Opportunistic detection of type 2 diabetes using deep learning from frontal chest radiographs.

Deep learning (DL) models can harness electronic health records (EHRs) to predict diseases and extract radiologic findings for diagnosis. With ambulatory chest radiographs (CXRs) frequently ordered, we investigated detecting type 2 diabetes (T2D) by combining radiographic and EHR data using a DL model. Our model, developed from 271,065 CXRs and 160,244 patients, was tested on a prospective dataset of 9,943 CXRs. Here we show the model effectively detected T2D with a ROC AUC of 0.84 and a 16% prevalence. The algorithm flagged 1,381 cases (14%) as suspicious for T2D. External validation at a distinct institution yielded a ROC AUC of 0.77, with 5% of patients subsequently diagnosed with T2D. Explainable AI techniques revealed correlations between specific adiposity measures and high predictivity, suggesting CXRs' potential for enhanced T2D screening.

Deep Learning-Based Digitally Reconstructed Tomography of the Chest in the Evaluation of Solitary Pulmonary Nodules: A Feasibility Study.

RATIONALE AND OBJECTIVES: Computed tomography (CT) is preferred for evaluating solitary pulmonary nodules (SPNs) but access or availability may be lacking, in addition, overlapping anatomy can hinder detection of SPNs on chest radiographs. We developed and evaluated the clinical feasibility of a deep learning algorithm to generate digitally reconstructed tomography (DRT) images of the chest from digitally reconstructed frontal and lateral radiographs (DRRs) and use them to detect SPNs. METHODS: This single-institution retrospective study included 637 patients with noncontrast helical CT of the chest (mean age 68 years, median age 69 years, standard deviation 11.7 years; 355 women) between 11/2012 and 12/2020, with SPNs measuring 10-30 mm. A deep learning model was trained on 562 patients, validated on 60 patients, and tested on the remaining 15 patients. Diagnostic performance (SPN detection) from planar radiography (DRRs and CT scanograms, PR) alone or with DRT was evaluated by two radiologists in an independent blinded fashion. The quality of the DRT SPN image in terms of nodule size and location, morphology, and opacity was also evaluated, and compared to the ground-truth CT images RESULTS: Diagnostic performance was higher from DRT plus PR than from PR alone (area under the receiver operating characteristic curve 0.95-0.98 versus 0.80-0.85; p < 0.05). DRT plus PR enabled diagnosis of SPNs in 11 more patients than PR alone. Interobserver agreement was 0.82 for DRT plus PR and 0.89 for PR alone; and interobserver agreement for size and location, morphology, and opacity of the DRT SPN was 0.94, 0.68, and 0.38, respectively. CONCLUSION: For SPN detection, DRT plus PR showed better diagnostic performance than PR alone. Deep learning can be used to generate DRT images and improve detection of SPNs.

Detecting Racial/Ethnic Health Disparities Using Deep Learning From Frontal Chest Radiography.

