ABSTRACT
Background: Detection of heart failure with preserved ejection fraction (HFpEF) involves integration of multiple imaging and clinical features which are often discordant or indeterminate. Objectives: The authors applied artificial intelligence (AI) to analyze a single apical 4-chamber transthoracic echocardiogram video clip to detect HFpEF. Methods: A 3-dimensional convolutional neural network was developed and trained on apical 4-chamber video clips to classify patients with HFpEF (diagnosis of heart failure, ejection fraction ≥50%, and echocardiographic evidence of increased filling pressure; cases) vs without HFpEF (ejection fraction ≥50%, no diagnosis of heart failure, normal filling pressure; controls). Model outputs were classified as HFpEF, no HFpEF, or nondiagnostic (high uncertainty). Performance was assessed in an independent multisite data set and compared to previously validated clinical scores. Results: Training and validation included 2,971 cases and 3,785 controls (validation holdout, 16.8% patients), and demonstrated excellent discrimination (area under receiver-operating characteristic curve: 0.97 [95% CI: 0.96-0.97] and 0.95 [95% CI: 0.93-0.96] in training and validation, respectively). In independent testing (646 cases, 638 controls), 94 (7.3%) were nondiagnostic; sensitivity (87.8%; 95% CI: 84.5%-90.9%) and specificity (81.9%; 95% CI: 78.2%-85.6%) were maintained in clinically relevant subgroups, with high repeatability and reproducibility. Of 701 and 776 indeterminate outputs from the Heart Failure Association-Pretest Assessment, Echocardiographic and Natriuretic Peptide Score, Functional Testing (HFA-PEFF), and Final Etiology and Heavy, Hypertensive, Atrial Fibrillation, Pulmonary Hypertension, Elder, and Filling Pressure (H2FPEF) scores, the AI HFpEF model correctly reclassified 73.5% and 73.6%, respectively. During follow-up (median: 2.3 [IQR: 0.5-5.6] years), 444 (34.6%) patients died; mortality was higher in patients classified as HFpEF by AI (HR: 1.9 [95% CI: 1.5-2.4]). Conclusions: An AI HFpEF model based on a single, routinely acquired echocardiographic video demonstrated excellent discrimination of patients with vs without HFpEF, more often than clinical scores, and identified patients with higher mortality.
ABSTRACT
A cross-sectional study on clinical mastitis, intramammary infection (IMI) and blind quarters was conducted on 50 smallholder dairy farms in Jimma, Ethiopia. A questionnaire was performed, and quarters of 211 cows were sampled and bacteriologically cultured. Risk factors at the herd, cow, and quarter level for clinical mastitis and (pathogen-specific) intramammary infection were studied using multilevel modeling. As well, factors associated with quarters being blind were studied. Eleven percent of the cows and 4% of the quarters had clinical mastitis whereas 85% of the cows and 51% of the quarters were infected. Eighteen percent of the cows had one or more blind quarter(s), whereas 6% of the quarters was blind. Non-aureus staphylococci were the most frequently isolated pathogens in both clinical mastitis cases and IMI. The odds of clinical mastitis was lower in herds where heifers were purchased in the last year [odds ratio (OR) with 95% confidence interval: 0.11 (0.01-0.90)], old cows (>4 years) [OR: 0.45 (0.18-1.14)], and quarters not showing teat injury [OR: 0.23 (0.07-0.77)]. The odds of IMI caused by any pathogen was higher in herds not practicing teat drying before milking (opposed to drying teats with 1 towel per cow) [OR: 1.68 (1.05-2.69)], cows in later lactation (>180 DIM opposed to ≤90 DIM) [OR: 1.81 (1.14-2.88)], cows with a high (>3) body condition score (BCS) [OR: 1.57 (1.06-2.31)], right quarters (opposed to a left quarter position) [OR: 1.47 (1.10-1.98)], and quarters showing teat injury [OR: 2.30 (0.97-5.43)]. Quarters of cows in herds practicing bucket-fed calf feeding (opposed to suckling) had higher odds of IMI caused by Staphylococcus aureus [OR: 6.05 (1.31-27.90)]. Except for BCS, IMI caused by non-aureus staphylococci was associated with the same risk factors as IMI caused by any pathogen. No access to feed and water immediately after milking [OR: 2.41 (1.26-4.60)], higher parity [OR: 3.60 (1.20-10.82)] and tick infestation [OR: 2.42 (1.02-5.71)] were risk factors for quarters being blind. In conclusion, replacement of old cows, prevention of teat injuries/lesions, drying teats with 1 towel per cow before milking, improving fertility in order to shorten the lactation period, allowing (restricted) suckling, access to feed and water immediately after milking, and improving tick control could improve udder health in Jimma.
