteria for fatty pancreas. On non-contrast computed tomography (CT), adipose tissue shows negative Hounsfield Unit (HU) attenuations (-150 to -30 HU). Using whole organ segmentation on non-contrast CT, we aimed to describe whole gland pancreatic attenuation and establish 5th and 10th percentile thresholds across a spectrum of age and sex. Subsequently, we aimed to evaluate the association between low pancreatic HU and risk of pancreatic ductal adenocarcinoma (PDAC). METHODS: The whole pancreas was segmented in 19,456 images from 469 non-contrast CT scans. A convolutional neural network was trained to assist pancreas segmentation. Mean pancreatic HU, volume, and body composition metrics were calculated. The lower 5th and 10th percentile for mean pancreatic HU were identified, examining the association with age and sex. Pre-diagnostic CT scans from patients who later developed PDAC were compared to cancer-free controls. RESULTS: Less than 5th percentile mean pancreatic HU was significantly associated with increase in BMI (OR 1.07; 1.03-1.11), visceral fat (OR 1.37; 1.15-1.64), total abdominal fat (OR 1.12; 1.03-1.22), and diabetes mellitus type 1 (OR 6.76; 1.68-27.28). Compared to controls, pre-diagnostic scans in PDAC cases had lower mean whole gland pancreatic HU (-0.2 vs 7.8, p = 0.026). CONCLUSION: In this study, we report age and sex-specific distribution of pancreatic whole-gland CT attenuation. Compared to controls, mean whole gland pancreatic HU is significantly lower in the pre-diagnostic phase of PDAC.

A Deep Learning Tool for Automated Radiographic Measurement of Acetabular Component Inclination and Version After Total Hip Arthroplasty.

BACKGROUND: Inappropriate acetabular component angular position is believed to increase the risk of hip dislocation after total hip arthroplasty. However, manual measurement of these angles is time consuming and prone to interobserver variability. The purpose of this study was to develop a deep learning tool to automate the measurement of acetabular component angles on postoperative radiographs. METHODS: Two cohorts of 600 anteroposterior (AP) pelvis and 600 cross-table lateral hip postoperative radiographs were used to develop deep learning models to segment the acetabular component and the ischial tuberosities. Cohorts were manually annotated, augmented, and randomly split to train-validation-test data sets on an 8:1:1 basis. Two U-Net convolutional neural network models (one for AP and one for cross-table lateral radiographs) were trained for 50 epochs. Image processing was then deployed to measure the acetabular component angles on the predicted masks for anatomical landmarks. Performance of the tool was tested on 80 AP and 80 cross-table lateral radiographs. RESULTS: The convolutional neural network models achieved a mean Dice similarity coefficient of 0.878 and 0.903 on AP and cross-table lateral test data sets, respectively. The mean difference between human-level and machine-level measurements was 1.35° (σ = 1.07°) and 1.39° (σ = 1.27°) for the inclination and anteversion angles, respectively. Differences of 5° or more between human-level and machine-level measurements were observed in less than 2.5% of cases. CONCLUSION: We developed a highly accurate deep learning tool to automate the measurement of angular position of acetabular components for use in both clinical and research settings. LEVEL OF EVIDENCE: III.

Deep Learning Artificial Intelligence Model for Assessment of Hip Dislocation Risk Following Primary Total Hip Arthroplasty From Postoperative Radiographs.