PURPOSE: The aim of this study was to assess racial/ethnic and socioeconomic disparities in the difference between atherosclerotic vascular disease prevalence measured by a multitask convolutional neural network (CNN) deep learning model using frontal chest radiographs (CXRs) and the prevalence reflected by administrative hierarchical condition category codes in two cohorts of patients with coronavirus disease 2019 (COVID-19). METHODS: A CNN model, previously published, was trained to predict atherosclerotic disease from ambulatory frontal CXRs. The model was then validated on two cohorts of patients with COVID-19: 814 ambulatory patients from a suburban location (presenting from March 14, 2020, to October 24, 2020, the internal ambulatory cohort) and 485 hospitalized patients from an inner-city location (hospitalized from March 14, 2020, to August 12, 2020, the external hospitalized cohort). The CNN model predictions were validated against electronic health record administrative codes in both cohorts and assessed using the area under the receiver operating characteristic curve (AUC). The CXRs from the ambulatory cohort were also reviewed by two board-certified radiologists and compared with the CNN-predicted values for the same cohort to produce a receiver operating characteristic curve and the AUC. The atherosclerosis diagnosis discrepancy, Δvasc, referring to the difference between the predicted value and presence or absence of the vascular disease HCC categorical code, was calculated. Linear regression was performed to determine the association of Δvasc with the covariates of age, sex, race/ethnicity, language preference, and social deprivation index. Logistic regression was used to look for an association between the presence of any hierarchical condition category codes with Δvasc and other covariates. RESULTS: The CNN prediction for vascular disease from frontal CXRs in the ambulatory cohort had an AUC of 0.85 (95% confidence interval, 0.82-0.89) and in the hospitalized cohort had an AUC of 0.69 (95% confidence interval, 0.64-0.75) against the electronic health record data. In the ambulatory cohort, the consensus radiologists' reading had an AUC of 0.89 (95% confidence interval, 0.86-0.92) relative to the CNN. Multivariate linear regression of Δvasc in the ambulatory cohort demonstrated significant negative associations with non-English-language preference (ß = -0.083, P < .05) and Black or Hispanic race/ethnicity (ß = -0.048, P < .05) and positive associations with age (ß = 0.005, P < .001) and sex (ß = 0.044, P < .05). For the hospitalized cohort, age was also significant (ß = 0.003, P < .01), as was social deprivation index (ß = 0.002, P < .05). The Δvasc variable (odds ratio [OR], 0.34), Black or Hispanic race/ethnicity (OR, 1.58), non-English-language preference (OR, 1.74), and site (OR, 0.22) were independent predictors of having one or more hierarchical condition category codes (P < .01 for all) in the combined patient cohort. CONCLUSIONS: A CNN model was predictive of aortic atherosclerosis in two cohorts (one ambulatory and one hospitalized) with COVID-19. The discrepancy between the CNN model and the administrative code, Δvasc, was associated with language preference in the ambulatory cohort; in the hospitalized cohort, this discrepancy was associated with social deprivation index. The absence of administrative code(s) was associated with Δvasc in the combined cohorts, suggesting that Δvasc is an independent predictor of health disparities. This may suggest that biomarkers extracted from routine imaging studies and compared with electronic health record data could play a role in enhancing value-based health care for traditionally underserved or disadvantaged patients for whom barriers to care exist.

Predicting clinical outcomes among hospitalized COVID-19 patients using both local and published models.

BACKGROUND: Many models are published which predict outcomes in hospitalized COVID-19 patients. The generalizability of many is unknown. We evaluated the performance of selected models from the literature and our own models to predict outcomes in patients at our institution. METHODS: We searched the literature for models predicting outcomes in inpatients with COVID-19. We produced models of mortality or criticality (mortality or ICU admission) in a development cohort. We tested external models which provided sufficient information and our models using a test cohort of our most recent patients. The performance of models was compared using the area under the receiver operator curve (AUC). RESULTS: Our literature review yielded 41 papers. Of those, 8 were found to have sufficient documentation and concordance with features available in our cohort to implement in our test cohort. All models were from Chinese patients. One model predicted criticality and seven mortality. Tested against the test cohort, internal models had an AUC of 0.84 (0.74-0.94) for mortality and 0.83 (0.76-0.90) for criticality. The best external model had an AUC of 0.89 (0.82-0.96) using three variables, another an AUC of 0.84 (0.78-0.91) using ten variables. AUC's ranged from 0.68 to 0.89. On average, models tested were unable to produce predictions in 27% of patients due to missing lab data. CONCLUSION: Despite differences in pandemic timeline, race, and socio-cultural healthcare context some models derived in China performed well. For healthcare organizations considering implementation of an external model, concordance between the features used in the model and features available in their own patients may be important. Analysis of both local and external models should be done to help decide on what prediction method is used to provide clinical decision support to clinicians treating COVID-19 patients as well as what lab tests should be included in order sets.

Predicting Prolonged Hospitalization and Supplemental Oxygenation in Patients with COVID-19 Infection from Ambulatory Chest Radiographs using Deep Learning.

