Characterizing the Progression of Pulmonary Edema Severity: Can Pairwise Comparisons in Radiology Reports Help?

Accurately assessing the severity of pulmonary edema is critical for making treatment decisions in congestive heart failure patients. However, the current scale for quantifying pulmonary edema based on chest radiographs does not have well-characterized severity levels, with substantial inter-radiologist disagreement. In this study, we investigate whether comparisons documented in radiology reports can provide accurate characterizations of pulmonary edema progression. We propose a rules-based natural language processing approach to assess the change in a patient's pulmonary edema status (e.g. better, worse, no change) by performing pairwise comparisons of consecutive radiology reports, using regular expressions and heuristics derived from clinical knowledge. Evaluated against ground-truth labels from expert radiologists, our labeler extracts comparisons describing the progression of pulmonary edema with 0.875 precision and 0.891 recall. We also demonstrate the potential utility of comparison labels in providing additional fine-grained information over noisier labels produced by models that directly estimate severity level.

Investigation of the vertical structure of the lower atmosphere during heat wave conditions

<p>The research on heat waves is strongly motivated by their impacts on human<br />life and the economy. Consequently, less research has been done on the<br />state of the lower atmosphere as a whole during these extreme events,<br />although it may play a role in the formation and persistence of heat<br />waves. Miralles et al. (2014) show that different factors must come<br />together to produce extremes such as the pronounced heat waves<br />in the year 2003 in France and 2010 in Russia. One interesting phenomenon<br />in this context is the emergence of an unusually deep boundary layer. The aim<br />of this work is to analyse whether this feature is a common trait of European<br />heat waves in general. To this end, we systematically investigate the vertical<br />structure and evolution of the lower atmosphere during heat waves in the<br />time period from 2014 to 2018. COSMO-REA6 data is used to find heatwaves<br />and provides vertical profiles of the atmosphere which we also compare<br />to radio sonde measurements. The results of our work could possibly be<br />used to improve the discriminability of different severity levels of heat waves or to<br />formulate a heat wave measure that is not based solely on surface variables.</p>

On the transience or stability of subthreshold psychopathology

Symptoms of psychopathology lie on a continuum ranging from mental health to psychiatric disorders. Although much research has focused on progression along this continuum, for most individuals, subthreshold symptoms do not escalate into full-blown disorders. This study investigated how the stability of psychopathological symptoms (attractor strength) varies across severity levels (homebase). Data were retrieved from the TRAILS TRANS-ID study, where 122 at-risk young adults (mean age 23.6 years old, 57% males) monitored their mental states daily for a period of six months (± 183 observations per participant). We estimated each individual’s homebase and attractor strength using generalized additive mixed models. Regression analyses showed no association between homebases and attractor strengths (linear model: B = 0.02, p = 0.47, R2 < 0.01; polynomial model: B < 0.01, p = 0.61, R2 < 0.01). Sensitivity analyses where we (1) weighed estimates according to their uncertainty and (2) removed individuals with a DSM-5 diagnosis from the analyses did not change this finding. This suggests that stability is similar across severity levels, implying that subthreshold psychopathology may resemble a stable state rather than a transient intermediate between mental health and psychiatric disorder. Our study thus provides additional support for a dimensional view on psychopathology, which implies that symptoms differ in degree rather than kind.

A universal severity classification for natural disasters

The magnitude of a disaster’s severity cannot be easily assessed because there is no global method that provides real magnitudes of natural disaster severity levels. Therefore, a new universal severity classification scheme for natural disasters is developed and is supported by data. This universal system looks at the severity of disasters based on the most influential impact factor and gives a rating from zero to ten: Zero indicates no impact and ten is a worldwide devastation. This universal system is for all types of natural disasters, from lightning strikes to super-volcanic eruptions and everything in between, that occur anywhere in the world at any time. This novel universal severity classification system measures, describes, compares, rates, ranks, and categorizes impacts of disasters quantitatively and qualitatively. The severity index is useful to diverse stakeholder groups, including policy makers, governments, responders, and civilians, by providing clear definitions that help convey the severity levels or severity potential of a disaster. Therefore, this universal system is expected to avoid inconsistencies and to connect severity metrics to generate a clear perception of the degree of an emergency; the system is also expected to improve mutual communication among stakeholder groups. Consequently, the proposed universal system will generate a common communication platform and improve understanding of disaster risk, which aligns with the priority of the Sendai Framework for Disaster Risk Reduction 2015–2030. This research was completed prior to COVID-19, but the pandemic is briefly addressed in the discussion section.

Adaptive Behavior, Emotional/Behavioral Problems and Parental Stress in Children With Autism Spectrum Disorder

Background: The aim of our study was to compare adaptive skills, emotional/behavioral problems, and parental stress among children with different severity levels of Autism Spectrum Disorder (ASD) symptoms.Methods: This study included a sample of 88 subjects with ASD (mean age = 6.00 ± 2.70). All subjects underwent standardized neuropsychological tests for the assessment of symptoms of the autism spectrum (Autism Diagnostic Observation Schedule-Second Edition), adaptive level (The Vineland Adaptive Behavior Scales, Survey Interview, 2nd edition), behavioral and emotional problems (Child Behavior CheckList CBCL), and parental stress (Parental Stress Index Short Form-PSI-SF). Non-parametric statistical methods (Kruskal-Wallis test and Mann-Whitney U-test for post hoc analysis) and linear regression analysis were used in this study.Results:Children who had higher severity levels of ASD symptoms had less adaptive functioning; younger children showed more severe symptoms of ASD; older children had better communication skills. The presence of greater adaptive difficulties was related to a greater presence of internalizing problems. An increase in parental stress levels was related to an higher severity of ASD symptoms, fewer adaptive skills, and a greater presence of internalizing and externalizing problems.Conclusion: This study suggests that the adaptive behavior should be considered in order to planning a habilitation intervention in children with autism. It is also important to monitor emotional/behavioral problems and parental stress levels in order to provide parenting support and improve the family quality of life.