BACKGROUND: Dislocation is a common complication following total hip arthroplasty (THA), and accounts for a high percentage of subsequent revisions. The purpose of this study is to illustrate the potential of a convolutional neural network model to assess the risk of hip dislocation based on postoperative anteroposterior pelvis radiographs. METHODS: We retrospectively evaluated radiographs for a cohort of 13,970 primary THAs with 374 dislocations over 5 years of follow-up. Overall, 1490 radiographs from dislocated and 91,094 from non-dislocated THAs were included in the analysis. A convolutional neural network object detection model (YOLO-V3) was trained to crop the images by centering on the femoral head. A ResNet18 classifier was trained to predict subsequent hip dislocation from the cropped imaging. The ResNet18 classifier was initialized with ImageNet weights and trained using FastAI (V1.0) running on PyTorch. The training was run for 15 epochs using 10-fold cross validation, data oversampling, and augmentation. RESULTS: The hip dislocation classifier achieved the following mean performance (standard deviation): accuracy = 49.5 (4.1%), sensitivity = 89.0 (2.2%), specificity = 48.8 (4.2%), positive predictive value = 3.3 (0.3%), negative predictive value = 99.5 (0.1%), and area under the receiver operating characteristic curve = 76.7 (3.6%). Saliency maps demonstrated that the model placed the greatest emphasis on the femoral head and acetabular component. CONCLUSION: Existing prediction methods fail to identify patients at high risk of dislocation following THA. Our radiographic classifier model has high sensitivity and negative predictive value, and can be combined with clinical risk factor information for rapid assessment of risk for dislocation following THA. The model further suggests radiographic locations which may be important in understanding the etiology of prosthesis dislocation. Importantly, our model is an illustration of the potential of automated imaging artificial intelligence models in orthopedics. LEVEL OF EVIDENCE: Level III.

Deep Learning Improves the Temporal Reproducibility of Aortic Measurement.

Imaging-based measurements form the basis of surgical decision making in patients with aortic aneurysm. Unfortunately, manual measurement suffer from suboptimal temporal reproducibility, which can lead to delayed or unnecessary intervention. We tested the hypothesis that deep learning could improve upon the temporal reproducibility of CT angiography-derived thoracic aortic measurements in the setting of imperfect ground-truth training data. To this end, we trained a standard deep learning segmentation model from which measurements of aortic volume and diameter could be extracted. First, three blinded cardiothoracic radiologists visually confirmed non-inferiority of deep learning segmentation maps with respect to manual segmentation on a 50-patient hold-out test cohort, demonstrating a slight preference for the deep learning method (p < 1e-5). Next, reproducibility was assessed by evaluating measured change (coefficient of reproducibility and standard deviation) in volume and diameter values extracted from segmentation maps in patients for whom multiple scans were available and whose aortas had been deemed stable over time by visual assessment (n = 57 patients, 206 scans). Deep learning temporal reproducibility was superior for measures of both volume (p < 0.008) and diameter (p < 1e-5) and reproducibility metrics compared favorably with previously reported values of manual inter-rater variability. Our work motivates future efforts to apply deep learning to aortic evaluation.

Phases of Metabolic and Soft Tissue Changes in Months Preceding a Diagnosis of Pancreatic Ductal Adenocarcinoma.

BACKGROUND & AIMS: Identifying metabolic abnormalities that occur before pancreatic ductal adenocarcinoma (PDAC) diagnosis could increase chances for early detection. We collected data on changes in metabolic parameters (glucose, serum lipids, triglycerides; total, low-density, and high-density cholesterol; and total body weight) and soft tissues (abdominal subcutaneous fat [SAT], adipose tissue, visceral adipose tissue [VAT], and muscle) from patients 5 years before the received a diagnosis of PDAC. METHODS: We collected data from 219 patients with a diagnosis of PDAC (patients) and 657 healthy individuals (controls) from the Rochester Epidemiology Project, from 2000 through 2015. We compared metabolic profiles of patients with those of age- and sex-matched controls, constructing temporal profiles of fasting blood glucose, serum lipids including triglycerides, cholesterol profiles, and body weight and temperature for 60 months before the diagnosis of PDAC (index date). To construct the temporal profile of soft tissue changes, we collected computed tomography scans from 68 patients, comparing baseline (>18 months before diagnosis) areas of SAT, VAT, and muscle at L2/L3 vertebra with those of later scans until time of diagnosis. SAT and VAT, isolated from healthy individuals, were exposed to exosomes isolated from PDAC cell lines and analyzed by RNA sequencing. SAT was collected from KRAS+/LSLG12D P53flox/flox mice with PDACs, C57/BL6 (control) mice, and 5 patients and analyzed by histology and immunohistochemistry. RESULTS: There were no significant differences in metabolic or soft tissue features of patients vs controls until 30 months before PDAC diagnosis. In the 30 to 18 months before PDAC diagnosis (phase 1, hyperglycemia), a significant proportion of patients developed hyperglycemia, compared with controls, without soft tissue changes. In the 18 to 6 months before PDAC diagnosis (phase 2, pre-cachexia), patients had significant increases in hyperglycemia and decreases in serum lipids, body weight, and SAT, with preserved VAT and muscle. In the 6 to 0 months before PDAC diagnosis (phase 3, cachexia), a significant proportion of patients had hyperglycemia compared with controls, and patients had significant reductions in all serum lipids, SAT, VAT, and muscle. We believe the patients had browning of SAT, based on increases in body temperature, starting 18 months before PDAC diagnosis. We observed expression of uncoupling protein 1 (UCP1) in SAT exposed to PDAC exosomes, SAT from mice with PDACs, and SAT from all 5 patients but only 1 of 4 controls. CONCLUSIONS: We identified 3 phases of metabolic and soft tissue changes that precede a diagnosis of PDAC. Loss of SAT starts 18 months before PDAC identification, and is likely due to browning. Overexpression of UCP1 in SAT might be a biomarker of early-stage PDAC, but further studies are needed.