RATIONALE AND OBJECTIVES: The clinical prognosis of outpatients with coronavirus disease 2019 (COVID-19) remains difficult to predict, with outcomes including asymptomatic, hospitalization, intubation, and death. Here we determined the prognostic value of an outpatient chest radiograph, together with an ensemble of deep learning algorithms predicting comorbidities and airspace disease to identify patients at a higher risk of hospitalization from COVID-19 infection. MATERIALS AND METHODS: This retrospective study included outpatients with COVID-19 confirmed by reverse transcription-polymerase chain reaction testing who received an ambulatory chest radiography between March 17, 2020 and October 24, 2020. In this study, full admission was defined as hospitalization within 14 days of the COVID-19 test for > 2 days with supplemental oxygen. Univariate analysis and machine learning algorithms were used to evaluate the relationship between the deep learning model predictions and hospitalization for > 2 days. RESULTS: The study included 413 patients, 222 men (54%), with a median age of 51 years (interquartile range, 39-62 years). Fifty-one patients (12.3%) required full admission. A boosted decision tree model produced the best prediction. Variables included patient age, frontal chest radiograph predictions of morbid obesity, congestive heart failure and cardiac arrhythmias, and radiographic opacity, with an internally validated area under the curve (AUC) of 0.837 (95% CI: 0.791-0.883) on a test cohort. CONCLUSION: Deep learning analysis of single frontal chest radiographs was used to generate combined comorbidity and pneumonia scores that predict the need for supplemental oxygen and hospitalization for > 2 days in patients with COVID-19 infection with an AUC of 0.837 (95% confidence interval: 0.791-0.883). Comorbidity scoring may prove useful in other clinical scenarios.

Cognitive Decline, Body Mass Index, and Waist Circumference in Community-Dwelling Elderly Participants.

OBJECTIVE: The aim of this study was to explore the association of body mass index (BMI), waist circumference (WC), and BMI and WC changes over time with cognitive decline in a nationally representative sample. METHODS: A total of 5239 participants (≥65 years) were followed for 3 years as part of the National Health and Aging Trends Study. Cox proportional hazard regression was applied to model the risk of cognitive decline. RESULTS: BMI, after adjusting for WC and main confounders, was associated with reduced risk of cognitive decline (hazard ratio [HR] 0.97 for each unit BMI increase, 0.95-0.99). After stratifying by gender and age, this effect remained significant among females and young elders ≤80 years. A BMI decrease and WC increase >10% over the study period were associated with increased risk of cognitive decline (HR 1.98, 1.16-3.38; HR 1.30, 1.04-1.62, respectively). CONCLUSION: In the elderly individuals, lean mass, as measured by BMI adjusted for WC, was associated with reduced risk of cognitive decline. Loss of lean mass and gain of fat mass, as measured by WC adjusted for BMI, were associated with elevated risk of cognitive decline.

Generalizability of Pharmacologic and Psychotherapy Clinical Trial Results for Posttraumatic Stress Disorder to Community Samples.

OBJECTIVE: The present study sought to quantify the generalizability of pharmacologic and psychotherapy clinical trial results in individuals with a DSM-IV diagnosis of posttraumatic stress disorder (PTSD) to a large representative community sample. METHODS: Data were derived from the 2004-2005 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), a large nationally representative sample of the adult US population. We applied a standard set of exclusion criteria representative of pharmacologic and psychotherapy clinical trials to all adults with a DSM-IV diagnosis of PTSD in the previous 12 months (n = 1,715) and then to a subsample of participants seeking treatment (n = 366). Our aim was to assess how many participants with PTSD would fulfill typical eligibility criteria. RESULTS: We found that more than 6 of 10 respondents from the overall PTSD sample and more than 7 of 10 respondents seeking treatment for PTSD would have been excluded by 1 exclusion criterion or more in a typical pharmacologic trial. In contrast, about 2 of 10 participants in the full sample and about 3 of 10 participants seeking treatment for PTSD would have been excluded in a typical psychotherapy efficacy trial. CONCLUSIONS: We found that psychotherapy trial results may be applied to most patients with PTSD in routine clinical practice. The designers of pharmacologic clinical trials should carefully consider the trade-offs between the application of each exclusion criterion and its impact on representativeness. Specification a priori of the goals of the study, better justification for each exclusion criterion, and estimation of the proportion of individuals ineligible for the trial would assist study design. Developing integrated forms of pharmacotherapy and psychotherapy that simultaneously target commonly overlapping psychiatric disorders may yield more informative results for mental health care providers and research funding agencies.

Generalizability of Neuroimaging Studies in 5 Common Psychiatric Disorders Based on the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC).

