Explore the vast range of titles available.

With age, the prevalence of diseases such as fatty liver disease, cirrhosis, and type two diabetes increases. Approaches to both predict abdominal age and identify risk factors for accelerated abdominal age may ultimately lead to advances that will delay the onset of these diseases. We build an abdominal age predictor by training convolutional neural networks to predict abdominal age (or “AbdAge”) from 45,552 liver magnetic resonance images [MRIs] and 36,784 pancreas MRIs (R-Squared = 73.3 ± 0.6; mean absolute error = 2.94 ± 0.03 years). Attention maps show that the prediction is driven by both liver and pancreas anatomical features, and surrounding organs and tissue. Abdominal aging is a complex trait, partially heritable (h_g 2  = 26.3 ± 1.9%), and associated with 16 genetic loci (e.g. in  PLEKHA1 and EFEMP1 ), biomarkers (e.g body impedance), clinical phenotypes (e.g, chest pain), diseases (e.g. hypertension), environmental (e.g smoking), and socioeconomic (e.g education, income) factors. Approaches to both determine abdominal age and identify risk factors for accelerated abdominal age will help delay the onset of several diseases. Here, the authors build an abdominal age predictor by training convolutional neural networks to predict abdominal age from liver and pancreas MRIs.

Physical activity improves quality of life and protects against age-related diseases. With age, physical activity tends to decrease, increasing vulnerability to disease in the elderly. In the following, we trained a neural network to predict age from 115,456 one week-long 100Hz wrist accelerometer recordings from the UK Biobank (mean absolute error = 3.7±0.2 years), using a variety of data structures to capture the complexity of real-world activity. We achieved this performance by preprocessing the raw frequency data as 2,271 scalar features, 113 time series, and four images. We defined accelerated aging for a participant as being predicted older than one’s actual age and identified both genetic and environmental exposure factors associated with the new phenotype. We performed a genome wide association on the accelerated aging phenotypes to estimate its heritability (h_g 2 = 12.3±0.9%) and identified ten single nucleotide polymorphisms in close proximity to genes in a histone and olfactory cluster on chromosome six (e.g HIST1H1C , OR5V1 ). Similarly, we identified biomarkers (e.g blood pressure), clinical phenotypes (e.g chest pain), diseases (e.g hypertension), environmental (e.g smoking), and socioeconomic (e.g income and education) variables associated with accelerated aging. Physical activity-derived biological age is a complex phenotype associated with both genetic and non-genetic factors.

Physical activity improves quality of life and protects against age-related diseases. With age, physical activity tends to decrease, increasing vulnerability to disease in the elderly. In the following, we trained a neural network to predict age from 115,456 one week-long 100Hz wrist accelerometer recordings from the UK Biobank (mean absolute error = 3.7±0.2 years), using a variety of data structures to capture the complexity of real-world activity. We achieved this performance by preprocessing the raw frequency data as 2,271 scalar features, 113 time series, and four images. We defined accelerated aging for a participant as being predicted older than one’s actual age and identified both genetic and environmental exposure factors associated with the new phenotype. We performed a genome wide association on the accelerated aging phenotypes to estimate its heritability (h_g2 = 12.3±0.9%) and identified ten single nucleotide polymorphisms in close proximity to genes in a histone and olfactory cluster on chromosome six (e.g HIST1H1C, OR5V1). Similarly, we identified biomarkers (e.g blood pressure), clinical phenotypes (e.g chest pain), diseases (e.g hypertension), environmental (e.g smoking), and socioeconomic (e.g income and education) variables associated with accelerated aging. Physical activity-derived biological age is a complex phenotype associated with both genetic and non-genetic factors. Author summary Physical activity improves quality of life and is also an important protective factor for prevalent age-related diseases and outcomes, such as diabetes and mortality. With age, physical activity tends to decrease, increasing vulnerability to disease in the elderly. Does physical activity measured from digital health devices predict one’s biological age? Biological age, as contrast to chronological age (the time that has elapsed since birth), is an indicator of the biological changes that accrue through time that are hypothesized to be one causal factor for age-related diseases. In the following, we trained machine learning models to predict age from 115,456 one week-long wrist accelerometer recordings from participants of the UK Biobank. We then found genetic, environmental, and behavioral factors associated with accelerated age, the difference between biological and chronological age, adding to the evidence of the biological plausibility of our new predictor. If reversable, summarizing complex physical activity into a biological age predictor may be a way of observing the effect of preventative efforts in real-time.