Automated Abdominal Segmentation of CT Scans for Body Composition Analysis Using Deep Learning.

Purpose To develop and evaluate a fully automated algorithm for segmenting the abdomen from CT to quantify body composition. Materials and Methods For this retrospective study, a convolutional neural network based on the U-Net architecture was trained to perform abdominal segmentation on a data set of 2430 two-dimensional CT examinations and was tested on 270 CT examinations. It was further tested on a separate data set of 2369 patients with hepatocellular carcinoma (HCC). CT examinations were performed between 1997 and 2015. The mean age of patients was 67 years; for male patients, it was 67 years (range, 29-94 years), and for female patients, it was 66 years (range, 31-97 years). Differences in segmentation performance were assessed by using two-way analysis of variance with Bonferroni correction. Results Compared with reference segmentation, the model for this study achieved Dice scores (mean ± standard deviation) of 0.98 ± 0.03, 0.96 ± 0.02, and 0.97 ± 0.01 in the test set, and 0.94 ± 0.05, 0.92 ± 0.04, and 0.98 ± 0.02 in the HCC data set, for the subcutaneous, muscle, and visceral adipose tissue compartments, respectively. Performance met or exceeded that of expert manual segmentation. Conclusion Model performance met or exceeded the accuracy of expert manual segmentation of CT examinations for both the test data set and the hepatocellular carcinoma data set. The model generalized well to multiple levels of the abdomen and may be capable of fully automated quantification of body composition metrics in three-dimensional CT examinations. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Chang in this issue.

Effects of Alcohol and Estrogen Receptor Blockade Using ICI 182,780 on Bone in Ovariectomized Rats.

BACKGROUND: Estrogen signaling is essential for the sexual dimorphism of the skeleton, is required for normal bone remodeling balance in adults, and may influence the skeletal response to alcohol. High levels of alcohol consumption lower bone mass in ovary-intact but not ovariectomized (ovx) rats. However, the extremely rapid rate of bone loss immediately following ovx may obscure the effects of alcohol. We therefore determined (i) whether heavy alcohol consumption (35% caloric intake) influences bone in sexually mature ovx rats with established cancellous osteopenia and (ii) whether ICI 182,780 (ICI), a potent estrogen receptor signaling antagonist, alters the skeletal response to alcohol. METHODS: Three weeks following ovx, rats were randomized into 5 groups, (i) baseline, (ii) control + vehicle, (iii) control + ICI, (iv) ethanol (EtOH) + vehicle, or (v) EtOH + ICI, and treated accordingly for 4 weeks. Dual-energy X-ray absorptiometry, microcomputed tomography, blood measurements of markers of bone turnover, and gene expression in femur and uterus were used to evaluate response to alcohol and ICI. RESULTS: Rats consuming alcohol had lower bone mass and increased fat mass. Bone microarchitecture of the tibia and gene expression in femur were altered; specifically, there was reduced accrual of cortical bone, net loss of cancellous bone, and differential expression of 19/84 genes related to bone turnover. Furthermore, osteocalcin, a marker of bone turnover, was lower in alcohol-fed rats. ICI had no effect on weight gain, body composition, or cortical bone. ICI reduced cancellous bone loss and serum CTX-1, a biochemical marker of bone resorption; alcohol antagonized the latter 2 responses. Neither alcohol nor ICI affected uterine weight or gene expression. CONCLUSIONS: Alcohol exaggerated bone loss in ovx rats in the presence or absence of estrogen receptor blockade with ICI. The negligible effect of alcohol on uterus and limited effects of ICI on bone in alcohol-fed ovx rats suggest that estrogen receptor signaling plays a limited role in the action of alcohol on bone in a rat model for chronic alcohol abuse.

