Uncertainty Estimation and Out-of-Distribution Detection for Deep Learning-Based Image Reconstruction using the Local Lipschitz.

Accurate image reconstruction is at the heart of diagnostics in medical imaging. Supervised deep learning-based approaches have been investigated for solving inverse problems including image reconstruction. However, these trained models encounter unseen data distributions that are widely shifted from training data during deployment. Therefore, it is essential to assess whether a given input falls within the training data distribution. Current uncertainty estimation approaches focus on providing an uncertainty map to radiologists, rather than assessing the training distribution fit. In this work, we propose a method based on the local Lipschitz metric to distinguish out-of-distribution images from in-distribution with an area under the curve of 99.94% for True Positive Rate versus False Positive Rate. We demonstrate a very strong relationship between the local Lipschitz value and mean absolute error (MAE), supported by a Spearman's rank correlation coefficient of 0.8475, to determine an uncertainty estimation threshold for optimal performance. Through the identification of false positives, we demonstrate the local Lipschitz and MAE relationship can guide data augmentation and reduce uncertainty. Our study was validated using the AUTOMAP architecture for sensor-to-image Magnetic Resonance Imaging (MRI) reconstruction. We demonstrate our approach outperforms baseline techniques of Monte-Carlo dropout and deep ensembles as well as the state-of-the-art Mean Variance Estimation (MVE) network approach. We expand our application scope to MRI denoising and Computed Tomography (CT) sparse-to-full view reconstructions using UNET architectures. We show our approach is applicable to various architectures and applications, especially in medical imaging, where preserving diagnostic accuracy of reconstructed images remains paramount.

NPSV-deep: a deep learning method for genotyping structural variants in short read genome sequencing data.

MOTIVATION: Structural variants (SVs) play a causal role in numerous diseases but can be difficult to detect and accurately genotype (determine zygosity) with short-read genome sequencing data (SRS). Improving SV genotyping accuracy in SRS data, particularly for the many SVs first detected with long-read sequencing, will improve our understanding of genetic variation. RESULTS: NPSV-deep is a deep learning-based approach for genotyping previously reported insertion and deletion SVs that recasts this task as an image similarity problem. NPSV-deep predicts the SV genotype based on the similarity between pileup images generated from the actual SRS data and matching SRS simulations. We show that NPSV-deep consistently matches or improves upon the state-of-the-art for SV genotyping accuracy across different SV call sets, samples and variant types, including a 25% reduction in genotyping errors for the Genome-in-a-Bottle (GIAB) high-confidence SVs. NPSV-deep is not limited to the SVs as described; it improves deletion genotyping concordance a further 1.5 percentage points for GIAB SVs (92%) by automatically correcting imprecise/incorrectly described SVs. AVAILABILITY AND IMPLEMENTATION: Python/C++ source code and pre-trained models freely available at https://github.com/mlinderm/npsv2.

Understanding anxiety through uncertainty quantification.

Uncertainty has been a central concept in psychological theories of anxiety. However, this concept has been plagued by divergent connotations and operationalizations. The lack of consensus hinders the current search for cognitive and biological mechanisms of anxiety, jeopardizes theory creation and comparison, and restrains translation of basic research into improved diagnoses and interventions. Drawing upon uncertainty decomposition in Bayesian Decision Theory, we propose a well-defined conceptual structure of uncertainty in cognitive and clinical sciences, with a focus on anxiety. We discuss how this conceptual structure provides clarity and can be naturally applied to existing frameworks of psychopathology research. Furthermore, it allows formal quantification of various types of uncertainty that can benefit both research and clinical practice in the era of computational psychiatry.

Incident dementia and long-term exposure to constituents of fine particle air pollution: A national cohort study in the United States.