OBJECTIVE: Although neuroimaging studies have an important role in psychiatric nosology and treatment development, little is known about the representativeness of participants in neuroimaging research. We estimated the effects of commonly used study eligibility criteria on the representativeness of neuroimaging research participants in relation to the general population with the psychiatric disorders of interest. METHODS: Common eligibility criteria were applied from 112 published neuroimaging studies of DSM-IV nicotine dependence (13 studies), alcohol dependence (12 studies), drug use disorders (13 studies), major depressive disorder (MDD) (37 studies), and posttraumatic stress disorder (PTSD) (36 studies) to representative US samples with these conditions from the 2001-2002 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) (n = 43,093). The analyses were repeated with NESARC respondents with the disorders and substantial psychosocial impairment. RESULTS: Most NESARC respondents with nicotine dependence (64.1%), alcohol dependence (57.7%), drug use disorders (86.6%), and PTSD (66.9%), though not with MDD (18.2%), would have been excluded by eligibility criteria used in at least half of the relevant neuroimaging studies. Across the diagnostic groups, comorbid psychiatric and general medical conditions resulted in the largest percentages of exclusions. Corresponding analyses limited to respondents with substantial impairment excluded larger percentages with nicotine dependence (77.6%), alcohol dependence (75.8%), drug use disorders (93.5%), and PTSD (76.8%), though not MDD (18.3%). CONCLUSIONS: Neuroimaging studies tend to recruit highly selected samples with the psychiatric disorders of interest that markedly underrepresent individuals with common comorbid conditions. Larger studies with less restrictive eligibility criteria may promote translation of advances in neuroimaging research to populations commonly encountered in clinical practice.

A dimensional liability model of age differences in mental disorder prevalence: evidence from a national sample.

Recent theories have proposed a metastructure that organizes related mental disorders into broad dimensions of psychopathology (i.e., internalizing and externalizing dimensions). Prevalence rates of most mental disorders, when examined independently, are substantially lower in older than in younger adults, which may affect this metastructure. Within a nationally representative sample, the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC; N = 43,093), we developed a dimensional liability model of common psychiatric disorders to clarify whether aging affects specific disorders or general dimensions of psychopathology. Significant age differences existed across age groups (18-24, 25-34, 35-44, 45-54, 55-64, 65-75 and 75+), such that older adults showed lower prevalence rates of most disorders compared to younger adults. We next investigated patterns of disorder comorbidity for past-year psychiatric disorders and found that a distress-fear-externalizing liability model fit the data well. This model was age-group invariant and indicated that the observed lower prevalence of mental disorders with advancing age originates from lower average means on externalizing and internalizing liability dimensions. This unifying dimensional liability model of age and mental disorder comorbidity can help inform the role of aging on mental disorder prevalence for research and intervention efforts, and service planning for the impending crisis in geriatric mental health.

Probability and predictors of treatment-seeking for substance use disorders in the U.S.

BACKGROUND: Little is known about to what extent treatment-seeking behavior varies across individuals with alcohol abuse, alcohol dependence, drug abuse, and drug dependence. METHODS: The sample included respondents from the Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) who reported a lifetime diagnosis alcohol abuse, alcohol dependence, drug abuse, or drug dependence. Unadjusted and adjusted hazard ratios are presented for time to first treatment contact by sociodemographic characteristics and comorbid psychiatric disorders. Individuals were censored from the analyses if their condition remitted prior to seeking treatment. RESULTS: In the first year after disorder onset, rates of treatment-seeking were 13% for drug dependence, 5% for alcohol dependence, 2% for drug abuse, and 1% for alcohol abuse. The lifetime probability of seeking treatment among individuals who did not remit was also highest for drug dependence (90%), followed by drug abuse (60%), alcohol dependence (54%), and alcohol abuse (16%). Having had previous treatment contact for a substance use disorder (SUD) increased the probability of seeking treatment for another SUD. By contrast, an early age of SUD onset, belonging to an older cohort, and a higher level of education decreased the lifetime probability of treatment contact for SUD. The role of comorbid mental disorders was more complex, with some disorders increasing and other decreasing the probability of seeking treatment. CONCLUSIONS: Given high rates of SUD and their substantial health and economic burden, these patterns suggest the need for innovative approaches to increase treatment access for individuals with SUD.

Age at onset of major depression and adulthood cardiovascular risk.

Childhood-onset compared to adulthood-onset of major depression is associated with increased rates of serious cardiovascular events, independently of cardiovascular risk factors. This could be explained by a longer duration of exposure to depression. Cardiovascular disease risk should be systematically assessed in individuals with long duration of major depression.

