Introduction & MotivationA reliable method for early detection for respiratory infections could result in better health outcomes. The gene expression levels of a cell change upon infection and may be a good predictor of illness. These gene expression levels of a cell correspond to the state of the cell and are measured by the abundance of mRNA produced during transcription of each gene. However, for any cell there are tens of thousands of genes measured and limited patient data available. Because of these constraints, we believe a k-nearest classifier would work well to classify infection and we seek to answer the question of which k values and distance measures perform best.
AcknowledgmentsThank you to to Prof. Bryan Pardo, Prof. Rosemary Braun, and Prof. Doug Downey for the guidance in developing this project.
|