Deep learning methods have become very popular for the processing of natural images, and were then successfully adapted to the neuroimaging field. As these methods are non-transparent, interpretability methods are needed to validate them and ensure their reliability. Indeed, it has been shown that deep learning models may obtain high performance even when using irrelevant features, by exploiting biases in the training set. Such undesirable situations can potentially be detected by using interpretability methods. Recently, many methods have been proposed to interpret neural networks. However, this domain is not mature yet. Machine learning users face two major issues when aiming to interpret their models: which method to choose, and how to assess its reliability? Here, we aim at providing answers to these questions by presenting the most common interpretability methods and metrics developed to assess their reliability, as well as their applications and benchmarks in the neuroimaging context. Note that this is not an exhaustive survey: we aimed to focus on the studies which we found to be the most representative and relevant.

Vaccination strategies aimed at maturing broadly neutralizing antibodies (bnAbs) from naïve precursors are hindered by unusual features that characterize these Abs, including insertions and deletions (indels). Longitudinal studies of natural HIV infection cases shed light on the complex processes underlying bnAb development and have suggested a role for superinfection as a potential enhancer of neutralization breadth. Here we describe the development of a potent bnAb lineage that was elicited by two founder viruses to inform vaccine design. The V3-glycan targeting bnAb lineage (PC39-1) was isolated from subtype C-infected IAVI Protocol C elite neutralizer, donor PC39, and is defined by the presence of multiple independent insertions in CDRH1 that range from 1-11 amino acids in length. Memory B cell members of this lineage are predominantly atypical in phenotype yet also span the class-switched and antibody-secreting cell compartments. Development of neutralization breadth occurred concomitantly with extensive recombination between founder viruses before each virus separated into two distinct population “arms” that evolved independently to escape the PC39-1 lineage. Ab crystal structures show an extended CDRH1 that can help stabilize the CDRH3. Overall, these findings suggest that early exposure of the humoral system to multiple related Env molecules could promote the induction of bnAbs by focusing Ab responses to conserved epitopes.

Measurements of particulate organic carbon (POC) are critical for understanding the ocean carbon cycle, including biogenic particle formation and removal processes, and for constraining models of carbon cycling at local, regional, and global scales. Despite the importance and ubiquity of POC measurements, discrepancies in methods across platforms and users, necessary to accommodate a multitude of needs and logistical constraints, commonly result in disparate results. Considerations of filter type and pore size, sample volume, collection method, and contamination sources underscore the potential for dissimilar measurements of the same variable assessed using similar and different approaches. During the NASA EXport Processes in the Ocean from RemoTe Sensing (EXPORTS) 2018 field campaign in the North Pacific Ocean, multiple methodologies and sampling approaches for determining POC were applied, including surface inline flow-through systems and depth profiles using Niskin bottles, in situ pumps, and Marine Snow Catchers. A comparison of results from each approach and platform often resulted in significant differences. Supporting measurements, however, provided the means to normalize results across datasets. Using knowledge of contrasting protocols and synchronous or near-synchronous measurements of associated environmental variables, we were able to reconcile dataset differences to account for undersampling of some particle types and sizes, possible sample contamination and blank corrections. These efforts resulted in measurement agreement between initially contrasting datasets and insights on long-acknowledged but rarely resolved discrepancies among contrasting methods for assessing POC concentrations in the ocean.