MANAGEMENT AND OUTCOMES OF GIANT PROLACTINOMA: A SERIES OF 71 PATIENTS.

Objective: To describe outcomes of patients with giant prolactinoma (≥4 cm) and identify predictors of therapeutic response. Methods: In this retrospective study, complete biochemical and structural response were defined as prolactin (PRL) ≤25 ng/mL and no visible tumor at follow-up, respectively. Results: Giant prolactinoma (median size, 4.8 cm [range, 4 to 9.8 cm]; median PRL, 5,927 ng/mL [range, 120 to 100,000 ng/mL]) was diagnosed in 71 patients. Treatments included: dopamine agonists (DAs) (n = 70, 99%), surgery (n = 30, 42%), radiation (n = 10, 14%), and somatostatin analogs (n = 2, 3%). Patients treated with DA monotherapy were older compared with those who received subsequent therapies (47 years vs. 28 years; P = .003) but had similar initial PRL and tumor size. Surgically managed patients were younger compared with the nonsurgical group (35 years vs. 46 years; P = .02) and had lower initial PRL (3,121 ng/mL vs. 6,920 ng/mL; P = .02), yet they had similar tumor response. Hypopituitarism was more common following surgery compared to medical management: adrenal insufficiency (69% vs. 27%; P<.001), hypothyroidism (67% vs. 38%; P = .02), growth hormone deficiency (24% vs. 6%; P = .04), and diabetes insipidus (17% vs. 3%; P = .04). Therapeutic response did not correlate with sex, age, initial PRL, tumor size, or first-line therapy mode. At median follow-up of 4.8 years, the median PRL was 18.3 ng/mL (range, 0.6 to 12,680 ng/mL), and final volume was 0.9 cm3 (range, 0 to 43.0 cm3). In those with available data, 36/65 (55%) patients achieved PRL normalization, and 16/61 (26%) had no visible tumor at follow-up. Conclusion: Most patients with giant prolactinoma have excellent response to DA. Sex, age, initial PRL, and tumor size do not predict therapeutic response. Abbreviations: BRC = bromocriptine; CAB = cabergoline; CSF = cerebrospinal fluid; DA = dopamine agonist; MRI = magnetic resonance imaging; PRL = prolactin.

RIL-Contour: a Medical Imaging Dataset Annotation Tool for and with Deep Learning.

Deep-learning algorithms typically fall within the domain of supervised artificial intelligence and are designed to "learn" from annotated data. Deep-learning models require large, diverse training datasets for optimal model convergence. The effort to curate these datasets is widely regarded as a barrier to the development of deep-learning systems. We developed RIL-Contour to accelerate medical image annotation for and with deep-learning. A major goal driving the development of the software was to create an environment which enables clinically oriented users to utilize deep-learning models to rapidly annotate medical imaging. RIL-Contour supports using fully automated deep-learning methods, semi-automated methods, and manual methods to annotate medical imaging with voxel and/or text annotations. To reduce annotation error, RIL-Contour promotes the standardization of image annotations across a dataset. RIL-Contour accelerates medical imaging annotation through the process of annotation by iterative deep learning (AID). The underlying concept of AID is to iteratively annotate, train, and utilize deep-learning models during the process of dataset annotation and model development. To enable this, RIL-Contour supports workflows in which multiple-image analysts annotate medical images, radiologists approve the annotations, and data scientists utilize these annotations to train deep-learning models. To automate the feedback loop between data scientists and image analysts, RIL-Contour provides mechanisms to enable data scientists to push deep newly trained deep-learning models to other users of the software. RIL-Contour and the AID methodology accelerate dataset annotation and model development by facilitating rapid collaboration between analysts, radiologists, and engineers.

