Impact of multi-output and stacking methods on feed efficiency prediction from genotype using machine learning algorithms.

Feeding represents the largest economic cost in meat production; therefore, selection to improve traits related to feed efficiency is a goal in most livestock breeding programs. Residual feed intake (RFI), that is, the difference between the actual and the expected feed intake based on animal's requirements, has been used as the selection criteria to improve feed efficiency since it was proposed by Kotch in 1963. In growing pigs, it is computed as the residual of the multiple regression model of daily feed intake (DFI), on average daily gain (ADG), backfat thickness (BFT), and metabolic body weight (MW). Recently, prediction using single-output machine learning algorithms and information from SNPs as predictor variables have been proposed for genomic selection in growing pigs, but like in other species, the prediction quality achieved for RFI has been generally poor. However, it has been suggested that it could be improved through multi-output or stacking methods. For this purpose, four strategies were implemented to predict RFI. Two of them correspond to the computation of RFI in an indirect way using the predicted values of its components obtained from (i) individual (multiple single-output strategy) or (ii) simultaneous predictions (multi-output strategy). The other two correspond to the direct prediction of RFI using (iii) the individual predictions of its components as predictor variables jointly with the genotype (stacking strategy), or (iv) using only the genotypes as predictors of RFI (single-output strategy). The single-output strategy was considered the benchmark. This research aimed to test the former three hypotheses using data recorded from 5828 growing pigs and 45,610 SNPs. For all the strategies two different learning methods were fitted: random forest (RF) and support vector regression (SVR). A nested cross-validation (CV) with an outer 10-folds CV and an inner threefold CV for hyperparameter tuning was implemented to test all strategies. This scheme was repeated using as predictor variables different subsets with an increasing number (from 200 to 3000) of the most informative SNPs identified with RF. Results showed that the highest prediction performance was achieved with 1000 SNPs, although the stability of feature selection was poor (0.13 points out of 1). For all SNP subsets, the benchmark showed the best prediction performance. Using the RF as a learner and the 1000 most informative SNPs as predictors, the mean (SD) of the 10 values obtained in the test sets were: 0.23 (0.04) for the Spearman correlation, 0.83 (0.04) for the zero-one loss, and 0.33 (0.03) for the rank distance loss. We conclude that the information on predicted components of RFI (DFI, ADG, MW, and BFT) does not contribute to improve the quality of the prediction of this trait in relation to the one obtained with the single-output strategy.

Mediastinitis After Congenital Heart Surgery: Epidemiology, Clinical Features and Outcomes.

Mediastinitis after surgery for congenital heart disease has a great impact on morbidity and mortality. However, there are scarce studies focused on the epidemiology of postsurgical mediastinitis in pediatric patients. In this 18-year period retrospective study, the cumulative incidence of mediastinitis was low: 0.64%, (95% confidence interval: 0.36-1.1). Gram-negative bacilli were common (35%). The mortality rate was 7.1%, associated with fungal infection.

Scimitar syndrome and left pulmonary vein stenosis: A serious and rare association.

Scimitar syndrome is a rare constellation of cardiovascular anomalies consisting of partial right anomalous pulmonary venous drainage to the inferior vena cava, hypoplasia of the right lung, cardiac dextroposition, and in many cases, one or more aortopulmonary collaterals from the descending aorta to the hypoplastic lung. The "infantile form" commonly presents with significant heart failure and pulmonary hypertension. Stenosis of the left pulmonary veins associated with scimitar syndrome is a rare but well-described combination, usually associated with a poor prognosis. We describe two patients with this association in whom the left pulmonary vein stenosis manifested months after the initial diagnosis as a progressive lesion that complicated the course of the disease due to severe pulmonary hypertension. Both patients were successfully treated with a combination of surgical, percutaneous, and hybrid treatment.

Friedreich's ataxia-associated childhood hypertrophic cardiomyopathy: a national cohort study.

