ABSTRACT
The performance of adaptive systems that consist of microscale on-chip elements [microelectromechanical mirror (mu-mirror) arrays and a VLSI stochastic gradient descent microelectronic control system] is analyzed. The mu-mirror arrays with 5 x 5 and 6 x 6 actuators were driven with a control system composed of two mixed-mode VLSI chips implementing model-free beam-quality metric optimization by the stochastic parallel perturbative gradient descent technique. The adaptation rate achieved was near 6000 iterations/s. A secondary (learning) feedback loop was used to control system parameters during the adaptation process, further increasing the adaptation rate.
ABSTRACT
Since there was a scientific need to conduct electrophysiology measurements to detect possible ocular (electroretinography, ERG), central neurotoxic (quantitative electroencephalography, qEEG), and cardiac (electrocardiography, ECG) effects in animals used in certain regulatory studies, the acquisition of suitable automated PC software systems were required. This article describes the process by which these systems were validated to ensure that they met the scientific requirements, while also addressing the principles of Good Automated Laboratory Practices (GALP). After a thorough search of existing commercial packages, a plan was developed specific for each PC-based collection system selected for evaluation. The common elements of each plan included consideration of both scientific and GALP elements, such as necessary biological response variables, raw data acquisition and identification, acceptance criteria, security, protection, storage media, data integrity, audit requirements and standard operating procedures. The authors' approach to validation for each electrophysiology system was to determine scientific needs for accuracy, precision, and detection limit of biological effects concurrent with GALP requirements. The selected software systems were employed in separate scientific GLP studies conducted in dogs, rats, and mini-pigs to demonstrate the ability to detect cholinesterase effects due to multiple infusions of physostigmine, based on parallel measurement of cholinesterase biomarkers. Since the systems were designed for human usage, certain adaptations were necessary. A critical assumption to be tested was the ability of the system's algorithms to adequately capture and assimilate the data in an accurate fashion. Concomitantly, the related GALP needs, such as data integrity, security, CD-ROM archive, and personnel training requirements were evaluated, implemented, and defined to accommodate the application and process needs. The biological approach to validation of these PC-based electrophysiology systems met the necessary scientific acceptance criteria as well as compliance requirements in order to be used in regulatory studies.
