A Biologically Motivated Chip and Its Application to the Analysis of Weight Change of Prematurely Born Babies

Prematurely born babies usually suffer from nutrition deficiency. Total parental nutrition (TPN) has been one of the major treatments commonly used by clinicians to improve the nutritional needs of prematurely born babies. This paper described the application of an artificial neuromolecular (ANM) chip to a database of prematurely born babies who were treated with TPN. The objective was to investigate the weight changes of these babies from the data of physical examination, laboratory tests, and TPN elements. Experimental results showed that the ANM chip achieved satisfactory results, outperforming both the back-propagation neural networks and SAS (a statistical tool). Out of our expectation, the result was even better than that of human judgment, suggesting that it could be used as an assisting tool to clinicians.