Scientific financial funding in Colombia from 2000 to 2012 / Presupuesto de financiación científica en Colombia del 2000 al 2012 / Captação de recursos científicos na Colômbia 2000-2012

This study evaluates trends in funding for Science, Technology and Innovation, Research and Development and COLCIENCIAS (Administrative Department for Science, Technology and Innovation) between 2000-2006 and 2007-2012. Available data from the World Bank, OCYT (Colombian observatory of science and technology), DANE (National statistics department), Banco de la República and COLCIENCIAS to evaluate funding source by sector (private, public and international), financial growth rate, financial expenditure, and activity related expenses from 2000 to 2012, and regression models to estimate financial trends. COLCIENCIAS funding increased in the past years; Science, Technology and Innovation, and Research and Development funding increased from $1,296.7 million US dollars in 2000-2006 to $2,766.4 million US dollars in 2007-2012. The financial analysis showed a significant increase in public funding mainly by government (p<0.05); however, government and corporation expenditure did not vary from 2000 to 2012.

Este estudio evalúa las tendencias financieras de Ciencia, Tecnología e innovación (STI), Investigación y Desarrollo (RD) y COLCIENCIAS (Departamento Administrativo de Ciencia, Tecnología e Innovación) entre el 2000-2006 y 2007-2012. Se usó información disponible del World Bank, OCyT (Observatorio Colombiano de Ciencia y Tecnología), DANE (Departamento Administrativo Nacional de Estadística), Banco de la República y COLCIENCIAS, se analizó: la fuente (privada, pública, internacional), tasa de crecimiento y ejecución financiera, así como ejecución por actividad del 2000 al 2012. Se usaron modelos de regresión para estimar tendencias financieras. La financiación Colombiana en STI, RD y COLCIENCIAS aumentó en los últimos años. La inversión en STI y RD aumentó entre 2000-2006 y 2007-2012 de $1,296.7 a $2.766.4 millones de dólares, respectivamente. Análisis evidenció un incremento significativo (p<0.05) en la inversión pública, siendo el gobierno el principal partícipe. Sin embargo, la ejecución financiera del gobierno y empresas no mostró cambios entre 2000-2012.

Este estudo avaliou as tendências financeiras da Ciência, Tecnologia e Inovação, Investigação e Desenvolvimento e COLCIENCIAS (Departamento Administrativo de Ciência, Tecnologia e Inovação), entre 2000-2006 e 2007-2012. Dados do Banco Mundial, OCyT (Observatório Colombiano de Ciência e Tecnologia), DANE (Departamento Nacional de Estatística), Banco de la República e COLCIENCIAS foram utilizados para analisar: origem do financiamento por setor (privado, público e internacional), taxa de crescimento, despesas, e atividades relacionadas às despesas, entre 2000-2012. Modelos de regressão foram utilizados para se chegar às tendências financeiras. O investimento em STI, RD e COLCIENCIAS tem crescido nos últimos anos. O investimento em Ciência, Tecnologia e Inovação,e Investigação e Desenvolvimento aumentou de US $1,296.7 milhões de dólares em 2000-2006 para US $2,766.4 milhões de dólares em 2007-2012. A análise mostrou um aumento significativo no financiamento público (p<0,05), mesmo que as despesas do governo e das empresas não se tenham alterado entre 2000-2012.

El papel del receptor de glucocorticoide en el estrés temprano: [revisión] / The relevance of glucocorticoid receptor in early stress: [review]

En estudios previos se han relacionado las alteraciones funcionales del eje hipotálamohipofisario-adrenal y el estrés temprano; por ejemplo, el aumento en la producción de corticotropina (ACTH) y glucocorticoide como factor clave en la fisiopatología de trastornos del estrés como la depresión. En este artículo se presentan los resultados de estudios en epigenética en busca del posible nexo entre el estrés temprano, la disminución en la expresión del receptor de glucocorticoide y la hiperactividad del eje hipotálamo-hipofisario-adrenal. De esta manera, se identifica al estrés temprano como modulador del neurodesarrollo de las estructuras cerebrales implicadas en la respuesta frente al estrés, así como el papel del receptor de glucocorticoide en dicho proceso.

Previous studies have shown how Hypothalamic-Pituitary-adrenal Axis dysfunction is related to early life stress; several works show that Hypothalamic-Pituitary-adrenal Axishyperactivity increases production of ACTH and glucocorticoids, indicating a pathophysiological key factor in stress related diseases like depression. This review will discuss results of some epigenetical studies linking early life stress, decreased production of the glucocorticoid receptor and Hypothalamic-Pituitary-adrenal Axis hyperactivity. We conclude how early life stress modulates the expression of the glucocorticoid receptor affecting the development of several brain structures involved in the stress response.