C Robin Walker | PubMed Health

PubMed2006

Predicting high-risk preterm birth using artificial neural networks.

A reengineered approach to the early prediction of preterm birth is presented as a complimentary technique to the current procedure of using costly and invasive clinical testing on high-risk maternal populations. Artificial neural networks (ANNs) are employed as a screening tool for preterm birth on a heterogeneous maternal population; risk estimations use obstetrical variables available to physicians before 23 weeks gestation. The objective was to assess if ANNs have a potential use in obstetrical outcome estimations in low-risk maternal populations. The back-propagation feedforward ANN was trained and tested on cases with eight input variables describing the patient's obstetrical history; the output variables were: 1) preterm birth; 2) high-risk preterm birth; and 3) a refined high-risk preterm birth outcome excluding all cases where resuscitation was delivered in the form of free flow oxygen. Artificial training sets were created to increase the distribution of the underrepresented class to 20%. Training on the refined high-risk preterm birth model increased the network's sensitivity to 54.8%, compared to just over 20% for the nonartificially distributed preterm birth model.

Artificial Intelligence↗

PubMed2003

Clinical outcomes following institution of universal leukoreduction of blood transfusions for premature infants.

CONTEXT: Leukocytes present in stored blood products can have a variety of biological effects, including depression of immune function, thereby increasing nosocomial infections and possibly resulting in organ failure and death. Premature infants, given their immature immune state, may be uniquely predisposed to the effects of transfused leukocytes. OBJECTIVE: To evaluate the clinical outcomes following implementation of a universal prestorage red blood cell (RBC) leukoreduction program in premature infants admitted to neonatal intensive care units (NICUs). DESIGN AND SETTING: Retrospective before-and-after study conducted in 3 Canadian tertiary care NICUs from January 1998 to December 2000. PATIENTS: A total of 515 premature infants weighing less than 1250 g who were admitted to the NICU, received at least 1 RBC transfusion, and survived at least 48 hours were enrolled. The intervention group consisted of infants admitted in the 18-month period following the introduction of universal leukoreduction (n = 247) and the control group consisted of infants admitted during the 18 months prior to the introduction of universal leukoreduction (n = 268). MAIN OUTCOME MEASURES: Primary outcomes were nosocomial bacteremia and NICU mortality, compared before and after implementation of universal leukoreduction using multivariate regression. Secondary outcomes included bronchopulmonary dysplasia, retinopathy of prematurity, necrotizing enterocolitis, and intraventricular hemorrhage. RESULTS: The proportion of infants who acquired bacteremia after an RBC transfusion was 79/267 (29.6%) in the nonleukoreduction period and 63/246 (25.6%) in the leukoreduction period. For NICU mortality, there were 45 deaths (16.8%) in the nonleukoreduction period and 44 deaths (17.8%) in the leukoreduction period. The adjusted odds ratio (OR) for bacteremia was 0.59 (95% confidence interval [CI], 0.34-1.01) and for mortality was 1.22 (95% CI, 0.59-2.50). The adjusted ORs for bronchopulmonary dysplasia and retinopathy of prematurity were 0.42 (95% CI, 0.25-0.70) and 0.56 (95% CI, 0.33-0.93), respectively. The adjusted ORs for necrotizing enterocolitis and grade 3 or 4 intraventricular hemorrhage were 0.39 (95% CI, 0.17-0.90) and 0.65 (95% CI, 0.35-1.19), respectively. The adjusted OR for a composite measure of any major neonatal morbidity was 0.31 (95% CI, 0.17-0.56). Crude and adjusted rates for all secondary outcomes suggest that leukoreduction was associated with improved outcomes. CONCLUSION: Implementation of universal prestorage leukoreduction was not associated with significant reductions in NICU mortality or bacteremia but was associated with improvement in several clinical outcomes in premature infants requiring RBC transfusions.

Bacteremia↗