What Does Deep Learning See? Insights From a Classifier Trained to Predict Contrast Enhancement Phase From CT Images.

OBJECTIVE: Deep learning has shown great promise for improving medical image classification tasks. However, knowing what aspects of an image the deep learning system uses or, in a manner of speaking, sees to make its prediction is difficult. MATERIALS AND METHODS: Within a radiologic imaging context, we investigated the utility of methods designed to identify features within images on which deep learning activates. In this study, we developed a classifier to identify contrast enhancement phase from whole-slice CT data. We then used this classifier as an easily interpretable system to explore the utility of class activation map (CAMs), gradient-weighted class activation maps (Grad-CAMs), saliency maps, guided backpropagation maps, and the saliency activation map, a novel map reported here, to identify image features the model used when performing prediction. RESULTS: All techniques identified voxels within imaging that the classifier used. SAMs had greater specificity than did guided backpropagation maps, CAMs, and Grad-CAMs at identifying voxels within imaging that the model used to perform prediction. At shallow network layers, SAMs had greater specificity than Grad-CAMs at identifying input voxels that the layers within the model used to perform prediction. CONCLUSION: As a whole, voxel-level visualizations and visualizations of the imaging features that activate shallow network layers are powerful techniques to identify features that deep learning models use when performing prediction.

BCL11B regulates sutural patency in the mouse craniofacial skeleton.

The transcription factor BCL11B plays essential roles during development of the immune, nervous, and cutaneous systems. Here we show that BCL11B is expressed in both osteogenic and sutural mesenchyme of the developing craniofacial complex. Bcl11b(-/-) mice exhibit increased proliferation of osteoprogenitors, premature osteoblast differentiation, and enhanced skull mineralization leading to synostoses of facial and calvarial sutures. Ectopic expression of Fgfr2c, a gene implicated in craniosynostosis in mice and humans, and that of Runx2 was detected within the affected sutures of Bcl11b(-/-) mice. These data suggest that ectopic expression of Fgfr2c in the sutural mesenchyme, without concomitant changes in the expression of FGF ligands, appears to induce the RUNX2-dependent osteogenic program and craniosynostosis in Bcl11b(-/-) mice.

Toolkits and Libraries for Deep Learning.

Deep learning is an important new area of machine learning which encompasses a wide range of neural network architectures designed to complete various tasks. In the medical imaging domain, example tasks include organ segmentation, lesion detection, and tumor classification. The most popular network architecture for deep learning for images is the convolutional neural network (CNN). Whereas traditional machine learning requires determination and calculation of features from which the algorithm learns, deep learning approaches learn the important features as well as the proper weighting of those features to make predictions for new data. In this paper, we will describe some of the libraries and tools that are available to aid in the construction and efficient execution of deep learning as applied to medical images.

Docosahexaenoic acid attenuates hepatic inflammation, oxidative stress, and fibrosis without decreasing hepatosteatosis in a Ldlr(-/-) mouse model of western diet-induced nonalcoholic steatohepatitis.

The incidence of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) has increased in parallel with the incidence of obesity. While both NAFLD and NASH are characterized by hepatosteatosis, NASH is characterized by hepatic damage, inflammation, oxidative stress, and fibrosis. We previously reported that feeding Ldlr(-/-) mice a high-fat, high-cholesterol diet containing menhaden oil attenuated several markers of NASH, including hepatosteatosis, inflammation, and fibrosis. Herein, we test the hypothesis that DHA [22:6 (n-3)] is more effective than EPA [20:5 (n-3)] at preventing Western diet (WD)-induced NASH in Ldlr(-/-) mice. Mice were fed the WD supplemented with either olive oil (OO), EPA, DHA, or EPA + DHA for 16 wk. WD + OO feeding induced a severe NASH phenotype, characterized by robust hepatosteatosis, inflammation, oxidative stress, and fibrosis. Whereas none of the C20-22 (n-3) fatty acid treatments prevented WD-induced hepatosteatosis, all 3 (n-3) PUFA-containing diets significantly attenuated WD-induced inflammation, fibrosis, and hepatic damage. The capacity of dietary DHA to suppress hepatic markers of inflammation (Clec4F, F4/80, Trl4, Trl9, CD14, Myd88), fibrosis (Procol1α1, Tgfß1), and oxidative stress (NADPH oxidase subunits Nox2, p22phox, p40phox, p47phox, p67phox) was significantly greater than dietary EPA. The effects of DHA on these markers paralleled DHA-mediated suppression of hepatic Fads1 mRNA abundance and hepatic arachidonic acid content. Because DHA suppression of NASH markers does not require a reduction in hepatosteatosis, dietary DHA may be useful in combating NASH in obese humans.

