Congenital heart disease (CHD) is the most common birth defect. Fetal survey ultrasound is recommended worldwide, including five views of the heart that together could detect 90% of complex CHD. In practice, however, sensitivity is as low as 30%. We hypothesized poor detection results from challenges in acquiring and interpreting diagnostic-quality cardiac views, and that deep learning could improve complex CHD detection. Using 107,823 images from 1,326 retrospective echocardiograms and surveys from 18-24 week fetuses, we trained an ensemble of neural networks to (i) identify recommended cardiac views and (ii) distinguish between normal hearts and complex CHD. Finally, (iii) we used segmentation models to calculate standard fetal cardiothoracic measurements. In a test set of 4,108 fetal surveys (0.9% CHD, >4.4 million images, about 400 times the size of the training dataset) the model achieved an AUC of 0.99, 95% sensitivity (95%CI, 84-99), 96% specificity (95%CI, 95-97), and 100% NPV in distinguishing normal from abnormal hearts. Sensitivity was comparable to clinicians' task-for-task and remained robust on external and lower-quality images. The model's decisions were based on clinically relevant features. Cardiac measurements correlated with reported measures for normal and abnormal hearts. Applied to guidelines-recommended imaging, ensemble learning models could significantly improve detection of fetal CHD and expand telehealth options for prenatal care at a time when the COVID-19 pandemic has further limited patient access to trained providers. This is the first use of deep learning to approximately double standard clinical performance on a critical and global diagnostic challenge.
Critical Care 2017, 21(Suppl 1):P349 Introduction Imbalance in cellular energetics has been suggested to be an important mechanism for organ failure in sepsis and septic shock. We hypothesized that such energy imbalance would either be caused by metabolic changes leading to decreased energy production or by increased energy consumption. Thus, we set out to investigate if mitochondrial dysfunction or decreased energy consumption alters cellular metabolism in muscle tissue in experimental sepsis. Methods We submitted anesthetized piglets to sepsis (n = 12) or placebo (n = 4) and monitored them for 3 hours. Plasma lactate and markers of organ failure were measured hourly, as was muscle metabolism by microdialysis. Energy consumption was intervened locally by infusing ouabain through one microdialysis catheter to block major energy expenditure of the cells, by inhibiting the major energy consuming enzyme, N+/K + -ATPase. Similarly, energy production was blocked infusing sodium cyanide (NaCN), in a different region, to block the cytochrome oxidase in muscle tissue mitochondria. Results All animals submitted to sepsis fulfilled sepsis criteria as defined in Sepsis-3, whereas no animals in the placebo group did. Muscle glucose decreased during sepsis independently of N+/K + -ATPase or cytochrome oxidase blockade. Muscle lactate did not increase during sepsis in naïve metabolism. However, during cytochrome oxidase blockade, there was an increase in muscle lactate that was further accentuated during sepsis. Muscle pyruvate did not decrease during sepsis in naïve metabolism. During cytochrome oxidase blockade, there was a decrease in muscle pyruvate, independently of sepsis. Lactate to pyruvate ratio increased during sepsis and was further accentuated during cytochrome oxidase blockade. Muscle glycerol increased during sepsis and decreased slightly without sepsis regardless of N+/K + -ATPase or cytochrome oxidase blocking. There were no significant changes in muscle glutamate or urea during sepsis in absence/presence of N+/K + -ATPase or cytochrome oxidase blockade. ConclusionsThese results indicate increased metabolism of energy substrates in muscle tissue in experimental sepsis. Our results do not indicate presence of energy depletion or mitochondrial dysfunction in muscle and should similar physiologic situation be present in other tissues, other mechanisms of organ failure must be considered. , and long-term follow up has shown increased fracture risk [2]. It is unclear if these changes are a consequence of acute critical illness, or reduced activity afterwards. Bone health assessment during critical illness is challenging, and direct bone strength measurement is not possible. We used a rodent sepsis model to test the hypothesis that critical illness causes early reduction in bone strength and changes in bone architecture. Methods 20 Sprague-Dawley rats (350 ± 15.8g) were anesthetised and randomised to receive cecal ligation and puncture (CLP) (50% cecum length, 18G needle single pass through anterior and posterior wa...
