Force Characterization And Motion Planning In Automated Cell Manipulation By Optical Tweezers

Optical tweezers, which are based on the transfer of photon momentum, can trapand move microscale and nanoscale particles without physical contact. Rapid andprecise transportation of live cells can benefit cell microsurgery, rare cell isolation,tissue engineering and cell-to-cell interactions. Increasing demands for both accuracyand efficiency in biological cell manipulation highlight the need for automation withrobotics technology. Understanding the forces exerted on live cells is essential tobiomechanical characterization and cell manipulation. However, traditional numericalforce measurement assumes live cells to be ideal objects, ignoring their complicatedinner structures and rough membranes. Furthermore, little reported research hasspecifically considered the synergy of dynamic analysis in motion planning duringautomated transportation. The problem of planning cell motion with optimized motionparameters, using cell dynamics analysis, is still very challenging.This thesis aims to characterize the mechanical forces applied to live cells inoptical traps, and use the mechanical parameters thus obtained to plan cell motionduring automated transportation. The research principally consists of the followingthree elements.First, the forces applied to live cells are calibrated by a novel staticviscous-drag-force method. Unlike existing approaches, the proposed method doesnot assume the live cells to share the same optical and/or drag parameters as those ofpolystyrene/silica beads. By binding a micro polystyrene sphere to the live cell andmoving the mixture with optical tweezers, the drag force on the cell can be obtainedby subtracting the drag force on the sphere from the total drag force on the mixture,under the condition of an extremely low Reynolds number. The trapping force on thecell is then obtained from the drag force when the cell is in the force equilibrium state.Second, motion planning strategy, which is designed using dynamics analysis ofthe optically trapped cell, is used to determine the ideal movement velocity of the cell. Due to property changes in the aqueous medium and laser during cell transportation,the calibrated dynamic parameters may vary, and thus, the cell movement velocitydesigned using these parameters should be adjusted online. A proportional-integral (PI)scheme is used to adjust the cell movement velocity online, to ensure that the cellmoves at an ideal speed and does not escape from the laser focus. Dynamics analysisresults are used to design the PI scheme.Third, an optimal path for cell movement is planned, using a modified A-staralgorithm, which introduces an additional cost to penalize waypoints where thedirection of movement changes. The algorithm balances smoothness and movementcost. Finally, experiments on moving yeast cells are conducted to demonstrate theeffectiveness of the proposed approach.The main contribution of this study lies in the development of a newexperimental method to characterize the mechanical forces exerted on live cells, andthe application of the dynamic analysis results to motion planning for automated celltransportation.