An active machine learning approach for optimal design of magnesium alloys using Bayesian optimisation.

In the pursuit of magnesium (Mg) alloys with targeted mechanical properties, a multi-objective Bayesian optimisation workflow is presented to enable optimal Mg-alloy design. A probabilistic Gaussian process regressor model was trained through an active learning loop, while balancing the exploration and exploitation trade-off via an acquisition function of the upper confidence bound. New candidate alloys suggested by the optimiser within each iteration were appended to the training data, and the performance of this sequential strategy was validated via a regret analysis. Using the proposed approach, the dependency of the prediction error on the training data was overcome by considering both the predictions and their associated uncertainties. The method developed here, has been packaged into a web tool with a graphical user-interactive interface (GUI) that allows the proposed optimal Mg-alloy design strategy to be deployed.

Validation of a diagnostic multiplex polymerase chain reaction assay for infectious posterior uveitis.

OBJECTIVE: To validate a multiplex polymerase chain reaction (PCR) assay capable of simultaneously screening vitreous biopsy specimens for a panel of common pathogens in posterior uveitis. METHODS: A multiplex PCR assay using novel primer sets for cytomegalovirus (CMV), herpes simplex virus (HSV), varicella zoster virus (VZV), and Toxoplasma gondii was developed. The sensitivity of the assay was determined for purified pathogen DNA. Twenty-one vitreous specimens from patients with posterior uveitis were tested by both multiplex and monoplex PCR. RESULTS: Fewer than 10 genomes of VZV and fewer than 100 genomes of HSV, CMV, and T gondii could be detected using the new primer sets. When used in multiplex, the assay lost less than 1 log of sensitivity. Monoplex PCR detected pathogen DNA in 18 of 21 patient samples; multiplex PCR detected pathogen DNA in 15 of the 18 samples positive by monoplex PCR. None of 10 negative control samples were positive for pathogen DNA. CONCLUSIONS: Multiplex PCR has adequate sensitivity to simultaneously screen a substantial differential diagnosis for posterior uveitis in a single reaction, without loss of specificity. This assay may reduce the time and cost involved in PCR-based molecular diagnostics of infectious pathogens. CLINICAL RELEVANCE: Mutiplex PCR may allow rapid diagnosis of infectious posterior uveitis.