Growing evidence suggests that fine particulate matter (PM2.5) likely increases the risks of dementia, yet little is known about the relative contributions of different constituents. Here, we conducted a nationwide population-based cohort study (2000 to 2017) by integrating the Medicare Chronic Conditions Warehouse database and two independently sourced datasets of high-resolution PM2.5 major chemical composition, including black carbon (BC), organic matter (OM), nitrate (NO3-), sulfate (SO42-), ammonium (NH4+), and soil dust (DUST). To investigate the impact of long-term exposure to PM2.5 constituents on incident all-cause dementia and Alzheimer's disease (AD), hazard ratios for dementia and AD were estimated using Cox proportional hazards models, and penalized splines were used to evaluate potential nonlinear concentration-response (C-R) relationships. Results using two exposure datasets consistently indicated higher rates of incident dementia and AD for an increased exposure to PM2.5 and its major constituents. An interquartile range increase in PM2.5 mass was associated with a 6 to 7% increase in dementia incidence and a 9% increase in AD incidence. For different PM2.5 constituents, associations remained significant for BC, OM, SO42-, and NH4+ for both end points (even after adjustments of other constituents), among which BC and SO42- showed the strongest associations. All constituents had largely linear C-R relationships in the low exposure range, but most tailed off at higher exposure concentrations. Our findings suggest that long-term exposure to PM2.5 is significantly associated with higher rates of incident dementia and AD and that SO42-, BC, and OM related to traffic and fossil fuel combustion might drive the observed associations.

A Cross-validated Ensemble Approach to Robust Hypothesis Testing of Continuous Nonlinear Interactions: Application to Nutrition-Environment Studies.

Gene-environment and nutrition-environment studies often involve testing of high-dimensional interactions between two sets of variables, each having potentially complex nonlinear main effects on an outcome. Construction of a valid and powerful hypothesis test for such an interaction is challenging, due to the difficulty in constructing an efficient and unbiased estimator for the complex, nonlinear main effects. In this work we address this problem by proposing a Cross-validated Ensemble of Kernels (CVEK) that learns the space of appropriate functions for the main effects using a cross-validated ensemble approach. With a carefully chosen library of base kernels, CVEK flexibly estimates the form of the main-effect functions from the data, and encourages test power by guarding against over-fitting under the alternative. The method is motivated by a study on the interaction between metal exposures in utero and maternal nutrition on children's neurodevelopment in rural Bangladesh. The proposed tests identified evidence of an interaction between minerals and vitamins intake and arsenic and manganese exposures. Results suggest that the detrimental effects of these metals are most pronounced at low intake levels of the nutrients, suggesting nutritional interventions in pregnant women could mitigate the adverse impacts of in utero metal exposures on children's neurodevelopment.

Health effects of air pollutant mixtures on overall mortality among the elderly population using Bayesian kernel machine regression (BKMR).

It is well documented that fine particles matter (PM2.5), ozone (O3), and nitrogen dioxide (NO2) are associated with a range of adverse health outcomes. However, most epidemiologic studies have focused on understanding their additive effects, despite that individuals are exposed to multiple air pollutants simultaneously that are likely correlated with each other. Therefore, we applied a novel method - Bayesian Kernel machine regression (BKMR) and conducted a population-based cohort study to assess the individual and joint effect of air pollutant mixtures (PM2.5, O3, and NO2) on all-cause mortality among the Medicare population in 15 cities with 656 different ZIP codes in the southeastern US. The results suggest a strong association between pollutant mixture and all-cause mortality, mainly driven by PM2.5. The positive association of PM2.5 with mortality appears stronger at lower percentiles of other pollutants. An interquartile range change in PM2.5 concentration was associated with a significant increase in mortality of 1.7 (95% CI: 0.5, 2.9), 1.6 (95% CI: 0.4, 2.7) and 1.4 (95% CI: 0.1, 2.6) standard deviations (SD) when O3 and NO2 were set at the 25th, 50th, and 75th percentiles, respectively. BKMR analysis did not identify statistically significant interactions among PM2.5, O3, and NO2. However, since the small sub-population might weaken the study power, additional studies (in larger sample size and other regions in the US) are in need to reinforce the current finding.