D Dimer – Prognostic indicator for disease severity in patients hospitalised with COVID 19

Patients of Covid 19 infections present with different severity. Levels of D Dimer in these patients can be correlated with disease severity for management and prognosis. To evaluate the usefulness of D-Dimer levels in blood to correlate with disease severity in COVID 19 patients. Retrospective study was done in Department of Pathology of Secondary Care hospital that became designated covid hospital from May 2021 to June 2021 on 60 COVID 19 positive admitted patients. D dimer levels were analysed and correlated with clinical severity of disease. Out of total 60 patients, 33 were in mild, 23 in moderate and 4 were in severe category. In mild cases D Dimer varies from 43 ng/ml to 183 ng/ml. In moderate cases D Dimer varies from 270 ng/ml to 991 ng/ml. In severe cases D Dimer varies from 1043 ng/ml to 2463 ng/ml. The study suggests cut off levels for D Dimer as up to 200 ng/ml for mild, 200-1000 ng/ml for moderate and more than 1000 ng/ml for severe category in COVID 19 patients. D dimer helps in identifying severe disease and can be used as an essential biomarker in developing the management protocol for COVID 19 patients.

A Generic Deep Learning Based Cough Analysis System from Clinically Validated Samples for Point-of-Need Covid-19 Test and Severity Levels

We seek to evaluate the detection performance of a rapid primary screening tool of Covid-19 solely based on the cough sound from 8,380 clinically validated samples with laboratory molecular-test (2,339 Covid-19 positive and 6,041 Covid-19 negative). Samples were clinically labelled according to the results and severity based on quantitative RT-PCR (qRT-PCR) analysis, cycle threshold and lymphocytes count from the patients. Our proposed generic method is a algorithm based on Empirical Mode Decomposition (EMD) with subsequent classification based on a tensor of audio features and deep artificial neural network classifier with convolutional layers called DeepCough'. Two different versions of DeepCough based on the number of tensor dimensions, i.e. DeepCough2D and DeepCough3D, have been investigated. These methods have been deployed in a multi-platform proof-of-concept Web App CoughDetect to administer this test anonymously. Covid-19 recognition results rates achieved a promising AUC (Area Under Curve) of 98.800.83%, sensitivity of 96.431.85%, and specificity of 96.201.74%, and 81.08%5.05% AUC for the recognition of three severity levels. Our proposed web tool and underpinning algorithm for the robust, fast, point-of-need identification of Covid-19 facilitates the rapid detection of the infection. We believe that it has the potential to significantly hamper the Covid-19 pandemic across the world.

Crash Injury Severity Prediction Using an Ordinal Classification Machine Learning Approach

In many related works, nominal classification algorithms ignore the order between injury severity levels and make sub-optimal predictions. Existing ordinal classification methods suffer rank inconsistency and rank non-monotonicity. The aim of this paper is to propose an ordinal classification approach to predict traffic crash injury severity and to test its performance over existing machine learning classification methods. First, we compare the performance of the neural network, XGBoost, and SVM classifiers in injury severity prediction. Second, we utilize a severity category-combination method with oversampling to relieve the class-imbalance problem prevalent in crash data. Third, we take advantage of probability calibration and the optimal probability threshold moving to improve the prediction ability of ordinal classification. The proposed approach can satisfy the rank consistency and rank monotonicity requirement and is proved to be superior to other ordinal classification methods and nominal classification machine learning by statistical significance test. Important factors relating to injury severity are selected based on their permutation feature importance scores. We find that converting severity levels into three classes, minor injury, moderate injury, and serious injury, can substantially improve the prediction precision.

Soil Mesofauna Responses to Fire Severity in a Sclerophyllous Forest in Central Chile

Forest fires may have severe impacts on the aboveground biodiversity and soil chemical and biological properties. Edaphic organisms are highly sensitive to disturbances and are typically used to measure the magnitude of these events. Overall, little is known about the responses of these organisms to fires differing in their severity levels. This study aimed to assess the effect of fire severity on the soil mesofauna community diversity and structure in a site located in a Mediterranean zone of central Chile. In postfire conditions, we use spectral indexes from satellite images to map fire severity at four levels (non-damage (ND), low damage (L), medium damage (M), high damage (H)). Soil samples were collected at each severity level, and the mesofauna abundance was quantified. Although the metrics describing species diversity and dominance were similar among fire severity levels, the abundance and composition of the mesofauna were specifically altered at the high severity level. The edaphic mesofauna can be considered suitable bioindicators to evaluate the postfire ecosystem recovery, especially in the areas highly damaged by fire.