Babies born too early often struggle to survive. But doctors can have a telling of the difficulty that preemies are developing serious health problems such as respiratory failure, and those who are well. Now researchers have developed a model that can predict the results of preemie with over 0% accuracy, an advance that could help doctors identify the sickest babies and save billions of dollars in health care costs.
thereFifty years, physician Virginia Apgar of Columbia University has developed a rating system for evaluating the health of a newborn. Apgar score still the standard-method takes into account factors such as whether a baby flexes his arms and legs or is still, breathing well or not at all, and if the skin is healthy and pink or blue. Regarding the prediction of serious diseases such as bleeding in the lungs, however, the Apgar score is just only about 70% of the time. New models that take into account the number of white blood cells and blood pH do better, but "they require a lot of invasive tests," says Anna Penn, a neonatologist at the Children's Hospital Lucile Packard (LPCH) in Palo Alto, in California.
The researchers, including Penn and co-lead author Daphne Koller, a computer scientist at Stanford University, has undertaken to develop an even more precise noninvasive tool to predict major complications in newborns smaller. The researchers selected 138 children born LPCH who spent less than 35 weeks in the womb and weighs less than 2 kg. The team classified the preemies as high or low risk based on the diseases they have developed. Babies in the high-risk group died or developed serious complications such as infections, bleeding and lung and heart problems. Infants in the low risk group suffered only minor ailments, such as mild respiratory distress.
Next, the researchers examined the physiological data collected regularly in the first 3 hours of life by bedside monitors such as heart rate, respiratory rate, and the amount of oxygen in the blood. When they modeled these data, they observed signatures in sick babies were different from those they have observed in healthy ones. They used these differences to develop a mathematical algorithm that integrates physiological data monitors, birth weight, and length of time spent in the womb to predict the likelihood that a preemie will develop severe illness. "These are very simple things," says Penn. "But when combined with sophisticated tools that come from the computer, we can make sense of these in a way that doctors are not normally."
The output of the model is a number between 0 and 1, the researchers call "PhysiScore." A higher score indicates a higher risk of complications. For example, a child with a score of 0.8 would have a risk of developing a serious disease by 80%.
PhysiScore outperformed not only the Apgar scale but also three models based on invasive laboratory tests, reports online today in Science Translational Medicine team . Using PhysiScore, the researchers were able to predict severe complications with an accuracy of between 91% and 98%. The accuracy of the Apgar score ranged from 70% to 74%, and other models have accuracies of 82% to 91%.
The researchers plan monitors that could calculate and display PhysiScore a baby automatically 3 hours after birth. That number could help doctors decide whether the baby should receive more aggressive care or be transferred to a better equipped hospital. "[The monitors] are already measuring these signals," says Suchi Saria, a computer scientist at Stanford University, who led the work. So it would be a way to "use existing resources to make better use data that has already collected, "she said.
" This is a huge progress in the field, "said Rosemary Higgins, a neonatologist at the National Institute of child health and human development in Rockville , Maryland. "Predicting the outcome of premature babies is a major challenge for physicians." Still, she would see how the prices of models in the smallest preemies-those weighing less than 1 kg. "This is really the group at high risk for major development problems," she said.
Namasivayam Ambalavanan, neonatologist at the University of Alabama, Birmingham, said doctors often use their judgment to identify preterm FARE wrong. He would like to see a study that pits PhysiScore against clinical judgment.
Penn said the same techniques used to create PhysiScore could also work to identify high-risk surgical patients or adults the most likely to suffer complications following a heart attack. "One of the most interesting things will be to see if we can apply this model to other parameters," she said.
This article identified Suchi Saria Stanford University as co-author of the study. She led the research. the text has been amended accordingly.
0 Komentar