Immune Markers and Their Association with Bone Density in Children, Adolescents, and Young Adults with Perinatally Acquired HIV.

To describe distributions of immune markers in children and young adults by sex and HIV status, and within groups, investigate associations of immune markers with bone density across Tanner stage. Using data and samples from 353 participants in a cross-sectional study in youth with perinatally acquired HIV (PHIV) and matched HIV-negative controls, distributions of inflammation and activation immune markers were described by sex and HIV status. Correlations and structural equation models (SEM) were used to explore marginal and multivariable associations of the immune markers with bone density and to assess whether patterns of association varied by sex and HIV status. Immune marker distributions did not differ by sex, but there were some differences by HIV status. Correlation patterns among bone, body composition, and immune markers were similar across the sex and HIV status groups. Conclusions from SEMs were limited by small sample sizes, but there was some indication that patterns of association between bone density and certain immune markers differed in male PHIV with more advanced Tanner stage compared to the other three groups. In conclusion, distributions of bone density, body composition, and immune markers may vary by sex and HIV status, although associations among these outcomes within sex and HIV status groups appear similar. Bone density of male PHIV appears to be more negatively affected than females, regardless of female HIV status. Larger longitudinal studies across Tanner stages are needed to further explore potential biological relationships between immune markers and bone density in youth living with HIV.

Image reconstruction by domain-transform manifold learning.

Image reconstruction is essential for imaging applications across the physical and life sciences, including optical and radar systems, magnetic resonance imaging, X-ray computed tomography, positron emission tomography, ultrasound imaging and radio astronomy. During image acquisition, the sensor encodes an intermediate representation of an object in the sensor domain, which is subsequently reconstructed into an image by an inversion of the encoding function. Image reconstruction is challenging because analytic knowledge of the exact inverse transform may not exist a priori, especially in the presence of sensor non-idealities and noise. Thus, the standard reconstruction approach involves approximating the inverse function with multiple ad hoc stages in a signal processing chain, the composition of which depends on the details of each acquisition strategy, and often requires expert parameter tuning to optimize reconstruction performance. Here we present a unified framework for image reconstruction-automated transform by manifold approximation (AUTOMAP)-which recasts image reconstruction as a data-driven supervised learning task that allows a mapping between the sensor and the image domain to emerge from an appropriate corpus of training data. We implement AUTOMAP with a deep neural network and exhibit its flexibility in learning reconstruction transforms for various magnetic resonance imaging acquisition strategies, using the same network architecture and hyperparameters. We further demonstrate that manifold learning during training results in sparse representations of domain transforms along low-dimensional data manifolds, and observe superior immunity to noise and a reduction in reconstruction artefacts compared with conventional handcrafted reconstruction methods. In addition to improving the reconstruction performance of existing acquisition methodologies, we anticipate that AUTOMAP and other learned reconstruction approaches will accelerate the development of new acquisition strategies across imaging modalities.

Use of a Digital Health Application for Influenza Surveillance in China.

OBJECTIVES: To examine whether a commercial digital health application could support influenza surveillance in China. METHODS: We retrieved data from the Thermia online and mobile educational tool, which allows parents to monitor their children's fever and infectious febrile illnesses including influenza. We modeled monthly aggregated influenza-like illness case counts from Thermia users over time and compared them against influenza monthly case counts obtained from the National Health and Family Planning Commission of the People's Republic of China by using time series regression analysis. We retrieved 44 999 observations from January 2014 through July 2016 from Thermia China. RESULTS: Thermia appeared to predict influenza outbreaks 1 month earlier than the National Health and Family Planning Commission influenza surveillance system (P = .046). Being younger, not having up-to-date immunizations, and having an underlying health condition were associated with participant-reported influenza-like illness. CONCLUSIONS: Digital health applications could supplement traditional influenza surveillance systems in China by providing access to consumers' symptom reporting. Growing popularity and use of commercial digital health applications in China potentially affords opportunities to support disease detection and monitoring and rapid treatment mobilization.