Artificial neural network for predicting intracranial haemorrhage in preterm neonates

Intraventricular haemorrhage (IVH) incidence is used to assess peri-/neonatal therapy, and to make intra-and inter-hospital quality assessments. Unbiased assessment is complicated by the amount of confounding factors. Is an artificial neural network (ANN) able to early and accurately forecast the occurrence of severe IVH in an individual patient? Is it superior to classic multiple logistic regression? We conducted an observational study on pre-existing routine data. Admission data were available from 890 preterm neonates (gestational age < 32 weeks, birthweight < 1500 g). Patients were randomly assigned to either a training, or a validation set (50%/50%). Using the training set data an ANN was trained. A second predictive model was developed by stepwise multiple logistic regression analysis. Using the validation set input data both models delivered estimates of the probability for severe IVH to occur in each individual patient. Receiver operating characteristic (ROC) curves were used to compare prognostic performance. The optimal ANN processed 13 input variables, whereas stepwise logistic regression analysis only identified five independent predictor variables. The area under the ROC curve was 0.935 for the ANN and 0.884 for the logistic regression model (p= 0:001). Adjusted for 95%, 90%, 85%, 80% and 75% specificity, the sensitivity of the ANN was significantly superior to that of the logistic regression model. Due to its ability to give an accurate prognosis based solely on admission data, a trained ANN qualifies as a tool for local quality control.     

Concentration and separation of erythrocytes infected with Plasmodium falciparum by gradient centrifugation.

Concentration of infected erythrocytes was achieved in cell suspensions derived from long-term culture of Plasmodium falciparum growing asynchronously in human erythrocytes. This new procedure involves the slow centrifugation (at 33 g) of erythrocyte suspensions through 5% Ficoll solutions. Mature asexual erythrocytic forms are preferentially retained in the gradient solution (top fraction). After further gradient centrifugation of these parasitized cells, the concentration of mature forms is increased 15- to 31-fold and a mature form parasitaemia of 71-81% is obtained in the final erythrocyte suspension. Furthermore, at least 75% of the total number of the mature forms can be retrieved by this method. Parasitized cells that are not retained in the gradient are sedimented to the bottom of the tube (bottom fraction) and consist predominantly of ring forms. Parasites from both the top and bottom fractions are viable and have been used to initiate short-term synchronous cultures. By providing purified parasite preparations, this simple procedure will facilitate immunological, chemotherapeutic, and biochemical studies with P. falciparum.     

Monitoring the response of malaria infections to treatment

A simplified schema for assessing the response of malaria patients to treatment is described. It involves collection and examination of blood films prior to the first dose and on days 2 and 7 after the onset of treatment.     

Effect of postpartum maternal infection on proteins and trace elements in colostrum and early milk

In developing countries, maternal infections during lactation are common. In this study, we evaluated the effect of acute maternal postpartum infection on the composition of colostrum and early milk with special emphasis on milk proteins and trace elements. The study was carried out in two maternity hospitals in Lima, Peru. Subjects were normally nourished women (body mass index (BMI) > 20.0) who intended to exclusively breastfeed their child and who had fever and clinical symptoms of infection within the first 48 h postpartum ( n = 34). Non-ill women of similar characteristics were selected as controls ( n = 23). Blood and milk samples were taken on days 1 and 14 postpartum. An acute phase response was confirmed by significantly increased serum levels of C-reactive protein in infected women. Serum zinc levels increased significantly from day 1 to day 14, but were not affected by infection. Serum copper levels were significantly higher in ill women than in non-ill women on day 1. All participating women were breastfeeding on day 14. Whey protein levels, the whey/casein ratio and total protein levels decreased significantly with time, but were not affected by infection. There were no differences in milk iron or copper levels with time or infection. Milk zinc levels decreased significantly with time, but were not affected by infection. Maternal infection during the early postpartum period does not appear to adversely affect the initiation of lactation or milk protein and trace element contents.