Kernel methods for biosensing application

This thesis examines the design noise robust information retrieval techniques based on kernel methods. Algorithms are presented for two biosensing applications: (1) High throughput protein arrays and (2) Non-invasive respiratory signal estimation. Our primary objective in protein array design is to maximize the throughput by enabling detection of an extremely large number of protein targets while using a minimal number of receptor spots. This is accomplished by viewing the protein array as a communication channel and evaluating its information transmission capacity as a function of its receptor probes. In this framework, the channel capacity can be used as a tool to optimize probe design; the optimal probes being the ones that maximize capacity. The information capacity is first evaluated for a small scale protein array, with only a few protein targets. We believe this is the first effort to evaluate the capacity of a protein array channel. For this purpose models of the proteomic channel's noise characteristics and receptor non-idealities, based on experimental prototypes, are constructed. Kernel methods are employed to extend the capacity evaluation to larger sized protein arrays that can potentially have thousands of distinct protein targets. A specially designed kernel which we call the Proteomic Kernel is also proposed. This kernel incorporates knowledge about the biophysics of target and receptor interactions into the cost function employed for evaluation of channel capacity.;For respiratory estimation this thesis investigates estimation of breathing-rate and lung-volume using multiple non-invasive sensors under motion artifact and high noise conditions. A spirometer signal is used as the gold standard for evaluation of errors. A novel algorithm called the segregated envelope and carrier (SEC) estimation is proposed. This algorithm approximates the spirometer signal by an amplitude modulated signal and segregates the estimation of the frequency and amplitude in-formation. Results demonstrate that this approach enables effective estimation of both breathing rate and lung volume. An adaptive algorithm based on a combination of Gini kernel machines and wavelet filltering is also proposed. This algorithm is titled the wavelet-adaptive Gini (or WAGini) algorithm, it employs a novel wavelet trans-form based feature extraction frontend to classify the subject's underlying respiratory state. This information is then employed to select the parameters of the adaptive kernel machine based on the subject's respiratory state. Results demonstrate significant improvement in breathing rate estimation when compared to traditional respiratory estimation techniques.