SPECIALLY COLLECTED PAPERS
Shimadzu Review 75[3・4] (2018.12)
In general, researchers conduct iterative experiments by changing experimental conditions by trial and error. If Artificial Intelligence (AI) could interpret experimental results and design the next experimental condition, then, by utilizing AI and robotic technology, the iterative experimental cycle could be automated, leading to reduced variation due to individual differences, reduced labor costs, and increased instrument utilization rates. This article presents a Bayesian optimization method to automate iterative experiments that include noisy observations. By applying the Bayesian optimization method that we have developed to LC-MS, the instrument parameters for high sensitivity LC-MS measurements can be found efficiently with a small number of measurements.
Keywords: Parameter optimization, Bayesian optimization, High sensitivity measurement, LC-MS, Electrospray ionization
1AI Solution Unit, Technology Research Laboratory, Shimadzu Corporation, Kyoto, Japan
2MS Business Unit, Life Science Business Department, Analytical & Measuring Instruments Division, Shimadzu Corporation, Kyoto, Japan
*The information contained in Shimadzu Review has not been modified since the original publication date. Please be aware that in some cases, products mentioned within the articles are no longer available.