Paper microfluidic devices are capable of processing fluids using capillary flow in materials that are lower cost than the standard materials used in microfluidics. Due to the low cost of their fabrication and minimal requirements for instrumentation, paper microfluidic devices are ideal for use in low-resource settings. They have the...
Remote health monitoring tools have significant potential to revolutionize the current healthcare paradigm. Such tools could improve efficiencies in healthcare by alerting patients and doctors to abnormal health conditions in real time, shortening the time to diagnose and treat ailments. To meet these needs, we pursue the development of fabric-based...
Transmit beamforming is an important technique employed to improve efficiency and signal quality in wireless communication systems by steering signals towards their in- tended users. It often arises jointly with the antenna selection problem due to various reasons, such as limited number of radio frequency (RF) chains and energy/resource effi-...
The types and rates of label noise in real-world data sets present a challenge to machine learning projects. In this thesis, we propose a novel approach to address this issue. Our method combines a noise modelling technique for correcting label noise across the entire data set with a robust loss...
Crowdsourcing is a popular paradigm to address the high demands for labeled data in big data deluge. It aims to produce accurate labels by effectively integrating noisy, non-expert labeling from crowdsourced workers (annotators). The machine learning community has been studying effective crowdsourcing methods for many years, and many models and...
Personal health monitoring tools, a subset of point-of care-diagnostics, enable more frequent testing of chronic health conditions outside of a clinic setting and can inform personalized therapy or support telehealth diagnostic capabilities. While there is evidence that regular monitoring of chronic conditions can lead to better health outcomes and lower...
We consider the problem of computing the cannonical polyadic decomposition (CPD) for large-scale dense tensors. This work is a combination of alternating least squares and fiber sampling. Data sparsity can be leveraged to handle large tensor CPD, but this route is not feasible for dense data. Inspired by stochastic optimization's...
Phenylketonuria (PKU) is a genetic inborn metabolic disorder which inhibits the functional production of the enzyme phenylalanine hydroxylase (PAH). It currently affects one in 15,000 Americans and is the most common amino acid metabolism error found in newborns. Individuals affected with PKU are incapable of metabolizing the amino acid, phenylalanine...
The anticonvulsant drug carbamazepine (CBZ) is a commonly prescribed drug that is used to treat patients with neurological disorders including epilepsy and bi-polar disorder. CBZ has a narrow range of efficacy: Below the therapeutic dose, seizures and other conditions are not controlled, and above the recommended window, patients experience side...
Paper-based assays for HIV diagnosis are cheap, portable, and require little to no skilled training, electricity, or expensive equipment to perform successfully. However, there is a need for tests with lower limits of detection (LOD) for diagnosing infection at the earliest stages. Effective labeling to increase signal-to-noise ratio is critical...