MicroRNAs control neurobehavioral development and function in zebrafish.

microRNAs (miRNAs) have emerged as regulators of a broad spectrum of neurodevelopmental processes, including brain morphogenesis, neuronal differentiation, and survival. While the role of miRNAs in establishing and maintaining the developing nervous system is widely appreciated, the developmental neurobehavioral role of miRNAs has yet to be defined. Here we show that transient disruption of brain morphogenesis by ethanol exposure results in behavioral hyperactivity in larval zebrafish challenged with changes in lighting conditions. Aberrations in swimming activity persist in juveniles that were developmentally exposed to ethanol. During early neurogenesis, multiple gene expression profiling studies revealed widespread changes in mRNA and miRNA abundance in ethanol-exposed embryos. Consistent with a role for miRNAs in neurobehavioral development, target prediction analyses identified multiple miRNAs misexpressed in the ethanol-exposed cohorts that were also predicted to target inversely expressed transcripts known to influence brain morphogenesis. In vivo knockdown of miR-9/9* or miR-153c persistently phenocopied the effect of ethanol on larval and juvenile swimming behavior. Structural analyses performed on adults showed that repression of miR-153c during development impacts craniofacial skeletal development. Together, these data support an integral role for miRNAs in the establishment of vertebrate neurobehavioral and skeletal systems.

Higher weight in partially leptin-resistant db/+ mice is associated with positive effects on bone.

Absence of leptin confers metabolic dysfunction resulting in morbid obesity. Bone growth and maturation are also impaired. Partial leptin resistance is more common than leptin deficiency and, when induced by feeding mice a high fat diet, often has a negative effect on bone. Here, we used a genetic model to investigate the skeletal effects of partial and total leptin resistance in mice. This was accomplished by comparing the skeletal phenotypes of 17-week-old female C57Bl6/J wild-type (WT) mice, partial leptin receptor-deficient (db/+) mice and leptin receptor-deficient (db/db) mice (n = 7-8/group), all fed a standard diet. Compared to WT mice, db/db mice were dramatically heavier and hyperleptinemic. These mice were also hypogonadal, hyperglycemic, osteopenic and had lower serum levels of bone turnover markers, osteocalcin and C-terminal telopeptide of type I collagen (CTX). Compared to WT mice, db/+ mice were 14% heavier, had 149% more abdominal white adipose tissue, and were mildly hyperglycemic. db/+ mice did not differ from WT mice in uterine weight or serum levels of markers of bone turnover, although there was a trend for lower osteocalcin. At the bone microarchitectural level, db/+ mice differed from WT mice in having more massive femurs and a trend (P = 0.072) for larger vertebrae. These findings suggest that db/+ mice fed a normal mouse diet compensate for partial leptin resistance by increasing white adipose tissue mass which results in higher leptin levels. Our findings suggest that db/+ mice are a useful diet-independent model for studying the effects of partial leptin resistance on the skeleton.

Severe Zinc Deficiency Impairs Accrual of Bone in Rapidly Growing Rats That Is Partially Corrected Following Short-term Zinc Repletion.