Objective To evaluate the predictive value of Computed Tomography Angiography (CTA) measurements of the RVOT for transcatheter valve sizing. Background Transcatheter pulmonary valve replacement (TPVR) provides an alternative to surgery in patients with right ventricular outflow tract (RVOT) dysfunction. We studied 18 patients who underwent catheterization for potential TPVR to determine whether CT imaging can be used to accurately predict implant size. Methods Cases were grouped by RVOT characteristics: native or transannular patch (n = 8), conduit (n = 5) or bioprosthetic valve (n = 5). TPVR was undertaken in 14/18 cases, after balloon-sizing was used to confirm suitability and select implant size. Retrospective CT measurements of the RVOT (circumference-derived (D circ ) and area-derived (D area ) diameters) were obtained at the level of the annulus, bioprosthesis or conduit. Using manufacturer sizing guidance, a valve size was generated and a predicted valve category assigned: (1) <18 mm, (2) 18–20 mm, (3) 22–23 mm, (4) 26–29 mm and (5) >29 mm. Predicted and implanted valves were compared for inter-rater agreement using Cohen’s kappa coefficient. Results The median age of patients was 37 years old (IQR: 30–49); 55% were male. Diagnoses included: Tetralogy of Fallot (12/18), d-Transposition repair (3/18), congenital pulmonary stenosis (2/18) and carcinoid heart disease (1/18). Measurements of D area (κ = 0.697, p < 0.01) and D circ (κ = 0.540, p < 0.01) were good predictors of implanted valve size. When patients with RVOT conduits were excluded, the predictive accuracy improved for D area (κ = 0.882, p < 0.01) and D circ (κ = 0.882, p < 0.01). Conclusions CT measurement of the RVOT, using D area or D circ , can predict prosthetic valve sizing in TPVR. These measurements are less predictive in patients with conduits, compared to those with a native RVOT or pulmonic bioprosthesis. Condensed abstract We studied 18 patients who underwent catheterization for TPVR to determine whether CT imaging could be used to accurately predict implant size. Retrospective RVOT measurements were used to generate a predicted valve size, which was compared with implanted valve size for inter-rater agreement. Measurements of D area (κ = 0.697, p < 0.01) and D circ (κ = 0.540, p < 0.01) were good predictors of implanted valve size. When cases with RVOT conduits were excluded, the predictive accuracy improved for D area (κ = 0.882, p < 0.01) and D circ (κ = 0.882, p < 0.01). CT measurement of the RVOT can accurately predict prosthetic valve sizing in TPVR. Thes...
Hypertrophic cardiomyopathy (HCM) is an important cause of morbidity and mortality, with rare pathogenic variants found in about a third of cases (sarcomere-positive). Large-scale genome-wide association studies (GWAS) demonstrate that common genetic variation contributes substantially to HCM risk. Here, we derive polygenic scores (PGS) from HCM GWAS, and multi-trait analysis of GWAS incorporating genetically-correlated traits, and test their performance in the UK Biobank, 100,000 Genomes Project, and across clinical cohorts. Higher PGS substantially increases population risk of HCM, particularly amongst sarcomere-positive carriers where HCM penetrance differs 10-fold between those in the highest and lowest PGS quintiles. In relatives of HCM patients, PGS stratifies risks of developing HCM and adverse outcomes. Finally, PGS strongly predicts risk of adverse outcomes in HCM, with a 4 to 6-fold increase in death between cases in the highest and lowest PGS quintiles. These findings promise broad clinical utility of PGS in the general population, in cases, and in families with HCM, enabling tailored screening and surveillance, and stratification of risk of adverse outcomes.