OBJECTIVE: Hypertrophic cardiomyopathy (HCM) is an important predictor of long-term outcomes in Friedreich's ataxia (FA), but the clinical spectrum and survival in childhood is poorly described. This study aimed to describe the clinical characteristics of children with FA-HCM. DESIGN AND SETTING: Retrospective, longitudinal cohort study of children with FA-HCM from the UK. PATIENTS: 78 children (<18 years) with FA-HCM diagnosed over four decades. INTERVENTION: Anonymised retrospective demographic and clinical data were collected from baseline evaluation and follow-up. MAIN OUTCOME MEASURES: The primary study end-point was all-cause mortality (sudden cardiac death, atrial arrhythmia-related death, heart failure-related death, non-cardiac death) or cardiac transplantation. RESULTS: The mean age at diagnosis of FA-HCM was 10.9 (±3.1) years. Diagnosis was within 1 year of cardiac referral in 34 (65.0%) patients, but preceded the diagnosis of FA in 4 (5.3%). At baseline, 65 (90.3%) had concentric left ventricular hypertrophy and 6 (12.5%) had systolic impairment. Over a median follow-up of 5.1 years (IQR 2.4-7.3), 8 (10.5%) had documented supraventricular arrhythmias and 8 (10.5%) died (atrial arrhythmia-related n=2; heart failure-related n=1; non-cardiac n=2; or unknown cause n=3), but there were no sudden cardiac deaths. Freedom from death or transplantation at 10 years was 80.8% (95% CI 62.5 to 90.8). CONCLUSIONS: This is the largest cohort of childhood FA-HCM reported to date and describes a high prevalence of atrial arrhythmias and impaired systolic function in childhood, suggesting early progression to end-stage disease. Overall mortality is similar to that reported in non-syndromic childhood HCM, but no patients died suddenly.

Epidemiology of Pediatric Heart Failure in the USA-a 15-Year Multi-Institutional Study.

The epidemiology of pediatric heart failure (HF) has been characterized for congenital heart disease (CHD) and cardiomyopathies (CM), but the impact of CM associated with CHD has not been studied. This study aims to describe the characteristics and outcomes of inpatient pediatric HF patients with CHD, CM, and CHD with CM (CHD + CM) across the USA. We included all HF patients with CM diagnoses with and without CHD using ICD 9/10 codes ≤ 19 years old from January 2004 to September 2019 using the Pediatric Health Information System database. We identified 67,349 unique patients ≤ 19 years old with HF, of which 87% had CHD, 7% had CHD + CM, and 6% had CM. Pediatric HF admissions increased significantly from 2004 to 2018 with an associated increase in extracorporeal circulatory support (ECLS) use. Heart transplantation (HTX) increased only in the CHD and CHD + CM groups. CHD patients required less ECLS with and without HTX; however, they had significantly higher inpatient mortality after those procedures than the other groups (p < 0.001). CM patients were older (median 115 months) and had the lowest inpatient mortality after HTX with and without ECLS (p < 0.05). CHD + CM showed the highest overall inpatient mortality (15%), and cumulative hospital billed charges (median US$ 541,374), all p < 0.001. Pediatric HF admissions have increased from 2004 to 2018. ECLS use and HTX have expanded in this population, with an associated decrease in inpatient mortality in the CHD and CM groups. CHD + CM patients are a growing population with the highest inpatient mortality.

Implante de un stent en la anastomosis del colector a la aurícula izquierda tras cirugía correctora de drenaje venoso pulmonar anómalo total / Stent Implantation in the Anastomosis After Correction of a Total Anomalous Pulmonary Venous Connection

No disponible

A framework for diagnosis of urinary incontinence disease based on scoring measures and automatic classifiers.

This work develops a decision support system based on machine learning and scoring measures to determine the type of urinary incontinence in women with low urinary tract symptoms. This system has two main branches. The former consists of selecting the feature set which best defines the UI type from the set of features (age, weight, etc.) characterizing a patient. This feature set is computed from several scoring measures. The patients characterized by the optimum feature set are then classified according to C4.5 and SVM classifiers. The results are evaluated according to Sensitivity and Specificity evaluation measures. The management of the final system is simple and its performance is high, getting Sensitivities over 80% and Specificities near 100% for some configurations.