Objectives: The optimal treatment approach to clavicle fractures in adolescents remains an area of significant controversy. The purpose of this study was to review the demographic characteristics, treatment approaches, and complications in a large series of adolescent clavicle fractures receiving operative and non-operative treatment. Methods: Radiographic and medical record review was conducted for all cases of patients ages 10-18 years-old who presented to a single tertiary care children’s hospital between 2003-2012 with a mid-diaphyseal clavicle fracture. Demographic data, radiographic features, such as fracture pattern, operative details when applicable, and post-treatment clinical course was analyzed, including the reported time to healing and any known complications. Results: Out of 641 cases reviewed (79% male; mean age 14.3 years), 408 (64%) fractures were sustained during sports, most frequently football (25%), hockey (18%), soccer (12%), snowboarding (12%) and skiing (9%). Other common mechanisms of injury were falls sustained outside of athletic activity (19%) and motor vehicle accidents (5%), with similar distribution of mechanism and similar rates of associated injuries seen within the operative (5%) and non-operative (6%) treatment groups. Greater numbers of clavicle fractures were seen annually over the study period. Among the overall cohort, 82% were treated non-operatively, while 18% were treated surgically, with increasing percentage of patients undergoing surgery over the course of the study period. The mean age was higher in the operative group (15.5 years) than the nonoperative group (14.1 years)(p<0.001). Fifty-eight documented complications occurred in 46 patients (7.2%), were significantly more common in the operative (16%) group than the non-operative (5%) group (p<0.001), and were more common in older patients (p=0.007). Only 1 case of nonunion occurred in each treatment group (p=0.56). The rate of symptomatic implants was 13% in the operative group (leading to plate removal in 9% cases), while the rate of symptomatic malunion was 2% in the nonoperative group. Refracture was significantly more common in the nonoperative group (3%) than the operative group (2%) (p=0.03). Refracture in the non-operative group most commonly occurred in the period before complete healing had occurred. Of the 2 cases of refracture in the operative group, 1 case was a peri-implant fracture and 1 case occurred over 1 year following plate removal. No infections were reported in either group. One of the nonoperative symptomatic malunion patients developed thoracic outlet syndrome requiring osteotomy, which led to symptom resolution. One of the operative patients developed contralateral recurrent laryngeal and hypoglossal neuropraxia (Tapia’s syndrome), causing vocal cord paralysis, tongue deviation, and hoarseness, with near complete resolution at the time of most recent follow up, four months post-operatively. Conclusion: Greater numbers of clavicle fractures are being seen in the adolescent population, with over 60% of cases occurring during sports and an increasing trend towards operative treatment in recent years. Nonunion and symptomatic malunion are rare in adolescents. While refracture is more common following nonoperative treatment, overall complication rates appear to be more common following operative management, the most common of which is symptomatic implants.

Hands are the primary means through which humans interact with the world. Reliable and always-available hand pose inference could yield new and intuitive control schemes for human-computer interactions, particularly in virtual and augmented reality. Computer vision is effective but requires one or multiple cameras and can struggle with occlusions, limited field of view, and poor lighting. Wearable wrist-based surface electromyography (sEMG) presents a promising alternative as an always-available modality sensing muscle activities that drive hand motion. However, sEMG signals are strongly dependent on user anatomy and sensor placement, and existing sEMG models have required hundreds of users and device placements to effectively generalize. To facilitate progress on sEMG pose inference, we introduce the emg2pose benchmark, the largest publicly available dataset of high-quality hand pose labels and wrist sEMG recordings. emg2pose contains 2kHz, 16 channel sEMG and pose labels from a 26-camera motion capture rig for 193 users, 370 hours, and 29 stages with diverse gestures - a scale comparable to vision-based hand pose datasets. We provide competitive baselines and challenging tasks evaluating real-world generalization scenarios: held-out users, sensor placements, and stages. emg2pose provides the machine learning community a platform for exploring complex generalization problems, holding potential to significantly enhance the development of sEMG-based human-computer interactions.

The composition of the marine phytoplankton community has been shown to impact many biogeochemical processes and marine ecosystem services. A variety of methods exist to characterize phytoplankton community composition (PCC), with varying degrees of taxonomic resolution. Accordingly, the resulting PCC determinations are dependent on the method used. Here, we use surface ocean samples collected in the North Atlantic and North Pacific Oceans to compare high performance liquid chromatography (HPLC) pigment-based PCC to four other methods: quantitative cell imaging, flow cytometry, and 16S and 18S rRNA amplicon sequencing. These methods allow characterization of both prokaryotic and eukaryotic PCC across a wide range of size classes. PCC estimates of many taxa resolved at the class level (e.g., diatoms) show strong positive correlations across methods, while other groups (e.g., dinoflagellates) are not well captured by one or more methods. Since variations in phytoplankton pigment concentrations are related to changes in optical properties, this combined dataset expands the potential scope of ocean color remote sensing by associating PCC at the genus- and species-level with group- or class-level PCC from pigments. Quantifying the strengths and limitations of pigment-based PCC methods compared to PCC assessments from amplicon sequencing, imaging, and cytometry methods is the first step toward the robust validation of remote sensing approaches to quantify PCC from space.