Background Methamphetamine misuse affects 27 million people worldwide and is associated with cardiovascular disease (CVD); however, risk factors for CVD among users have not been well studied. Methods and Results We studied hospitalized patients in California, captured by the Healthcare Cost and Utilization Project database, between 2005 and 2011. We studied the association between methamphetamine use and CVD (pulmonary hypertension, heart failure, stroke, and myocardial infarction). Among 20 249 026 persons in the Healthcare Cost and Utilization Project, 66 199 used methamphetamines (median follow‐up 4.58 years). Those who used were more likely younger (33 years versus 45 years), male (63.3% versus 44.4%), smoked, misused alcohol, and had depression and anxiety compared with nonusers. Methamphetamine use was associated with the development of heart failure (hazard ratio [HR], 1.53 [95% CI, 1.45–1.62]) and pulmonary hypertension (HR, 1.42 [95% CI, 1.26–1.60]). Among users, male sex (HR, 1.73 [95% CI, 1.37–2.18]) was associated with myocardial infarction. Chronic kidney disease (HR, 2.38 [95% CI, 1.74–3.25]) and hypertension (HR, 2.26 [95% CI, 2.03–2.51]) were strong risk factors for CVD among users. When compared with nonuse, methamphetamine use was associated with a 32% significant increase in CVD, alcohol abuse with a 28% increase, and cocaine use with a 47% increase in CVD. Conclusions Methamphetamine use has a similar magnitude of risk of CVD compared with alcohol and cocaine. Prevention and treatment could be focused on those with chronic kidney disease, hypertension, and mental health disorders.
Background Hypertrophic cardiomyopathy (HCM) is an important cause of sudden cardiac death associated with heterogeneous structural phenotypes but there is no systematic framework for classifying morphology or assessing associated risks. In this study we quantitatively survey genotype-phenotype associations in HCM to derive a data-driven taxonomy of disease expression for automated patient stratification. Methods An observational, single-centre study enrolled 436 HCM patients (median age 60 years; 28.8% women) with clinical, genetic and imaging data. An independent cohort of 60 HCM patients from Singapore (median age 59 years; 11% women) and a normative reference population from UK Biobank (n = 16,691, mean age 55 years; 52.5% women) with equivalent data were also recruited. We used machine learning to analyse the three dimensional structure of the left ventricle from cardiac magnetic resonance imaging and build a tree-based classification of HCM phenotypes. Genotype and mortality risk distributions were projected on the tree. Results The prevalence of pathogenic or likely pathogenic variants for HCM (P/LP) was 24.6%, while 66% were genotype negative. Carriers of P/LP variants had lower left ventricular mass, but greater basal septal hypertrophy, with reduced lifespan (mean follow-up 9.9 years) compared to genotype negative individuals (hazard ratio: 2.66; 95% confidence interval [CI]: 1.42-4.96; P < 0.002). Four main phenotypic branches were identified using unsupervised learning of three dimensional shape: 1) non-sarcomeric hypertrophy with co-existing hypertension; 2) diffuse and basal asymmetric hypertrophy associated with outflow tract obstruction; 3) isolated basal hypertrophy; 4) milder non-obstructive hypertrophy enriched for familial sarcomeric HCM (odds ratio for P/LP variants: 2.18 [95% CI: 1.93-2.28, P = 0.0001]). Phenotypic variation and associated risks could be visualised as a continuous distribution across the taxonomic tree. The model was generalisable to an independent cohort (trustworthiness M1: 0.86-0.88). Conclusions We report a data-driven taxonomy of HCM for identifying groups of patients with similar morphology while preserving a continuum of disease severity, genetic risk and outcomes. This approach will be of value for developing personalized clinical profiles to guide diagnosis, surveillance and intervention in patients with HCM, and improve understanding of the drivers of heterogeneity.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.