Zinc (Zn) deficiency impairs bone growth. However, the precise skeletal effects of varying levels of Zn deficiency and response to subsequent Zn repletion on the growing skeleton are incompletely understood. To address this gap in knowledge, we investigated the effects of dietary Zn ((severe deficiency (< 0.5 mg Zn/kg diet) and short-term Zn repletion (30 mg/kg diet), marginal deficiency (6 mg Zn/kg diet)) on bone mass, density, and cortical and cancellous bone microarchitecture in growing male Sprague Dawley rats. Marginal Zn intake for 42 days had no effect on bone mass or cortical and cancellous bone microarchitecture. Twenty-one days of severe Zn deficiency lowered serum osteocalcin and C terminal telopeptide of type I collagen (CTX-1), decreased tibial bone mineral content and density, and lowered cross-sectional volume, cortical volume, and cortical thickness in tibial diaphysis as compared to both Zn-adequate (30 mg/kg diet) and pair-fed controls. Severe Zn deficiency similarly lowered cancellous bone volume in proximal tibial metaphysis. Zn repletion (10 days) accelerated weight gain, indicative of catch-up growth, normalized CTX-1 and osteocalcin, but did not normalize bone mass (unadjusted and adjusted for body weight) or cortical and cancellous bone microarchitecture. In summary, severe but not marginal Zn deficiency in rapidly growing rats impaired acquisition of cortical and cancellous bone, resulting in abnormalities in bone microarchitecture. Zn repletion accelerated weight gain compared to Zn-adequate controls but absence of a compensatory increase in serum osteocalcin or bone mass suggests Zn repletion may be insufficient to fully counteract the detrimental effects of prior Zn deficiency on skeletal growth.

Association between computerized tomography (CT) study of body composition and severity of acute pancreatitis: Use of a novel Z-score supports obesity paradox.

BACKGROUND & AIMS: The association between body composition parameters measured on computed tomography (CT) and severity of acute pancreatitis (AP) is conflicting because these composition parameters vary considerably by sex and age. We previously developed normative body composition data, in healthy subjects. Z-score calculated from the normative data gives age and sex adjusted body composition parameters. We studied the above association using this novel Z-score in a large cohort of patients with AP. METHODS: Between January 2014 and March 2018, patients admitted with AP and had CT scans within a week of admission, were enrolled. Body composition data including skeletal muscle (SM), subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) were calculated from the CT scan using deep learning automated algorithm. Then we converted the value to Z-score, and then compared the same score between mild AP, moderately severe AP and severe AP defined by revised Atlanta criteria. RESULTS: Out of 514 patients, 336 (65.4%) are mild AP, 130 (25.3%) moderately severe AP, and 48 (9.3%) severe AP. Patients with moderately severe AP had significantly lower SM-z-score than those with mild AP (1.21 vs1.73, p = 0.048) and patients with severe AP had significantly lower SAT-z-score than those with mild AP (0.70 vs.1.29, p = 0.016). VAT-z-score was not significantly different between three groups. (p = 0.76). CONCLUSION: Lower SM-z-score and SAT-z-score were associated with moderately severe and severe types of AP, respectively. Future prospective studies in patients with AP using Z-scores, may define the association between body composition and severity of AP, and explain the inconsistencies reported in previous studies.

SOUP-GAN: Super-Resolution MRI Using Generative Adversarial Networks.

There is a growing demand for high-resolution (HR) medical images for both clinical and research applications. Image quality is inevitably traded off with acquisition time, which in turn impacts patient comfort, examination costs, dose, and motion-induced artifacts. For many image-based tasks, increasing the apparent spatial resolution in the perpendicular plane to produce multi-planar reformats or 3D images is commonly used. Single-image super-resolution (SR) is a promising technique to provide HR images based on deep learning to increase the resolution of a 2D image, but there are few reports on 3D SR. Further, perceptual loss is proposed in the literature to better capture the textural details and edges versus pixel-wise loss functions, by comparing the semantic distances in the high-dimensional feature space of a pre-trained 2D network (e.g., VGG). However, it is not clear how one should generalize it to 3D medical images, and the attendant implications are unclear. In this paper, we propose a framework called SOUP-GAN: Super-resolution Optimized Using Perceptual-tuned Generative Adversarial Network (GAN), in order to produce thinner slices (e.g., higher resolution in the 'Z' plane) with anti-aliasing and deblurring. The proposed method outperforms other conventional resolution-enhancement methods and previous SR work on medical images based on both qualitative and quantitative comparisons. Moreover, we examine the model in terms of its generalization for arbitrarily user-selected SR ratios and imaging modalities. Our model shows promise as a novel 3D SR interpolation technique, providing potential applications for both clinical and research applications.